Win with the Caro-Kann

Sverre Johnsen and Torbjørn Ringdal Hansen

A choice of counterattacking options from a sound base

Contents

Symbols

Conventions

Bibliography

Preface by

Sverre

Introduction

1: Classical: Korchnoi Variation Lesson

1: Introduction Lesson 2:

Forgacs’s 6 Bc4 Lesson 3:

Introduction to the Main Line Lesson 4:

Introduction to the Hansen Variation Lesson

5: Hansen Variation: Lines with 0-0-0

2: Classical: Capablanca

Variation Lesson 6: Rare Fifth

Moves Lesson 7: Rare Sixth Moves

Lesson 8: Rare Seventh Moves

Lesson 9: Bareev Introduction and Campora Hybrid 8

h5 Lesson 10: Introduction to the Bareev Main Line

Lesson 11: Bareev Main Line: Menchik’s Line

Lesson 12: Bareev Main Line: 11...Ngf6

3: Advance Variation

Lesson 13: Arkell/Khenkin Introduction

Lesson 14: Arkell/Khenkin 4 dxc5 e6 Introduction and 5

Nf3 Lesson 15: Arkell/Khenkin 5 Be3 and Rare 5th Moves

Lesson 16: Arkell/Khenkin Introduction to 5 a3

Lesson 17: Arkell/Khenkin Main Battleground

Lesson 18: Experimental Line: Johnsen Variation (3...Qc7)

4: Exchange Variation and Panov Attack

Lesson 19: Exchange: Introduction to the Carls Variation

Lesson 20: Classical Exchange Main Line 6 Ne2

Lesson 21: Panov Attack: Introduction and 5 Nf3 g6

Lesson 22: Panov Attack Euwe System: 6 cxd5 and Other

Moves Lesson 23: Panov Attack Euwe System: 6

Qb3 Lesson 24: Panov Attack: Miles’s 4...dxc4

5: Early Deviations

Lesson 25: Maroczy’s 3 f3 and Rare Third Moves

Lesson 26: 2 Nc3 and Introduction to the Two

Knights
Lesson 27: Two Knights Main Line: 3...Bg4

Lesson 28: Steiner Variation (2 c4)

Lesson 29: King’s Indian Attack

Lesson 30: Rare Second Moves

Index of Variations

Copyright Information

About the Authors

About Gambit Publications

About Gambit Chess

Studio

Other Gambit Titles on Chess Studio and Kindle

Symbols

+ check
++ double check
# checkmate
!! brilliant move
! good move
!? interesting move
?! dubious move
? bad move
?? blunder
+– White is winning
+/– White is much better
+= White is slightly better
= the game is equal
=+ Black is slightly better
–/+ Black is much better
–+ Black is winning
Ch championship
tt team event
corr. correspondence game
1-0 the game ends in a win for White
½-½ the game ends in a draw
0-1 the game ends in a win for Black
(n) nth match game
(D) see next diagram
Conventions

For convenience, and in order to avoid trivial repetitions, we have

followed these conventions:

At the start of the ‘Magnifiers’ in each Lesson, there is a brief

summary of what is covered in each one. When a move appears in
brackets, this means that alternatives to that move are also
examined.
In general we have only attributed analysis where we have been
confident that an author gives his own analysis and is not just
quoting older, unattributed analysis.
Analyses and discoveries by our editor have only been attributed
in a few cases where the analysis provided by far exceeded an
editor’s normal duties.
Bibliography

Main Sources

The Caro-Kann Revisited; F.Rambaldi; Thinkers Publishing 2020

Bologan’s Caro-Kann; V.Bologan; New in Chess 2018

A Practical Black Repertoire with d5, c6. Volume 2; A.Kornev; Chess

Stars 2017

The Caro-Kann Defense: Classical Variation; A.Karpov,

M.Podgaets; Russian Chess House 2016

Opening Repertoire: The Caro-Kann; J.Houska; Everyman Chess 2015

Other Repertoire Books for Black

Grandmaster Repertoire 7: The Caro-Kann; L.Schandorff; Quality

Chess 2010

Smerdon’s Scandinavian; D.Smerdon; Everyman Chess 2015

Repertoire Books for White and Various

Playing 1 e4: Caro-Kann, 1...e5 & Minor Lines; J.Shaw; Quality Chess 2016

How to Build Your Chess Opening Repertoire; S.Giddins;

Gambit Publications 2003

The Blackmar-Diemer Gambit; C.Scheerer; Everyman Chess 2011

Preface by Sverre

I have for a long time had an ambition to complement the books Win with
the London System and Win with the Stonewall Dutch with a book for
Black against 1 e4. Having finally concluded that the Caro-Kann must be
the right opening to complete the trilogy, I contacted my friend Torbjørn
Ringdal Hansen and asked if he would like to co-author a book based on
the Hansen Variation in the Caro-Kann. “You mean the ...h5 line in the
Korchnoi Variation?”, he asked modestly.

Black to play

Modesty can go too far, and in my opinion there is no doubt that this is the
Hansen Variation. While Torbjørn wasn’t the first to play 9...h5 in the
diagram position and isn’t the strongest GM to adopt it, he is undoubtedly
the top authority on this fashionable line. Getting Torbjørn on the team was
a scoop on a par with getting Vlatko Kovačević for the London book and
Ivar Bern for the Stonewall. For a repertoire book it was also a lucky strike
that Torbjørn plays the 3...c5 variation against the Advance Variation,
saving a lot of space compared to 3...Bf5.
Space considerations are important, as the tournament scene has
changed a lot lately. Thanks to huge databases and easily available super
chess engines, today it’s easier and more common to prepare for a game
than ever. The flip side of this is the importance of being unpredictable so
that you face your opponent’s actual skills rather than the analysis of
Stockfish. One of our ambitions when starting out with this project was to
offer some choice against all of White’s major systems – preferably one
well-established main line and one rarer but playable sideline. Overall we
believe we succeeded in this, even if some challenges proved tougher
than expected.

Most lessons have been written by one of us and then reviewed by the
other. If you are curious about the author of a particular lesson, you will find
it right at the beginning. “By: TRH” means that this lesson was mainly
written by Torbjørn Ringdal Hansen and that ‘I’ means him. “By: SJ” means
that this lesson was mainly written by Sverre Johnsen and that ‘I’ means
him.

This book is structured around 30 Model Games. If you read the

introductions and play through these games, you will already know a lot
about the Caro- Kann. Actually, if you are below master level, you should
be ready to play the Caro-Kann with some confidence against opponents of
your own strength. You may not be theoretically well prepared, but you will
have the basics in place. One of the strengths of the Caro-Kann is its
resilience.
Playing Black is never easy, but if there is an opening where you can
survive with common sense and a general understanding of the strategy,
it’s probably the Caro-Kann.

So as not to clutter the text too much, we have separated the

supplementary details and minor lines into what we have called Theory
Magnifiers at the end of each lesson. An ambitious player will want to study
this material as a part of his general preparation. However, for most
readers we expect them to be mostly for looking up a line after it occurs in
one of their games. In theory that might seem the wrong order, but you may
be surprised how quickly you’ll cover the basics even with this approach.
And if you want to speed it up tenfold, you should do the same with your
Internet blitz games.

Here and there throughout the book, you will find a few questions in
italics. These are questions that we imagine some of our less
experienced readers might ask and are followed by a more than usually
detailed verbal
explanation.

The COVID Games

Due to the global COVID-19 virus close-down, there have been unusually
few recent over-the-board tournaments during 2020. This has to some
extent been compensated by Internet activity, but generally with shorter
time-limits and you may question the quality of these games. In general we
consider this an important information source and have included quite a
few recent Internet blitz and rapid games. Nevertheless some points to be
noted are:

Well-prepared players will play the opening well as long as they stay
in their preparation and even for some more moves, as long as the
position remains familiar.
There may be a tendency to choose tricky or rare continuations as
the surprise effect will be greater than in standard games.
Some players will be more willing to experiment or test out new ideas
as there are no ‘real’ rating points at risk.
There is less correlation between the outcome of the opening and
the game result, so don’t be fooled by statistics.
Introduction

Fighting From a Firm Foundation

This book offers an aggressive repertoire for Black against 1 e4. It may
seem paradoxical to base it on the Caro-Kann Defence, 1...c6, given that
this is generally considered one of Black’s more solid defences. However,
although not apparent in Black’s first move, you’ll find the aggression
manifesting itself in the variations we recommend.

With the black pieces it’s not always advisable to go straight for an attack
on the opponent’s king – although there will be enough examples of that
approach too. More frequently our ambition will be more modest – to
create imbalance and lure our opponent into positions that we know better.

1 e4 c6 (D)

White to play

Black carefully prepares to stake a claim in the centre with 2...d5.

Having already seen White advance his king’s pawn as his first move,
Black to a large degree decides the character of the game with his reply. If
you add the fact that White has quite likely spent most of his general
preparation on the two ‘big’ defences, 1...e5 and 1...c5, chances are good
that you will to some degree be playing the game on your home turf.

2 d4

This is the normal move. However, quite a few other moves are also fairly
challenging. There is no way to ‘punish’ 2 Nc3, 2 Nf3, 2 c4 or even 2 d3.
On the bright side, you may (and should) safely play 2...d5 against all of
them.
For details on White’s alternatives to 2 d4, see Lessons 26 to 30.

2...d5 (D)

White to play

Black attacks the unprotected e-pawn, so White’s options are limited. Still,
he can choose between a number of moves, leading to positions of a very
different nature. Here we’ll take a preliminary look at White’s main options:

3 Nc3 or 3 Nd2: The Classical Complex

 3 e5: The Advance Complex
3 exd5: The Exchange Complex

In addition, there are of course some minor third-move alternatives, but

only Maroczy’s 3 f3 of Lesson 25 has some bite if Black is unprepared.

The Classical Complex

3 Nc3

This is known as the Classical Variation. The name is fitting, as White

follows all classical principles by developing the knight to its most
natural square while retaining the central pawn-tension.

Somewhat confusingly, 3 Nd2, which usually leads to exactly the same

position, is sometimes referred to as the Modern Variation. For our
repertoire it’s fortunately not relevant to differentiate between the two
knight moves.
This less natural knight development (after all it blocks the queen and
dark- squared bishop) is usually intended to avoid the lines 3 Nc3 g6 (see
the next note for some elaboration) and/or 3...b5 (which is pointless
without a knight on c3) but gives Black a few additional options which are
not really relevant for us, as they seem less promising than capturing on
e4, transposing to our normal repertoire.

3...dxe4

By classical norms, this is quite a serious concession. Black gives up his

foothold in the centre while helping White’s knight to a nice central square.
Of course it isn’t quite that simple, as the knight may also be seen as a
target, and in any case there isn’t really a better option.

Gurgenidze’s 3...b5, although very rare, is of some interest to us as the

common recommendation 4 a3 may lead to a pawn-structure related to
the Korchnoi Variation of Lessons 1 to 5 after 4...dxe4 5 Nxe4 Nf6 6
Nxf6+ exf6. It’s harder to come up with a good plan against 4 e5.

Gurgenidze’s other patent, 3...g6, is a relatively respected opening and

also somewhat relevant to our repertoire as Black’s set-up may be an
option in
some lines where White plays an early e5. A typical line goes 4 e5 (4 Nf3
Bg7 5 h3 dxe4 6 Nxe4 Nf6 7 Nxf6+ exf6! is a further analogy to the
Korchnoi line) 4...h5 (yes, this is actually a popular move) 5 f4 (White does
better to omit this and seek piece-play) 5...Nh6 (Black is defying all
conventional wisdom) 6 Nf3 Bg4 7 Be3 Nf5 8 Bf2 e6 9 h3 Bxf3 10 Qxf3 h4,
and Black has achieved an interesting blockading position.

4 Nxe4 (D)

Black to play

Now it’s Black’s turn to decide what kind of position he wants:

The Finnish Variation, 4...h6!?, is provocative but probably borderline

playable. Black’s point is that after 5 Nf3 Bf5 6 Ng3 Bh7 he has
arguably saved a move compared to the similar lines in the
Capablanca Variation where Black plays ...Bf5-g6-h7.
The rare 4...Qc7 has been tested by Carlsen but is more or less
nameless. If it weren’t for the Hansen Variation you will find in Lessons
4 and 5, we might also call this the Hansen Variation, though after
Danish GM Lars Bo Hansen, who played it with some frequency for a
while.
 The solid 4...Nd7 is a popular main line. It’s often called the Smyslov
Variation, but also the Karpov Variation (or a combination of the
two). Black prepares ...Ngf6 without allowing any damage to his
pawn- structure, but blocks the deployment of his queen’s bishop.

Our two recommendations in this book are:

4...Bf5 initiates what we shall call the Capablanca Variation.

Black develops a bishop with a tempo gain on White’s knight.
This is the subject of Lessons 6 to 12.
The other developing move, 4...Nf6, is arguably more controversial as
Black volunteers the doubling of his f-pawns. After 5 Nxf6+, the
overtly aggressive 5...gxf6 is what we’ll call the Larsen Variation. It is
strategically risky and does not enjoy a good reputation at present.
We shall instead recommend the Korchnoi Variation, 5...exf6, which is
the subject of Lessons 1 to 5.

These options lead to play of a very different nature, so in order to say

anything meaningful about them before we move on with our overview,
a split is needed:

The Korchnoi Variation and the Endgame

Our main repertoire recommendation is the Korchnoi Variation:

 White to play

While there are a lot of nice things to be said about this line – we shall
see how it gives Black easy development and frequently ferocious
middlegame attacks – there is one clear reason why the Korchnoi
Variation isn’t even more popular: it may lead to a rather gloomy
endgame. So let’s right now have a look at how bad the various
endgames tend to be. And to get it over with, let’s check out what
happens if everything goes wrong and White manages to exchange into
a pure pawn endgame.

The Pawn Endgame

 Black to play

Before writing this book, I (TRH) wasn’t entirely convinced that this pawn
endgame was winning for White (I’ve never had it on the board or as a
realistic possibility after quite a number of Korchnoi games). However, after
analysing a bit, I had to conclude that it indeed is a lost cause for Black.
Even giving Black a couple of tempi to gain space on either side doesn’t
change the evaluation. So it’s clear that a pure pawn endgame must be
avoided. If we add a rook for both sides, the situation is still dangerous for
Black.

The Rook Endgame

 White to play

Let’s start with a look at this typical position.

1 Re1!

White takes advantage of the fact that Black can’t exchange rooks.

1...Re6

This is Black’s best defence as his king now isn’t cut off from the
queenside. With more pieces on the board, such a move is often
impossible, as playing
...fxe6 would lead to a weak backward pawn on e6. In a pure pawn
endgame, this weakness would not mean much at all. Nevertheless, Black
is still in a difficult situation.

2 Re3! f5

Relatively best. 2...Rxe3?! 3 fxe3 leads to a dubious pawn endgame where

White is probably winning. Also after 2...Ke7 3 Ke2 Kd6 4 Kd3 Black is in a
difficult situation, unless he now plays 4...f5. Taking on e3 leads to the lost
pawn endgame, while not capturing gives White a free hand about when to
capture on e6 or swing the rook over to another file. Meanwhile, White
will start advancing pawns on either flank. It’s vital not to be left with a
passive rook.

3 Ke2 Ke7 4 Kd3 (D)

Black to play

This position is much easier for White to play, but Black can probably
hold with accurate defence.

Let’s move on to a somewhat easier position for Black:

 White to play

With his king in the centre, Black has an improved version, as his king is
ready to stop White’s majority. In any case, White still can and should
take over the open file:

1 Re1!

Now in addition to 1...Re6 Black may also cede the open file:

1...Rb8

I believe this is best. Black avoids the rook exchange and is waiting for
a chance to open a new file for the rook. Black should hold.

Simplified Middlegames

The more pieces we add to the board, the harder it gets for White to exploit
his queenside majority. This is an example of a position that is fine for
Black:
 Black to play

The Korchnoi Variation should be played with confidence, and you

should not panic about seeing an endgame getting closer. Here a few
pieces have been exchanged, but there is a long way to go before a
pure rook or pawn endgame is reached, and much will happen before
then!

Trying to make use of the majority as early as the middlegame is a tough

challenge for White. It will most likely come under fire from the black rooks,
which, after all, enjoy one more open file than White’s rooks.

Next up is an example where White’s use of the pawn-majority failed badly.

 Black to play

J. Tan – T.R. Hansen

Internet rapid 2018

White played an early c4-c5 to remove Black’s bishop from d6. The
downside was of a long-term nature: the majority became almost useless
with a backward d-pawn. Black has succeeded with a light-square strategy,
and there is immense pressure on the d-pawn. Add this up with 26...g5!
and it’s clear that Black has a decisive advantage.

Last-Ditch Defence

What if White succeeds in advancing the d-pawn and creates a

protected passed pawn? Well, then I want you to think of this game!
 Black to play

Hebden – T.R. Hansen

Reykjavik 2017

Believe it or not, this position occurred from a King’s Indian. I do everything

to get my doubled f-pawns! The important thing is that the structure is the
same, and at first sight, it looks like White has achieved a lot with his
protected passed d-pawn. However, the position is a dead draw as Black
can set up an impenetrable fortress:

30...f5 31 Qc3 f6 32 Qd2 Kf7 33 Bc2

Here a draw was agreed!

I hope this discussion has set your mind at ease about the spectre of a
difficult endgame. In the introduction to Chapter 1 we shall see many ways
in which Black can use his active pieces to create much more real
problems for White earlier in the game!
Capablanca’s Classical Base

Our secondary repertoire recommendation is the natural bishop development:

4...Bf5 (D)

White to play

This is known as the Capablanca Variation, although he didn’t play it that

often. Probably the name stuck because the resulting positions fitted the
great Cuban’s natural playing style perfectly. Within this variation we aim
for a line that occurs a few more moves along the main road:

5 Ng3 Bg6 6 h4 h6 7 Nf3

At this crossroad we shall focus on 7...e6!?, known as the Bareev

Variation. It’s fairly theoretical and its main point is far from obvious.
Actually, one of the few things it has in common with the Korchnoi
Variation is the fact that it’s currently quite fashionable. Anyway, in order to
understand what’s going on, we need to look at the old main line 7...Nd7,
which prevents 8 Ne5. This was for a long time considered mandatory and
is still the most common
move. Play usually continues 8 h5 Bh7 9 Bd3 Bxd3 10 Qxd3 e6 11 Bd2
Ngf6 12 0-0-0 (D):

Black to play

Here Black has mostly played 12...Be7, usually leading to sharp play after
kingside castling and scoring a very decent 48%. However, Black’s
counterplay generally depends on the central break ...c5, and when the
pawn leaves c6, it’s clear that his queen’s knight would have loved to be
on that square.

So it was quite natural to look for ways to avoid an early ...Nd7. Enter
the Bareev Variation:

7...e6!? (D)

Black allows White to attack the light-squared bishop with his knight rather
than the h-pawn. Back in the 1990s this was considered a provocation or
even a mistake. Today it’s perhaps Black’s most respected move. Another
related idea is the Campora Variation, 7...Nf6.
 White to play

Now the critical reply is 8 Ne5, which is the subject of Lessons 10 to 12.
However, before examining the critical lines, let’s have a look at what
can happen if White continues routinely:

8 h5?!

This move is relatively common, and has been used by some very strong
players. We have therefore dedicated the whole of Lesson 9 to it, even
though it’s not of much theoretical importance.

8...Bh7 9 Bd3 Bxd3 10 Qxd3 Nf6 11 Bd2 (D)

 Black to play

Here White hasn’t castled queenside while Black’s knight hasn’t moved to
d7. This trade clearly favours Black. Here Black usually prefers 11...c5 12
0- 0-0 Nc6, scoring around 70%. However, that score may be a little
inflated by the fact that Black is the stronger player in many of these
games and wins fairly balanced endgames. In our analysis we shall
instead focus on the sharper 11...Be7 12 0-0-0 0-0 (followed by ...c5 and
usually ...Nc6) which also scores very well for Black at around 58%.

Well, now you know what to hope for. The real challenges will be tackled
in Lessons 10 to 12.

The Advance Complex

3 e5 (D)
 Black to play

Until the mid-1980s, the Advance Variation wasn’t considered much of a

challenge for Black, on the basis that after 3...Bf5 Black is playing a kind of
French with his light-squared bishop outside the pawn-chain. This benefit
was supposed to matter more than the loss of tempo when he later plays
...c6- c5.

White then started to have some success in the Van der Wiel Variation, 4
Nc3 e6 5 g4. While Black always seemed theoretically fine, the resulting
positions were sharp and required serious preparation. This made the
Advance a serious option for White, but with time and stronger engines
Black found many ways to get a good game. Therefore it was only when
Short a little later demonstrated that quiet development with moves like
Nf3, c3, Be3, Be2 and 0-0 could also be quite challenging, that the
Advance really came to the theoretical forefront. Rather than allowing
White to choose from this range of interesting options, we decided to
recommend Black’s second most popular choice:

3...c5!?
This is sometimes referred to as the Arkell/Khenkin Variation. Although
considerably less explored than the bishop development, we believe it’s a
very viable option. However, supplementing it with a minor alternative for
Black proved extremely challenging. We rejected moves like 3...Na6,
3...g6 and 3...h5, which have been played by strong players but lead to
cramped positions that are probably very good for White and even harder
for Black to play well in practice. Instead we ended up with a small
experimental lesson on the virtually unexplored 3...Qc7!?. It probably isn’t
brilliant but has some ideas in common with 3...c5 while avoiding the
pawn-sacrifice aspect. For White can indeed cling on to the pawn:

4 dxc5 e6 5 Be3 Nd7 6 Bb5 Ne7 (D)

White to play

There may be theoretically more challenging lines, but gambit play isn’t
for everybody and some players are always sceptical about nebulous-
looking compensation. Objectively there is little doubt that this position is
fine for Black. But if before every game you worry about reaching a pawn-
down endgame, you may want to spend some extra time boosting your
confidence in Black’s compensation. Model Game 15 is a good starting
point.
The Exchange Complex

3 exd5 cxd5 (D)

White to play

This move introduces the Exchange Complex. It consists of two major

components leading to very different kinds of play – the Panov Attack
and the Exchange Variation. The good news for us is that in contrast to
many ‘exchange variations’, this one unbalances the pawn-structure and
usually leads to quite interesting positions.

4 c4

This is White’s most challenging move, turning the opening into the Panov
Attack (or the ‘Panov-Botvinnik’ as some prefer). White increases his
central presence by offering to play with an isolated queen’s pawn (IQP).
White’s less critical moves, and in particular 4 Bd3, are generally referred
to as the Exchange Variation. It is a popular suggestion in repertoire books
for White because his play is direct and thematic, and may lead to a
dangerous kingside attack if Black is careless.
4...Nf6

Theoretically 4...dxc4 can hardly be the critical test of White’s central

challenge. However, it has some points and in Lesson 24 we shall offer
some analysis of this risky move for those who appreciate the element of
surprise. The fact that this variation mostly arises by transposition from the
Queen’s Gambit Accepted should greatly increase Black’s chances of
being the one on more familiar ground.

5 Nc3 g6 (D)

White to play

For at least a century a kingside fianchetto has been recognized as the

ideal set-up against an isolated queen’s pawn. Here Black’s fianchettoed
dark- squared bishop exerts a lot of pressure on d4 and endgames tend
to be promising for Black.

While the mirror-image of this set-up – with an extra tempo – has made
the Tarrasch a rare guest on the absolute elite level, this version is quite
controversial because it usually involves a pawn sacrifice. We believe we
have made a strong case that Black’s game is perfectly sound in the
critical
lines in Lessons 22 and 23.

The Panov IQP

Sometimes a certain opening line may lead more or less by force to a

specific endgame. In that case it makes sense to include guidelines for that
endgame in a book on the opening. However, when that endgame is of a
general nature and may arise from many different openings, that’s not
always possible.
Perhaps the most important endgame of this kind in our book is the
IQP Panov endgame which Black heads for in Lessons 21 to 23.

White to play

You may recognize this as a mirror-image of the Tarrasch endgame,

and quite promising for Black. Some essential features resulting from
Black’s kingside fianchetto include:

Black’s winning chances are improved because his dark-squared

bishop may apply direct pressure to the d-pawn from a distance.
Black didn’t have to play ...e6 so he developed his light-squared bishop
 actively and later exchanged it for its white counterpart.

For additional information on how to handle this endgame, you may wish
to consult a book (or digital equivalent) focusing on the subject. Quite
likely it will also offer valuable insights in how to handle the typical
middlegames too.
1: Classical: Korchnoi Variation

Lesson 1: Introduction

By: TRH

Before the endgame, the Gods have placed the middlegame.

Siegbert Tarrasch

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6

Black immediately declares his willingness to accept doubled pawns in

order to unbalance the play. This makes it a more controversial move than
the main alternatives 4...Bf5 (the subject of Lessons 6 to 12) and 4...Nd7
(which we don’t cover).

5 Nxf6+ exf6 (D)

 White to play

For the last few years, this line has been my favourite of all openings. The
resulting positions typically lead to complex games. One advantage is that
you won’t get far by memorizing endless variations. Instead, you have to
develop your understanding and feel for the positions. Despite some ups
and downs the last years, I have had a stable relationship with the
Korchnoi Variation, and it keeps serving me good results. Before entering
the theory and model games, I shall share some of my experience in this
variation.

Control of Important Squares

Usually when playing 1 e4, White places his knight on f3. In the Korchnoi
Variation, this is not dangerous. The f6-pawn covers e5 and g5, making
the knight on f3 look a bit silly. While in other openings Black must be
careful about playing ...f6 as it weakens the e6-square, here we don’t
have such worries. If White’s knight ever tries to reach f5 via h4, we can
keep on dominating it by playing ...g6.

We all know that control of the centre is essential, and it’s easy to
understand why the e5-square is vital, but actually so is control of the g5-
square, as can be seen in this position:
 Black to play

Kizatbay – T.R. Hansen

Oslo 2017

Thanks to the f6-pawn, Black doesn’t have to worry about a knight jump
to g5. Therefore:

17...h5!?

I intend to break up White’s kingside with ...h4. In this structure, the move
...h5 doesn’t weaken anything, whereas without the extra pawn on f6 it
would permanently weaken g5 and hardly be possible at all. I won the game
after the further moves 18 h4 Qc8 19 Qa5 b6 20 Qc3 Bd6 21 Bg2 a5 22
Qc1 Bh3 23 Rxe8+ Rxe8 24 d5?? Bxg2 25 Kxg2 Nf4+! 26 Kh2 Qh3+ 0-1
and I believe this reflects an interesting point regarding the Korchnoi
Variation. In many lines, Black goes for the same set-up no matter how
White plays. It usually takes approximately 15 moves to get the pieces
where you want, and then you start adding pressure to White’s position.
When things start to go wrong for White, it usually goes badly downhill well
before an endgame is reached. A huge majority of my wins in the Korchnoi
last 25 to 40 moves,
confirming my impression that it’s easy for White to go wrong in
these middlegames.

Before moving on, let me point out that although 17...h5 was effective,
another idea is to continue tickling White’s kingside with 17...Qd7!?,
after which I believe Black is also for choice. Actually, there is even a
third interesting option here in 17...a6!?, which introduces an idea that
might be relevant in many similar positions: after the natural 18 Bg2 my
point is 18...Ba7!. When the d-pawn is unable to reach d5, it can easily
end up as a weakness. Here it’s undoubtedly a liability for White.

The move ...h5 is a frequent guest in this structure. This position is

another example:

Black to play

J. Christiansen – T.R. Hansen

Norwegian League 2017/18

With the queen on a4, White is exposed to kingside attacks.

16...h5!

The white king is the weakness that Black must target.

17 Rad1?

This allows Black a dangerous kingside attack. White must create

counterplay, so 17 c4! is a better move. A possible continuation is
17...Qf5 18 Rad1 h4 19 d5 with a very sharp and unclear game.

17...h4 18 gxh4 Re4 19 c4? Qh5

Black is clearly in the driving seat; indeed, his attack should prove decisive.

Model Game 1

A. Svensson – T.R. Hansen

Lund 2018

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4

This may seem mandatory, but you will actually find some gambit options
in 1A.

4...Nf6

If you enjoy complex middlegames, this move is a step in the right direction.

5 Nxf6+

One of the Korchnoi Variation’s strengths is that only this move can be
critical. The passive 5 Ng3 is frequently seen, especially at lower levels.
It’s not dangerous, and Black has many ways to equalize. My
recommendation is to fight immediately for the initiative with 5...h5!?. The
threat of advancing the h-pawn is annoying for White. See 1B for some
analysis.

5...exf6 6 Nf3 (D)

This natural move has been played by many strong players but the theory
isn’t well developed. This is mostly due to the non-forcing nature that the
play usually takes. We discuss the rare but interesting 6 Be3 and some
other rare 6th moves in 1C.

Black to play
6...Bd6

This game serves as a perfect example of what can go wrong for White if
he doesn’t handle the opening well. The way Black arranges his minor
pieces is the best way to play against many of White’s options. Black
should first seek safety for his king on the kingside. Thereafter Black
frequently plays the moves ...Re8 and ...Nd7-f8 before choosing a set-up
for the rest of his pieces, but this is far from his only plan.

7 Be2

The active 7 Bd3 is somewhat more popular. For analysis of this and
other 7th-move alternatives, see 1D.

7...0-0 (D)
 White to play

Part of the Korchnoi Variation’s attraction is the additional safety provided

to Black’s king by his extra f-pawn. We have already seen how this leaves
his
h-pawn free to play an active role as it’s usually superfluous in the defence.

8 0-0 Re8

Another of the advantages of the Korchnoi Variation is that many of

Black’s moves can be played more or less automatically. The rook
belongs on the open e-file and the f8-square is freed for the knight.

9 h3

This is not very common but it’s a semi-useful move that White frequently
plays quite soon, so after the more common moves 9 Re1 and 9 c4 there
are plenty of transpositions to positions we see later in the game.

It makes sense to avoid h3, but also in this case Black experiences no
problems; e.g., 9 Be3 Nd7 10 c4 Nf8 11 Qd2 Ng6 12 Rfe1 Bf5 13 Bf1
Qd7, Arutyunova-Stefanova, St Petersburg blitz 2018. For more on the
non- transpositional lines without an early h3, see 1E.
9...Nd7 10 c4

The somewhat paradoxical tempo-loss 10 Bd3 has been played by

quite strong players. For some examples and analysis of this move,
see 1F.

10...Nf8 (D)

White to play

So far Black hasn’t revealed much of his plans but mobilizing all the pieces
with the moves ...Be6, ...Qd7, ...Ng6 and ...Rad8 is a natural prelude to a
full- scale kingside attack.

11 Be3

Both players are choosing very normal-looking moves but we are now in
the outskirts of current theory. I had actually had this position in the earlier
game Kizatbay-T.R.Hansen, Oslo 2017. After the moves 11 Re1 Be6 12
b3 Qd7 13 Bb2?! Rad8?! (both players missed that 13...Bxh3! is very
strong here, albeit understandably as the variations are both complex and
open-ended) 14 Bf1 Ng6 15 Qd2 Bb8 16 Rad1 Qc7 17 g3 we had reached
the position discussed in the introduction to this lesson under the heading
‘Control of Important
Squares’.

11...Be6

Putting the bishop on this square is my preference, but the following game
with 11...Bf5 saw Black defeating a higher-rated player in typical fashion:
12 Qd2?! Ng6 13 Rac1?! Qd7! 14 d5 c5 and at first sight, it looks like White
has achieved something with a protected passed pawn on d5, but this
won’t count before the endgame. In the middlegame the stable structure
and the access to the e5-square are beneficial to Black. Meskenas-
Bagdanskis, Plateliai 2019 continued 15 b4 b6 16 bxc5 bxc5 and White
couldn’t stop the sacrifice on h3: 17 Rfe1 Bxh3 18 gxh3 Qxh3 19 Bd1 Re4
20 Ng5 Qh2+ 21 Kf1 Qh1+ 0-1 (White is mated after 22 Ke2 Nf4#).

We now return to 11...Be6 (D):

White to play
12 Qd2?!

12 Qc2 gives Black a couple of options with his queen:

a) After 12...Qc7 13 Rfd1 Ng6 14 Bd3 Rad8 15 a4, as in S.Næss-Sidali,
Oslo 2017, 15...Qd7 and 15...Qc8 both oblige White to continue 16 d5,
with unclear play. 15...Qa5!? intends ...Qh5 and, if White parries that, then
...Bb8 and ...Qc7 is a clever regrouping idea.

b) The direct 12...Qd7!? is very dangerous for White as there is no

obvious way to prevent the sacrifice on h3. However, there is a narrow
path to survival: 13 Rfd1! Bxh3 (Black can play more slowly, of course) 14
gxh3 Qxh3 (Black threatens ...Ng6-h4) 15 c5! Bc7 16 d5 cxd5 (there’s no
time for 16...Ng6? due to 17 d6 +–) 17 Rxd5 and thanks to the resource
Qf5, Black is forced to give perpetual check.

12...Qd7!

Now ...Bxh3 is an immediate threat, and White must scramble to defend.

13 Rad1? (D)

After 13 Bd3, the sacrifice seems to fail, so I would have to bring in

the reserve troops with 13...Ng6 and ...Rad8 before contemplating
anything violent.
 Black to play
13...Rad8

This being a rapid game, I chose to complete development rather than

risk missing a defensive resource after 13...Bxh3! 14 gxh3 Qxh3, but
Stockfish assures me that this is a clear and safe win for Black.

14 Rfe1?

Even though not a single black piece has so far crossed his third rank,
Black now has a winning sacrifice. But there is no fully satisfactory
defence:

a) 14 Bf4? Bxf4 15 Qxf4 Bxh3 is much better for Black.

b) 14 d5 cxd5 15 Qa5 (after 15 cxd5?, the standard 15...Bxh3 works

again) 15...Bxh3 16 gxh3 Qxh3 17 Rd4 g5 and once again Black has an
obvious advantage.

14...Bxh3!

As you may already have discovered, this is a very typical sacrifice in

the Korchnoi Variation.

15 gxh3

After 15 Bf1 Bg4 16 Be2 Ng6 Black’s continuing kingside initiative in

combination with his extra pawn ensures a winning advantage.

15...Qxh3 (D)
 White to play

What makes this attack fascinating is how difficult it is for White to bring
defenders over to the kingside. Moreover, he has no way to prevent
Black from swinging his own rooks into the attack.

16 Bf4 Re4!

This forces White to allow another rook into play.

17 Bxd6 Rxd6 18 Nh2 Rh4 19 Bf3 f5! 20 Re8 Rg6+ 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 (4 Nxe4 Nf6 5 Nxf6+ exf6 6 Nf3 Bd6 7 Be2 0-0
8 0-0 Re8)

1A: (4 f3?!)

1B: (5 Ng3)
1C: (6 Be3)

1D: (7 Bd3)

1E: 9 Re1

1F: 9 h3 Nd7 10 Bd3

The current lesson isn’t very theoretical, so there isn’t much hard work
demanded here. If you are new to the Caro-Kann, the best you can do at
this point is probably to get on with Lesson 2, which is quite challenging.
But when you find the time for detailed preparation of side-lines, here are
some early deviations and our suggested antidotes.

1A: Blackmar-Diemer and Rare 4th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4)

Instead of the obvious recapture of the e-pawn, White may offer a gambit:

4 f3?!

4 Bc4?! Nf6 will usually transpose after 5 f3 exf3 6 Nxf3, as 5 Bg5?! b5 is

probably weaker; e.g., 6 Bxf6 exf6 7 Bb3 Bb4 (7...f5 is also strong) 8 a4 0-
0 9 Ne2 bxa4 10 Rxa4 a5 and Black was more or less winning in
Omanović- Dizdarević, Sarajevo Ch 2018.

4...exf3 5 Nxf3

For his pawn, White has roughly sufficient compensation in the form
of better development.

Q: So, is the gambit correct?

Not really. White starts the game with some plus. So if he is now equal (or
slightly worse), he is a bit worse off than he could be. However, if we
factor in that most club players attack better than they defend and that
Black may become nervous or overly optimistic, the gambit may not be
such a bad
practical decision below GM level.

5...Nf6 (D)

White to play

This position belongs to the Blackmar-Diemer Gambit (BDG among

aficionados) which normally arises from the 1 d4 d5 2 e4 dxe4 3 Nc3 Nf6
4 f3 exf3 5 Nxf3 move-order. Then Black can reach the diagram position
with 5...c6.

Q: That doesn’t seem a very natural move. From that move-order I

would hurry to develop with a move like 5...Bg4 or 5...Bf5 or perhaps
prepare kingside castling with 5...e6 or 5...g6.

I understand your concern, but although 5...c6 is only 5th in popularity

(after the moves you mention), Black has not been tricked into an inferior
line against the gambit. Actually it’s known as the Ziegler Defence and is
Black’s best-scoring reply! The point is that having sacrificed a pawn,
White is obliged to develop at lightning speed, but he doesn’t yet know
where his light-squared bishop belongs. After ...e6 it’s clear that it should
be on d3, while after ...g6 or ...Bf5 it just as clearly belongs on c4. Now
White has to
declare his intentions.

6 Bc4

Against 6 Bd3?!, 6...g6 seems logical given what we just said. Black
should prepare for kingside castling but perhaps delay it a while until the
plan Qe1- h4 plus Bh6 isn’t effective any more:

a) After 7 0-0 Bg7 8 Qe1 Bg4, some typical lines are:

a1) 9 Qh4 Nbd7! is untested but looks fine for Black. The following line is
harmless but illustrates a few tactical points that you should be aware of:
10 Ng5 Qb6 11 Nxf7?! Qxd4+ 12 Kh1 Kxf7 13 Ne2 Qd5 14 Qxg4 Ne5 –
/+.

a2) 9 Be3 Nbd7 10 h3 Bxf3 11 Rxf3 0-0 –/+ Wundt-B.Knorr, corr. 1992.

a3) 9 Bg5 gives Black a choice between the cautious 9...Nbd7?! 10 Qf2
Bxf3 11 Qxf3 0-0 12 Rae1 (Diemer-Gunderam, corr. 1983) 12...Qb6 –/+
and simply taking a second pawn with 9...Bxf3! and ...Qxd4, which
objectively should just be winning (10 Qe3 doesn’t help in view of 10...Ng4
or 10...h6).

b) 7 Bg5 is Scheerer’s recommendation but after 7...Bg7 8 Qd2 Bg4 it’s

hard to believe in White’s chances:

b1) 9 Ne5 Be6 10 0-0-0 Qxd4 11 Nf3 Qb4 12 a3 Qa5 13 Nd4 Nbd7 –/+
P.Leisebein-Tauscher, corr. 2013.

b2) Scheerer gives the unlikely retreat 9 Be2 0-0 (also 9...h6 10 Be3 Nbd7
is fine for Black; e.g., 11 h3 Be6 12 0-0 Nb6 13 Ne5 Nbd5 14 Nxd5 Nxd5
15 Bf2 Bxe5 16 dxe5 g5 and Black has some plus) 10 0-0-0 Nbd7 (the
direct 10...b5 is probably even stronger) 11 h3 Be6 12 g4?! b5 13 Kb1?!
b4 14 Na4 Qa5 15 b3 (Stummer-Shevchenko, corr. 1993) and now
15...Nd5 16 Qd3 N7b6 is maybe the most convincing of Black’s many
winning lines.

We now return to 6 Bc4 (D):

 Black to play
6...e
6

It may be tempting to bring out the light-squared bishop first (and 6...Bf5
is indeed a respectable choice) but it might become a target on f5, and it
looks sensible to cover against potential f7 tactics.

7 0-0

White has also tried to aim for queenside castling:

a) 7 Bg5 Be7 8 Qd2 Nbd7 9 0-0 (9 0-0-0?! b5!) 9...0-0 10 Bd3?! c5 –/+
Szenetra-Deckwerth, 2nd Bundesliga 1980/1.

b) 7 Qe2 Be7 8 Bd2 Nbd7 9 0-0-0 (Puig Pulido-Golombek, Madrid

1959) 9...b5 10 Bd3 b4 11 Ne4 Nxe4 12 Qxe4 a5 leads to a very sharp
position where it may be a plus for Black not yet to have committed his
king.

7...Be7 8 Bg5 Nbd7 9 Qe1 h6!

9...0-0?! could give White a dangerous attack after 10 Bd3!.

10 Qh4?! (D)

10 Bd2, planning Qg3, is a better try.

Black to play

In this position from Sirve-Pakarinen, Tampere 1998, White has some play
for his pawn – it could even be enough to save the game – but Black must
be for choice after 10...b5 11 Bd3 Bb7 12 Ne4 a6 13 Rae1 c5; e.g., 14
Nxc5 Nxc5 15 dxc5 Bxf3 16 Rxf3 Qd5.

1B: 5 Ng3 and Rare 5th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4)

4 Nxe4 Nf6 (D)

 White to play
5 Ng3

This is a harmless alternative to 5 Nxf6+. Even rarer moves:

a) 5 Nc3 makes little sense. There is no need to memorize variations in

case of this move, as Black can simply play 5...Bf5 (see also Game 6) 6
Nf3 e6 7 Bd3 Bxd3 8 Qxd3 Nbd7 with a very solid position. Black also has
good long- term prospects thanks to his potentially better bishop, and
White’s difficulty in finding a plan. A way to go for Black is to finish
development, castle and then play for ...c5.

b) While 5 Ng5 makes a lot of sense in the Karpov Variation (4...Nd7),

here it fails to impress. 5...c5 (there is no need to spend time chasing the
knight away) 6 Bc4 e6 7 Ne2 Nc6 and sadly for White, the knight is clearly
misplaced on g5. Koykka-Šulskis, Jyväskylä 2012 continued 8 Nf3 Bd7 9
c3 cxd4 10 Nexd4 Nxd4 11 Nxd4 Qc7 and it looked like a French Tarrasch
where White was a couple of tempi behind.

c) 5 Qd3 has been mostly ignored by strong players but looks sensible. I
suggest meeting this rare move with another rare move, namely 5...g6.
Then
Black seems fine:

c1) 6 Nxf6+ exf6 can’t be dangerous compared to the main lines of the
Korchnoi Variation. White’s queen is misplaced on d3, while the pawn
is often useful on g6.

c2) After 6 Nf3 Bg7 7 Nxf6+ exf6 the only test is 8 Qa3 to try to prevent
Black from castling. However, it cannot be worth moving the queen three
times only for it to be exchanged: 8...Qe7+ 9 Qxe7+ Kxe7 10 Be2 Re8 11
0-0 Kf8 12 Re1 Be6 =.

5...h5!? (D)

White to play

This move lays a nasty trap, and the good thing is that it’s a 100%
sound move too.

6 Bg5?

White misses the danger. He has a few better responses:

a) A superior way to ignore Black’s threat of advancing his h-pawn is 6
Bc4. However, after 6...h4 7 N3e2 Bf5, as in S.Das-Nisipeanu, Dubai
2012 (and many other games), Black is comfortable.

b) 6 h4 has the drawback of weakening the g4-square. 6...Bg4 7 N1e2

(other moves are no better: 7 f3 is best met with 7...Bd7 followed by ...e6
and ...c5, while 7 Be2 can be answered with 7...e6 8 Bxg4 hxg4 9 N1e2,
as in Plat- Dorner, Wroclaw rapid 2014; e.g., 9...c5 10 c3 Nc6 =+) and
after the new move 7...g6 the white knights look silly as they are far away
from the weakened g5-square.

6...h4 7 Bxf6??

One of the reasons why I can’t resist the aggressive h-pawn thrust is
the chance that White will fall into this beautiful trap.

7...hxg3 8 Be5 (D)

Black to play

Black wins by 8...Rxh2! 9 Rxh2 Qa5+! 10 c3 Qxe5+! 11 dxe5 gxh2,

when amazingly White has no defence – Black ends up with an extra
bishop.
1C: 6 Be3 and Other Rare 6th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6)

5 Nxf6+ exf6 (D)

White to play
6 Be3

Not a bad move, but it fails to pose Black any problems. The main idea is
to overprotect the d-pawn so that White can play 7 Bd3. Other moves:

a) It is possible to sacrifice the d-pawn, and indeed 6 Bd3?! Qxd4 7 Nf3

worked out perfectly for White in Sevian-Grandelius, Internet blitz 2020
where Black chose the passive 7...Qd8 and White achieved good play for
the pawn. The active 7...Qd5 is better; for instance, 8 0-0 Be6 9 Bf4 Nd7
when White has some compensation, but nothing more than that.

b) 6 h3 aims to play 7 Nf3 without having to worry about ...Bg4. However,

as you will see, ...Bg4 is usually not part of our plan anyway. After 6...Bd6 7
Nf3 0-0 8 Be2 Black is doing perfectly fine, and may attack White’s
kingside in standard fashion.

c) 6 g3 looks a bit strange as there are more natural ways to develop the
bishop. After 6...Bd6 7 Bg2 0-0 8 Ne2 Re8 9 0-0 Bg4 White is potentially
weak on the light squares. Montoliu Cervero-Salgado Lopez, Linares rapid
2016 continued 10 Re1 Na6 with a roughly equal position, but from here on
it went quickly downhill for White: 11 a3 Nc7 12 Qd3 Qd7 13 Bd2 Bf5 14
Qb3 Ne6 15 Be3 Ng5 16 Nf4? Be4 17 Bxe4? Rxe4 18 Kg2 Rae8 19 Rf1
Bxf4 20 Bxf4 Qh3+ 21 Kh1 Rxf4 0-1.

d) After 6 Be2 Bd6 7 Bf3 0-0 8 Ne2 Black played 8...Qc7 in

Nozdrachev- V.Moiseenko, Voronezh 2017, which is perfectly fine.
Another option is 8...Re8, followed by the standard knight manoeuvre
to f8.

e) With 6 Ne2 White is intending to fianchetto the bishop, but this is not
the best way to develop. After 6...Bd6 7 g3 0-0 8 Bg2 Bg4 9 Be3 (I
suspect Black was intending to meet 9 0-0 with 9...Qd7 followed by
...Na6) 9...Re8 10 Qd2 Nd7 Black was already better in J.Petr-J.Colas,
Pardubice 2019. The game continued 11 0-0?! Nb6 12 b3 Nd5 13 c4 Bb4
14 Qd3 Nxe3 15 fxe3 Qe7 16 e4 f5 17 e5 Rad8 (White’s centre is
collapsing) 18 Bf3 Qxe5 –+.

f) 6 Bf4 Bd6 (D) and now:

 White to play

f1) White has tried 7 Bxd6 Qxd6 8 c3 0-0 9 Bd3 Re8+ 10 Ne2 Bg4 11 f3
(11 h3 Bxe2 12 Bxe2 Qe7 is awkward for White) 11...Be6 12 Qc2 g6 13 0-
0 Nd7 and Black was doing fine in S.Dražić-Blübaum, European Ch,
Skopje 2019.

f2) 7 Qd2 0-0 (one advantage of the Korchnoi Variation is that Black’s king
is quickly brought to safety; in the line that follows, we see the dangers for
White if he delays castling) 8 Bd3 Na6 9 c3 (9 Bxa6?! Bxf4 10 Qxf4 Qa5+
is a trick worth remembering! 11 Qd2 Qxa6 =+ gives Black a bishop versus
a knight, and White’s king is, at least for the moment, stuck in the centre)
9...Re8+ 10 Ne2 leaves White one move away from castling, and Black
tries to prevent that: 10...Bxf4 11 Qxf4 Nc5 12 Bc2 Qe7 13 Qf3 Ne6 14
Bd3 and finally White is ready to bring his king to safety. After 14...b6
(Black has many alternatives to this move, such as 14...Bd7 and 14...Ng5,
both leading to equality) 15 0-0 Bb7 16 Qg4 g6, the position was equal,
and Black eventually won in Viczencz-Buzas, Balatonalmadi 2019.

6...Na6 (D)
 White to play

The knight heads for d5 via c7, or, if given the chance, via b4 to take
advantage of White’s early commitment of his dark-squared bishop.
Now:

a) 7 Bxa6 Qa5+ gives Black an advantage as he gets the bishop-pair

without compromising his structure.

b) 7 Qd2 was played in Lauk-V.Zhuravliov, Vilnius Zonal 1993, and here

I suggest 7...Nb4 threatening 8...Bf5. After 8 c3 Nd5 Black can pick up
the bishop whenever he wants to, and is already more comfortable.

c) 7 Bd3 Nb4 8 Ne2 and in Timmerman-Boersma, Enschede 1993

Black played the ambitious and creative 8...h5!?, although after the
simple 8...Nxd3+ Black is better.

d) 7 c3 prevents ...Nb4 but is the very move White hoped to save with his
6th move. After 7...Nc7 8 Bd3 Nd5 Black will develop his kingside,
reaching a fine position.

1D: 7 Bd3 and Rare 7th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6)

6 Nf3

I believe Black should be happy to see this move. First of all, the pawn on
f6 indicates that f3 is not the best square for the knight. Secondly, and
probably more important, White’s move does not prevent Black’s desired
set-up.

6...Bd6 (D)

White to play
7 Bd3

A natural square for the bishop, and somewhat more common than the
modest 7 Be2, which we examined in Model Game 1. Rarer options
include:

a) After 7 h3 0-0 8 Be3 Re8 9 Be2 Nd7 10 0-0 Nf8 11 c4 we have

transposed to Model Game 1.

b) 7 Qe2+? is pointless as Black has 7...Be6.

c) 7 Be3 hints at queenside castling. After 7...0-0 developing the light-
squared bishop will take us back to lines we have already looked at. 8 Qd2
Re8 9 0-0-0 Be6 already left White in trouble in A.Stanciu-Greibrokk,
European Girls Under-18 Ch, Riga 2018. Compared to the Hansen
Variation, a main line we will investigate in depth later in this book, Black
has avoided the double-edged ...h5, and simply has the better attacking
chances with opposite-side castling here.

d) 7 Bc4 targets the f7-pawn, but this is not so critical with the moves Nf3
and ...Bd6 included. White has lost the possibility of throwing in the
disruptive check on e2 (which is the subject of Lesson 2), as Black is just in
time to castle. After 7...0-0 8 0-0 the most common move is to pin the
knight by 8...Bg4, but you will see that I usually want to delay this move to
tempt White into playing h3. Instead I prefer 8...Re8, when after 9 c3 Black
has a pleasant choice between 9...Bg4 and 9...Nd7, in both cases with a
good game. In case of 9 h3, Black has 9...Nd7 with the well-known plan of
...Nf8, ...Be6,
...Qd7, threatening to sacrifice on h3. This attacking scheme normally
gives White a lot of headaches.

7...0-0 8 0-0 (D)

 Black to play
8...Re8

8...Bg4 used to be my preference, but there is no need to rush with this move.

9 h3

Or:

a) 9 Re1 Bg4 10 Rxe8+ Qxe8 11 Be3 Na6 12 h3 Bh5 13 c4 Rd8 14 a3

Nc7 15 b4 b5 16 Qe2 bxc4 17 Bxc4 Qd7 18 Re1 Re8 =+ Muñoz Pantoja-
Narciso Dublan, Barcelona 2014.

b) 9 c3 Nd7 10 Qc2 Nf8 11 Be3 and in one of my games I chose

11...Bg4. Instead, Black has a very interesting idea that has never
crossed my mind before: 11...Qa5 intending ...Qh5 followed by ...Bg4,
and there is not much White can do against it. For instance, 12 h3 Qh5
13 Qd1 gives Black a good position. 13...Ng6 is unproblematic for Black,
while 13...Bxh3!? is at least a draw, as after 14 gxh3 Qxh3 15 Re1 Black
can give perpetual check or continue the fight with 15...g5!?.

9...Nd7 10 c4 Nf8 (D)

 White to play

Even though I am a big fan of the black side of this variation, I accidentally
stumbled into this position with White in 2014. Black’s idea is to play
...Be6,
...Qd7 and sacrifice on h3. I was not aware of this idea at the time, but still
sensed the danger. After 11 Qc2 Be6, 12 Bf5 was not very ambitious but at
least I stopped Black’s plan (instead 12 Re1 Ng6 13 Bd2 Qd7 14 Be4
allows a typical attack: 14...Bxh3 15 gxh3 Qxh3 16 Re3 Nf4 17 Ne1 Ne2+
18 Rxe2 Bh2+ 19 Kh1 Bg3+ ½-½ Puliaev-V.Osipov, Russia Cup, Tomsk
2002). After 12...Qd7 13 Bxe6 fxe6 (13...Nxe6 would have been my choice,
and if allowed continue with ...Rad8) 14 Be3 Rad8 15 Rfd1, the position
was equal and quickly drawn in T.R.Hansen-Prohaszka, Oslo 2014.

1E: 9 Re1

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Nf3 Bd6)

7 Be2 0-0 8 0-0 Re8 9 Re1

In this line where White castles kingside many transpositions are

possible. However, they are all easy to meet as the standard set-up
works perfectly.
9...Nd7 (D)

White to play
10 Bf1

White is eager to exchange pieces. Some more enterprising continuations:

a) 10 c3 Nf8 11 Be3 Be6 12 Qc2 Qd7 13 Rad1 (Aasen-Sidali, Helsingør

2017) 13...Ng6 is slightly more comfortable for Black as White’s set-up
is toothless.

b) 10 Be3 Nf8 11 Qd2 Ng6 (this move is not strictly necessary, but it
makes sense to prevent White from playing Bf4) 12 c4 Bg4 13 h3 =
Mrkonjić- Cvorović, Croatian Team Ch, Šibenik 2005.

c) 10 Bd3 Rxe1+ 11 Qxe1 Nf8 and now:

c1) 12 Bd2 Bg4 13 Qe4 (Kolago-Antoniewski, Polish Team Ch,

Dzwirzyno 2019) and here I suggest 13...f5 14 Qe2 Ne6 15 c3 Qb6 16
Rb1 g6 =.

c2) After 12 h3, Black has more than one good move, but I like 12...Be6 13
Bd2 Bc7 intending ...Qd6, ...Re8 and ...Ng6 with a good game.

d) 10 c4 Nf8 11 h3 Be6 12 Be3 Qd7 13 Bf1 (Kollars-Feuerstack,

Hamburg 2016) and Black should continue 13...Rad8, followed by ...Bb8
and ...Qc7, with an eventual ...Ng6-h4 creating a lot of pressure.

10...Rxe1 11 Nxe1 Nf8 12 Nd3 Bf5

Black has no problems whatsoever; White is far away from proving

an opening advantage.

13 Be3 Qd7 14 Qd2 h5

A little spoiler – you will see more of Black’s last move!

15 Bf4 Be7 16 Re1 Ne6

Black is comfortable, and in Wu Xibin-Li Chao, Chinese Team Ch 2019,

eventually outplayed his lower-rated opponent: 17 c3 Rd8 18 Rd1 h4 19
Be3 Qc7 20 Qe2 g5 21 Nc5 Ng7 22 h3 Bg6 23 Re1 Re8 24 Qd2 b6 25
Nd3 Bd6
26 Be2 Be4 27 Qd1 Qe7 28 Bf3 f5 29 Ne5 Bxe5 30 dxe5 Bxf3 31 Qxf3 f4
32 Bc1 Qc5 33 Qe4? f6 34 Rd1 Rxe5 35 Rd8+? Kf7 36 Qh7 Re1+ 37 Kh2
Qxf2 38 Rd7+ Ke8 0-1.

1F: 9 h3 Nd7 10 Bd3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Nf3 Bd6 7 Be2 0-0 8
0-0 Re8)

9 h3 Nd7 10 Bd3 (D)

 Black to play

White has lost a tempo playing Be2-d3, but his pieces are now
reasonably active. Some players as White may now expect to exchange a
number of pieces and move into a better endgame. Well, that’s not how
we expect play to develop!

10...Nf8

We go on with our standard piece deployment. While h7 may not need

protection for the moment, the knight is also slowly moving towards
White’s king.

11 c4

White ambitiously occupies the centre while setting his queenside majority
in motion. Some less optimistic tries:

a) 11 Re1 Rxe1+ 12 Qxe1 Ng6 13 Be3 Be6 14 Qd2 and here instead
of 14...Qc7 (Oparin-S.Golubov, Moscow blitz 2016), I like 14...Qd7,
when White has to be careful not to allow ...Bxh3.
b) After 11 c3 Be6 White has a difficult task:

b1) 12 Nd2 Bc7 13 f4?! c5 14 f5? Bd5 15 Nc4 cxd4 16 cxd4 Bc6 17 Be2
Qd5 18 Rf3 Rad8 –+ A.Mnatsakanian-Harutjunyan, Poti 2018.

b2) 12 Be3 Qc7 13 Nd2 Ng6 14 Qh5 Bf4 15 Bxf4 Qxf4 ½-½ I.Mayer-
Borda, Budapest 2011.

b3) 12 Nh4 g6 13 Bd2 f5 14 Nf3 Bd5 15 c4? Bxf3 16 Qxf3 Ne6 17 Bc3?
and in Rakić-Bertok, Yugoslav Team Ch, Pula 1971, the players agreed a
draw following 17...Bf8?. Too bad, as after 17...Ng5 18 Qd1 Nxh3+ 19
gxh3 Qg5+ 20 Kh1 Qf4 White is mated!

b4) 12 Re1 Qd7 13 Be3 Ng6 14 Bf1 (Monpeurt-Steil Antoni, French

Team Ch 2019) and now 14...Rad8 reaches Black’s ideal set-up, with a
pleasant advantage.

11...Be6 (D)

White to play

This would be my choice: Black simply goes for the ideal set-up and takes
it
from there. Next up is ...Qd7 and ...Rad8.

12 Re1

After 12 Be3 Qd7 13 Re1 Ng6 14 Bf1 we have a choice:

a) After 14...Re7 15 Qc2 Rae8 16 Rad1 Bf5 17 Qb3 White ended up

winning in Teichmann-Tartakower, Karlsbad 1911. However, Black has
every reason to be happy after 17...Bb8. With moves like ...Be4 and ...Qd6
coming up, White has to be careful.

b) I recommend 14...Rad8 as in this structure I believe Black should try to

benefit from having two files for his rooks. A possible continuation is 15
Qc2 Bb8 16 Rad1 Qc7, when Black is about to create a typical kingside
attack. The threat is ...Nh4.

12...Qd7

I feel this is the most practical continuation. It is the most flexible move,
and I believe Black should finish development before launching an attack.
In B.Kovanova-K.Abdulla, European Women’s Ch, Riga 2017 Black
speeded things up with 12...Ng6 13 Bd2 Bc7 14 Bf1 Qd7 15 Bc3?! (the
bishop needs to protect the f4-square), when Black has various paths to a
slight advantage, such as 15...Qd6, probably followed by ...Rad8, and only
then considering how to start the attack.

13 Bd2

This position has occurred in two games, and in both Black played
13...Ng6. This is not a mistake, but I believe Black, if allowed, should
develop his queen’s rook as early as possible. After all, it’s not clear how
White should parry a slow development of Black’s attack. After 13...Rad8
a possible continuation is 14 Qb3 Ng6, with the ideal set-up for Black.
From here there is a pleasant choice of playing ...Bb8 and ...Qc7, or
immediately occupying the f4-square with ...Nf4. Either way, Black has
absolutely no worries out of the opening.

Lesson 2: Forgacs’s 6 Bc4

By: TRH

As a rule, the more mistakes there are in a game, the more memorable
it remains, because you have suffered and worried over each mistake
at the board.

Viktor Korchnoi

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6

Here we shall take a closer look at...

6 Bc4 (D)

Black to play

This is the Forgacs Variation, which used to worry me somewhat as the

queens tended to get exchanged on the e-file in the old main lines after
6...Qe7+. Skipping the middlegame and going directly for the endgame
has never been to my liking in general, and it is potentially dangerous for
Black due to the doubled f-pawns. Therefore I am happy that there is a
real fighting
alternative:

6...Bd6

Now Black is ready to castle kingside, something White should try

to prevent.

7 Qe2+

This is the critical variation, as White would welcome a queen exchange.

7 Ne2 shouldn’t divert us from our standard set-up, but 7 Qh5 is

another matter, as there are lines where the black king is chased
across the board, while White’s king and queen can also get in trouble.
See 2B.

7...Be7!? (D)

Exchanging queens with 7...Qe7 is possible and Black seems to do alright

in practice.

White to play
However, I prefer this more combative continuation, which leads to more
complex positions. It’s true that Black lost a tempo moving the bishop
twice, but as we shall see, the e2-square is not the best place for the white
queen. In practice these two suboptimal moves will often cancel each
other out.

Model Game 2

Bocharov –

Malofeev Peterhof

2007

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4 Bd6 (D)

If you dislike the potential tempo-loss inherent in this move, there is

some information on the even rarer 6...Na6!? in 2A.

7 Qe2+

This is the critical move. 7 Ne2 and 7 Qh5 are both discussed in 2B.

7...Be7!? 8 Nf3

8 Be3 is a move normally connected to queenside castling. As we shall

see, with the bishop on c4 this plan can easily turn out problematic for
White.
After 8...0-0 9 0-0-0?! (for 9 Nf3 see the note to White’s 9th move in
the game), 9...b5 (D) starts an effective attack:
 White to play

a) 10 Bb3? a5 leaves White in trouble:

a1) 11 a4 bxa4 12 Bxa4 (Asztalos-Tartakower, Vienna 1917) and here

Black can practically decide the game with 12...Qd5, threatening to take
on g2 as well as playing ...Qa2.

a2) After 11 a3 a4 12 Ba2 b4 Black was already close to winning

in T.Schuster-G.Pfeiffer, West German Ch, Höchst 1955.

a3) 11 c4 a4 12 Bc2 bxc4 13 Qxc4 Be6 14 Qd3 g6 15 a3 Nd7 16 Ne2 Nb6

–/+ E.Player-J.Houska, Douglas 2016.

b) 10 Bd3 Be6 and then:

b1) 11 Nf3?! Na6 (11...Qd5 –/+) 12 h4? (12 c3 Re8 =+) 12...Nb4 –+
B.Pineda-Frey Beckman, Cienfuegos 1976.

b2) 11 a3!? has the point that 11...Qd5?! 12 Nh3! is OK for White because
either pawn-grab (on g2 or b2) is fatal, and otherwise the game looks
unclear. I therefore suggest immediately attacking the queenside with
11...b4 12 a4
(12 axb4?! a5! gives Black a dangerous initiative) 12...Nd7 13 Nh3 Bd6
=+, when I am sure White will regret having the pawn on a4.

We now return to 8 Nf3 (D):

Black to play
8...0-0 9 0-0

At this point White could consider castling queenside, but it’s not to
be recommended as 9 Be3 Nd7 10 0-0-0?! b5 11 Bd3 Nb6 gives
Black the advantage.

In this game, the position after White’s 9th move was actually reached via
the move-order 2 Nc3 d5 3 Nf3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4 Bd6 7
Qe2+ Be7 8 0-0 0-0 9 d4.

9...Re8

When the rook appears on e8, White’s queen starts to feel uncomfortable
on e2. Black will continue by developing his queenside, and eventually the
queen’s placement will become a problem. White has a few ways to
develop,
none of them problematic for Black.

10 Re1 Nd7 11 Bd2

This probably isn’t best. For the more theoretically challenging moves
11 Bd3 and 11 c3, see 2C.

11...Nf8 12 h3 Bf5 13 Bf4

Moving the bishop for the second time can hardly be critical, but
nonetheless Black has no problems.

13...Qd7 (D)

White to play

Black has had no problems equalizing and an interesting game is to

be expected.

14 Rad1??

White brings his last piece into play with a horrible blunder.
14...Bb4

Let’s stop here. The game lasted for some more moves, but White had
no compensation for the exchange and Black’s victory wasn’t in doubt.
Although this was a rather silly blunder by White, it illustrates nicely why we
chose 7...Be7. At first sight, it seemed like Black loses a tempo, but as this
game shows, White’s queen isn’t well placed on e2.

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4 (6...Bd6 7 Qe2+

Be7 8 Nf3 0-0 9 0-0 Re8 10 Re1

Nd7) 2A: 6...Na6

2B: (7 Ne2)

2C: (11 Bd3)

If for some reason you don’t like my main recommendation, you should
take a closer look at 2A. I am sure it will be a surprise to your opponent.

Magnifier 2B takes on White’s alternatives to the main line. These are not
challenging to Black. It wouldn’t be my main focus when studying this
book, but it may help improve your general understanding.

2A: 6...Na6

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4)

6...Na6 (D)
 White to play

7 Ne2

Or:

a) 7 Bxa6?! Qa5+! =+.

b) 7 Nf3 Bd6 8 Qe2+ Qe7 (8...Be7!? also seems playable) 9 Qxe7+ (9

0-0 Qxe2 10 Bxe2 Bg4 shouldn’t worry Black) 9...Bxe7 10 Bf4 Be6 11
Bxe6 fxe6 12 0-0-0 Kf7 13 Rhe1 Rad8 14 c3 c5 = Linchevsky-Oparin,
Russian Team Ch, Sochi 2019.

c) 7 Qe2+ Be7 and now 8 Nf3 0-0 9 0-0 Re8 10 Re1 Nc7 11 Qd3 a5 12 a4
Nd5 = K.Martins-M.Drus, corr. 2011 or 8 Bxa6 bxa6 9 Nf3 0-0 10 0-0 h6
11 Bf4 Re8 12 Qd3 Qd5 13 c4 Qf5 14 Qxf5 Bxf5 = S.Henderson-
H.Buchmann, corr. 2011.

7...Nc7 8 0-0 Bd6 9 Bf4 0-0

Now:
a) 10 c3 was played in Deveque-Velten, Cappelle-la-Grande 2020.
Here Black has many possibilities and the easiest is probably
10...Be6 =.

b) 10 Qd2 Be6 11 Bd3 Re8 12 Rfe1 g6 13 c4 a5 14 a3 Qd7 (14...b5

looks unclear) 15 Rad1 += Rõtšagov-Kiik, Finnish Team Ch 2005/6.

c) 10 Re1 Be6 11 Bb3 Re8 12 Qd2 a5 13 c3 a4 14 Bxe6 ½-½

Savanović- Dizdarević, Bosnian Team Ch, Jahorina 2018.

2B: 7th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4)

6...Bd6 (D)

White to play
7 Ne2

You shouldn’t be shocked by 7 Qh5, an aggressive move tried by Keres

and Tal among others. After 7...0-0 Black should be able to neutralize any
danger with a few precise moves:
a) 8 Be3 Nd7 9 Nf3 (Sevillano-La Rota, Philadelphia 2006) 9...Nb6 10
Bb3 g6 11 Qh6 Nd5 gives Black the more comfortable position.

b) 8 Ne2 g6 and now:

b1) 9 Qf3 Re8 10 Bh6 Qa5+ 11 Bd2 Qf5 12 Qxf5 Bxf5 =+ L.Espig-Möhring,
East German Ch, Weimar 1968. White’s problem is that 13 0-0-0?? loses
to 13...b5.

b2) 9 Qh6 should probably be met by 9...Bf5; e.g.:

b21) After 10 h4? Bxc2 11 h5 g5 12 f4 Be7 –/+ White’s queen is in trouble.

b22) 10 Bb3 is best met by the immediate 10...Re8! (since 10...a5?! 11

a4 Na6 12 0-0! Re8 13 Bxf7+! is enough for crazy equality after 13...Kxf7
14 Qxh7+ Ke6 15 d5+! Kxd5! 16 Bf4!, when further superhuman play
should lead to perpetual check), when White lacks a sensible way to
develop (11 Be3? Re4! threatens to trap the queen with ...Bf8), while 11
Bxf7+ Kxf7 12 Qxh7+ Ke6 13 d5+ Kxd5! doesn’t give White quite enough
for the piece since 14 Qxb7 Rxe2+! =+ forces a favourable exchange of
queens.

b23) 10 0-0 Bxc2 11 Bf4 Bxf4 12 Nxf4 b5 and here too White doesn’t
get sufficient compensation.

7...0-0 (D)

Black has alternatives, including 7...Qc7, which delays White’s castling, but
I see no point in making things complicated in such a minor line.
 White to play

8 0-0

Or:

a) 8 Qd3 is well met by 8...b5 9 Bb3 Qc7, when White has some issues
with his king still in the centre.

b) 8 Bf4 looks wrong as it exposes the uncastled white king: 8...Bxf4 9

Nxf4 Re8+ 10 Ne2 and Black has many ways to get a good game. Even
after 10...Be6 11 Bxe6 Rxe6 12 0-0 Qd6 13 c4 Nd7 he is at least equal.

8...Re8

Black usually chooses 8...Qc7, but I am not sure this is the best square for
the queen. Even if this move is played at all, I don’t think it should be a
priority. Instead I want to stick to our main set-up.

9 Bf4

Or:
a) 9 Re1 (Schalkwijk-Kravchenko, corr. 2011) 9...Nd7 10 c3 Nf8 and
we have reached a typical situation where Black has nothing to fear.

b) After 9 c3 (E.Izi-Sariisik, Turkish Team Ch, Konya 2019) Black has an

instructive move in 9...Na6, heading for c7 and d5. With the bishop on c4,
the d5-square is ideal for Black’s knight. For instance, 10 Bf4 (10 Bxa6
bxa6 gives Black strong play on the light squares: 11 Be3 Qc7 12 h3 Be6
with a complex but equal position) 10...Nc7 11 Qd2 Be6 12 Bd3 Nd5 =.

9...Bg4 10 Bxd6 Qxd6 (D)

White to play

This position is fine for Black. White has the better pawn-structure, but
Black has the more active pieces. In return for his doubled f-pawns, Black
has an extra (half-)open file for his rooks. It will take White a long time to
make use of this queenside majority. Meanwhile, Black will bring his
queen’s rook to d8, putting pressure on White’s centre. White’s main
problem is that it’s difficult to organize an effective queenside
breakthrough.

Black’s knight is normally transferred to f8, where it does not stand in the
way and serves as an excellent defender. From there it can proceed to
e6 to
attack the d-pawn or to g6, where it can easily create some kingside
action. Now:

a) 11 Qd2 Nd7 is roughly equal and a draw was agreed in

D.Rousseau- L.Zsiros, Kecskemet 2010.

b) After 11 c3 Nd7 12 Qc2 Nb6 13 Bd3 (L.Arnold-C.Bastos, corr. 1999) I

like 13...g6. The idea is to neutralize the white line-up on the b1-h7
diagonal. A possible line is 14 f3 Be6 15 c4 Nd7 =.

2C: 11th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 Bc4 Bd6)

7 Qe2+ Be7 8 Nf3 0-0 9 0-0 Re8 10 Re1 Nd7 (D)

White to play

Obviously White’s play in Model Game 2 can be improved, but

the alternatives are nothing to lose sleep over.

11 Bd3
This looks more active than the modest 11 Bd2 of Model Game 2.
Another option is 11 c3 Nf8 12 Bf4 Be6 13 Bg3 Qd7 14 Rad1
(Kovalevskaya- L.Zaitseva, Svetlogorsk 1997), when after 14...Rad8 it’s
hard to come up with a sensible plan for White.

11...Nf8

Once again, we have successfully managed to manoeuvre the knight to

f8. This can be considered a small success.

12 h3 Bd7 13 Qf1 Be6 14 Bd2 Bd6 15 Nh4 Bc7 16 Re4

In Perenyi-Utasi, Budapest 1985, 16...g5? 17 Nf5! gave White the

advantage thanks to the short tactical sequence 17...Bxf5 18 Rxe8 Qxe8
19 Bxf5.
Instead 16...Ng6! leaves Black better. After 17 Nxg6 hxg6! the ‘square’
of pawns may be funny-looking, but they serve a defensive role by
covering many important squares, and can also advance and be used in
an attack.

Lesson 3: Introduction to the Main Line

By: TRH

Every time I win a tournament, I have to think that there is something

wrong with modern chess.

Viktor Korchnoi

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 (D)

For years, this has been considered the best continuation against the
Korchnoi Variation. The main idea is to attack Black’s h-pawn quickly with
Bd3 and Qc2. (Obviously the immediate 6 Bd3?! – see 1C – drops the d4-
pawn.) It leads to unbalanced positions where Black has his share of the
chances, so ambitious players should welcome it.
 Black to play

In the Korchnoi Variation, Black normally hurries to castle and bring his
knight to f8. White’s set-up aims to disrupt this plan. The following
moves are usually played automatically:

6...Bd6 7 Bd3 0-0 8 Qc2

8 Ne2 usually doesn’t make much difference as 8...Re8 9 Qc2 transposes.

For a few independent ideas, see Model Game 3.

The immediate attack on the h-pawn means that Black isn’t in time to
defend it by placing his knight on f8. This way, White manages to force
Black to weaken his kingside. But before committing to anything, Black
should give a check.

8...Re8+ (D)
 White to play
9 Ne2

This is the most frequently played move, but 9 Be3 caught my attention
in 2018 when Svidler played it against Carlsen in the European Clubs
Cup.
Carlsen chose the modest 9...h6, although I believe his original intention
was to go 9...h5. However, this gives White the option of placing his knight
on f3, which is presumably the point of not committing it to e2. In my
opinion the strongest move is indeed 9...h5, but caution is needed by
Black, and this will be discussed in detail in 5A.

Korchnoi’s Lean Mean Attacking Machine

Despite its modest appearance, the Korchnoi Variation is a great

attacking weapon. In most variations, Black is aiming for a set-up
looking like this:
The knight on f8 makes Black’s kingside bullet-proof, but it can also join an
attack via g6, or exert pressure on the d-pawn from e6. The black rooks
typically want to occupy the two open files, while the dark-squared bishop
is usually best on the b8-h2 diagonal. Sometimes you will see the light-
squared bishop go to f5 or g4 instead of e6, while the queen also has
some other options, such as c7, a5 or even b6. Of course, you have to pay
attention to what White is doing, but quite often this is the desired set-up.

In Model Game 1 we saw Black winning with a devastating ...Bxh3

sacrifice. You’ll be surprised to see how many people end up in positions
like my opponent did there.

In the main line White often decides to castle queenside. Fortunately

our attacking machine can just as easily be directed at that wing:
 Black to play

B. Vignesh – T.R. Hansen

London 2017

Notice that here too, from exactly the same squares, the black bishops
are aiming at the white king.

14...b5!

My idea is to advance with ...a5-a4 and slowly create an attack on the king.

15 h3 f5

This move blunts White’s light-squared bishop and the entire line-up on
b1- h7 diagonal.

16 Nxd6 Qxd6

You’ll see again and again that Black welcomes this exchange (i.e. the
white knight for his dark-squared bishop) when White castles queenside.
17 Bc1 a5

Black is planning ...a4 and ...Qd5 with a primitive yet dangerous attack.

In this game, you saw that White’s entire light-squared complex was weak,
which is what you should aim to exploit when White castles queenside. In
the following position, the light squares led to a disaster for White.

White to play

T. Dahl – T.R. Hansen

Norwegian Team Ch 2017

White has failed to parry Black’s attack.

22 a3 Qa2+ 23 Kc1 Bb3 24 Qg2 Qa1+ 25 Bb1 Bd5

This bishop is stronger than any rook!

26 Qf1 Ne6 27 Be5 Be4 28 Rd3 Nc5 29 dxc5 Rxe5 30 Qe2

The light-squared strategy has worked to perfection, and this is probably
the best bishop I have ever had.

30...Rae8 31 Rhd1 Bf5 0-1

These are just examples where things have gone wrong for White but I
genuinely think that Black has the easier task in the first part of the game,
and that he will benefit from the doubled f-pawns.

Model Game 3

Spassky –

Seirawan USA blitz

1990

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6

This is a hard-fought game between two world-class players. It touches

upon many critical questions and illustrates the complexity of the Korchnoi
Variation. However, we must note that since it’s a blitz game, the quality of
the play is lower than would be usual for these players, so don’t judge
them too harshly for the mistakes at the end.

6 c3 Bd6 7 Bd3 0-0 (D)

 White to play
8 Ne2

This is less common and a bit less flexible than 8 Qc2 Re8+ 9 Ne2, but
will usually transpose.

Q: Which extra options would White get by playing 8 Qc2 first?

Well, there are some relatively rare lines where White instead develops
his knight to f3 or even leaves it on g1 for quite a long time. This is mostly
connected to queenside castling and you will see several examples in
Lesson 5.

8 Qf3 was tried in E.Berg-Tikkanen, Swedish Team Ch 2018/19. The

game went 8...c5 9 dxc5 Bxc5 10 Ne2 Nd7 11 Bf4 Ne5 12 Bxe5 fxe5 13
Rd1 Qe7 14 Ng3 Be6 15 0-0 g6 16 Qe2 f6 17 Bc4 ½-½ but if anyone was
close to being better after the opening, it was Black.

8...Re8 (D)

8...Qc7 is relatively common, but I am not a fan. It’s true that it attacks the
h2-pawn, and thus prevents White from castling kingside. However, the
solution is simple: 9 Qc2! gives Black a slightly inferior version of the main
lines we see in later lessons. On d8 the queen is more flexible in general,
and moreover one of the key ideas to answer White’s queenside castling
is by aggressively placing the queen on d5.

White to play
9 Be3

White is developing sensibly, but it looks a bit strange not to attack the h-
pawn with 9 Qc2, which is normal and critical. However, that move
arguably provokes ...h5 and some players who fear our pawn might try to
avoid all provocations.

9 0-0 is inaccurate for the same reasons, as Black’s kingside remains

intact. Since positions with White castling kingside occur frequently, I will
examine 9...Nd7 in 3B. However, I must mention that I was tempted to
have 9...Qc7 as the main game, since J.Perez-Alekhine, Caracas 1939 is
very instructive: 10 f4? (a strategic mistake; 10 Ng3! is the reason why I
couldn’t go for this line as Black’s position is suspect after 10...Nd7 11 Nf5
Bxh2+ 12 Kh1 Bd6 13
Bh6! g6 14 Ng7, when to me it looks like Black is skating on thin ice)
10...f5 11 Ng3 g6 12 Bd2 c5 13 Rc1 Nd7 14 Re1 Rxe1+ 15 Qxe1 Nf6 16
Qf2 Be6
and Black had a close to decisive advantage, and won a smooth

game. We now return to 9 Be3 (D):

Black to play
9...Nd7 10
Qc2

10 Qd2 is similar, as also in this case White stays flexible about which way
to castle. Black has a pleasant choice:

a) 10...Nf8 11 0-0 Qc7 12 Kh1 Ng6 13 Rae1 Be6 14 Qc2 Qa5 15 c4 Rad8
and in this typical position, Black is not worse, Fedorchuk-
Gharamian, French Team Ch 2019.

b) However, to broaden our arsenal I would like to point out an alternative

continuation. As mentioned earlier, I’m normally careful about committing
the queen to c7, and instead advocate placing the knight on f8. However,
here it makes sense to wait and see White’s intentions with 10...Qc7:
b1) After 11 0-0-0, 11...b5 with the idea of playing 12...Nb6 looks
very strong for Black. White’s queen is misplaced on d2.

b2) Against 11 h3 I suggest a set-up a little different from the others in this
book: 11...b6 12 0-0 c5 followed by 13...Bb7. One of the advantages of
the Korchnoi Variation is that Black has many set-ups and it’s hard for
White to out-prepare him.

10...Nf8 (D)

White to play
11 Ng3

This is not the most common move, but transpositions are likely. Otherwise:

a) 11 0-0 is similar to variations discussed in 3B, but with the queen

already committed to c2.

b) 11 0-0-0 looks aggressive, but in reality Black’s attack is more

dangerous. 11...Be6 12 Kb1 b5 launches a pawn-storm at White’s king,
which provoked the typical mistake 13 Nf4? Bxf4! 14 Bxf4 in S.Robin-
Vardan, San Sebastian
2018, when 14...Qd5 15 b3 a5 (or 15...Qxg2) would have left Black
much better.

11...g6

This can be played almost automatically, as it blunts White’s line-up on the

b1-h7 diagonal. However, Black has an interesting alternative which could
also make sense in similar situations: 11...c5 targets the d4-pawn which
the white knight no longer protects. For instance, after 12 0-0 cxd4 13
Bxd4 Qc7 Black’s set-up is quite powerful. The rooks will occupy the open
files, while the queen has found a nice post where it plays on both sides of
the board.
Black is at least equal.

12 0-0-0 (D)

Black to play

White claims to have provoked a weakness on the kingside with his

11th move, and now he signals an attack with opposite-side castling.

12...f5!
If you are allowed to play this against a knight on g3, chances are good
it’s a strong move.

13 Ne2

The players in this game are incredibly strong, but even they didn’t
manage to make all the right decisions. This is partly due to the time-
limit, but the strategic complexity is also part of the explanation.

13...Qa5

I believe Seirawan underestimated his position. My choice would have

been 13...Be6 14 Kb1 b5, when I believe Black’s attack is much more
effective. For instance, after 15 h4 h5 Black is ready to continue the march
on the queenside, while g4 is out of the question for White, so I have
trouble finding a meaningful move for him.

14 Kb1 Be6 15 b3?! (D)

15 c4 is met by 15...b5, but even so, this was White’s best option.

Black to play
15...Qc7

White’s last move created a target for attack, so Black clears the way for
the a-pawn.

15...b5 also makes a ton of sense, but it’s well worth noting that the
engine screams enthusiastically in favour of including 15...Bd5 with the
idea of grabbing the g2-pawn. One entertaining line is 16 Rdg1 b5 17 h4
h5 18 Bd2?! Qb6 19 f3 a5 20 g4?! hxg4 21 fxg4 c5 and thanks to the
threat of ...c4 the engine claims Black is winning.

16 Qd2 a5 17 h4 a4 18 b4 b6

Black is playing too patiently and gives away chances to fight for
an advantage.

It is better to shut down White’s attack with 18...h5!. If White plays slowly,
Black can continue with ...b5 and play on the light squares, after which
exchanging bishops with ...Bc4 becomes a good idea. If White prepares
g4 with 19 f3, Black has an instructive manoeuvre beginning with 19...Nd7,
heading for b6 and d5. White’s position is very fragile while Black has
many ways to improve his game.

19 Bf4?!

19 h5 was called for. However, it’s not like Black is getting mated as
after 19...Bd5 20 hxg6?! fxg6 he can defend along his second rank. In
positions like this, the knight on f8 makes Black’s kingside extra
resilient.

19...b5

It’s interesting that neither player wants to push their pawn to h5. The
explanation might be that Seirawan wanted to keep some flexibility, and
as a matter of fact, later in the game, he played ...h6 and ...g5.

20 Bxd6 Qxd6 21 Nf4 Bd5 22 Rde1 Ne6

Black is slightly more comfortable after 22...h5; for instance, 23 Rhg1

Nd7 24 f3 Nf6 and neither player can break through any time soon. With
some
preparation, Black can try the manoeuvre ...Be6 and ...Nd5.

23 Nxd5 Qxd5 24 f3 Ng7 (D)

White to play

The position is becoming simplified, and we are moving towards an

endgame where Black is no worse.

25 Rxe8+ Rxe8 26 Re1 Rxe1+ 27 Qxe1 Ne6 28 Qe3 Kg7 29 Kb2

White can’t make progress, but now Black starts to push on the kingside.
If Black wants a draw, standing still should do the trick.

29...h6 30 g3 f6 31 Be2 g5 32 hxg5 hxg5 33 Bd3 f4 34 gxf4 Nxf4

The position is still drawn, but now White can make use of the e4-
square, which is why I liked having the pawns on g6 and h5. Still, with
this being a blitz game, one can’t expect very accurate play.

35 Bc2 Ne6 36 Qe4 Qxe4?

36...Kf8 is still equal.

37 Bxe4 Nd8 38 c4? bxc4 39 Kc3 Kf7 40 Kxc4

Black is under serious pressure.

40...Ke6 41 Kc5 f5??

41...Kd7 42 d5 looks really grim for Black, but the engine doesn’t share
my worries.

42 Bd3??

42 Bxc6! is winning for White. Such a move would normally be a piece

of cake for Spassky, but here I am pretty sure we can blame the clock.

42...g4 43 fxg4 fxg4 ½-½

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 (6...Bd6 7 Bd3 0-

0 8 Ne2 Re8 9 0-0)

3A: 6...Bf5!?

3B: 9 0-0

In this magnifier collection you’ll find the first one offering an alternative for
Black. This may be considered a luxury item so 3A could well wait until you
actually need to surprise a forewarned opponent. Remember that for the
first year or so you play the Korchnoi Variation, few opponents will expect
you to play it and therefore will not be very well prepared.

Preparing the lines in 3B should not top your list of priorities. A

good understanding of the basic ideas should take you a long way.

3A: 6...Bf5!?
(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3)

If you want to try something a bit different, I suggest you give the following
move a chance:

6...Bf5!? (D)

White to play

Occasionally I play this surprise weapon to prevent White from setting

up with Bd3 and Qc2. The more I looked at this variation when writing
this book, the more I liked it. It definitely needs further testing, but I
believe it may be a valid alternative to the main lines.

7 Ne2!

This is probably the real test. Other moves:

a) 7 Bd3 is modest but logical as it aims to swap pieces in the hope of

reaching an endgame. Black has many decent options, like 7...Bxd3 or
7...Qd7, but I suggest a more combative approach: 7...Bg6 8 Ne2 (after 8
Bxg6 hxg6 Black’s ‘square’ structure might look a bit strange, but I
believe
this is very promising for Black) 8...Bd6 9 0-0 0-0 and Black has nothing
to worry about as f4 will always be met by ...Bxd3, when White will have
weakened squares on the e-file for no good reason.

b) 7 Bc4 Bd6 8 Ne2 0-0 9 0-0 Nd7 10 Ng3 Bg6 11 f4 at first sight looks
dangerous for Black. However, Black has excellent prospects after
11...Nb6 12 Bb3 f5!. This is why White is better off delaying the
development of the light-squared bishop. It looks absurd to bury the bishop
like this, but in return White has a bad knight on g3 and Black can look
forward to exploiting the weakened e4-square, which is exactly what
happened in Alsina Leal-Salgado Lopez, Linares rapid 2015.

c) 7 Nf3 Bd6 (D) and now:

White to play

c1) After 8 Bd3, Black can equalize by simply taking on d3 followed by

...0- 0 and ...Re8. However, we can also aim for more complex play:
8...Bg6 9 Qe2+ and now instead of the peaceful 9...Qe7, as played in
Kokarev- V.Moiseenko, Khanty-Mansiisk 2017, I suggest the more
combative 9...Be7 10 0-0 0-0. This position has never occurred, and no
theory exists. I believe
it’s roughly equal.

c2) 8 Be2 Nd7 9 0-0 0-0 10 Be3 Qc7 = Hou Yifan-Ruan Lufei,
Women’s World Ch, Antakya 2010.

c3) 8 Nh4 Be6 9 Bd3 (Lecha Gonzalez-Cardoso Garcia, corr. 2008) 9...0-0
10 0-0 g6 and in this typical position Black has no problems.

c4) 8 Bc4 0-0 9 0-0 and in Percin-Gok, Turkish Team Ch, Konya 2019
Black played 9...Nd7. Instead, 9...Re8 is more flexible, as Black can
consider developing the knight to a6; e.g., 10 Nh4 Be6 11 Bxe6 Rxe6 =.

7...Bd6 8 Ng3

White tries to prove that the black bishop has reached f5 too early.

8...Bg6 9 Be2 0-0 10 0-0 (D)

Black to play
10...Re8!
I believe this move is a tough nut for White to crack. Originally my intention
was to recommend 10...Nd7 11 f4 f5 12 Bd3 Qa5 13 Qc2 c5, as in Eminov-
R.Ibrahimov, Azerbaijan Under-14 Ch, Baku 2003. I found this position
very interesting, but concluded that Black doesn’t get full compensation.

11 f4

This seems the only logical move. White must try to prove that the bishop
is misplaced.

11...f5

You have to be a bit crazy to play with such a bishop, but everything is
much better than it looks. The knight on g3 is misplaced and there are
weak squares on e3 and e4. Additionally, White’s dark-squared bishop can
only go to d2.

12 Bd3

White can win the f-pawn, but with dynamic play, Black can prove a lot of
compensation in all lines. If you check this position with an engine, you
will see that Black has many decent moves.

12...Qf6 (D)
 White to play

It took me a while to settle on this move, which was played in

Mehlhorn- J.Berg, corr. 2017, as some of the alternatives are just as
interesting.

13 Qc2

What is more logical than going for the f5-pawn? After 13 Qb3 Na6!
two sample lines are:

a) 14 a3 (to overprotect b4) 14...h5!? (this is what I love about this

variation: Black plays on the entire board) 15 Qxb7 Nc5 16 dxc5 Bxc5+ 17
Kh1 Rad8 and despite being a knight and a pawn down, Black is in the
driving seat.

b) After 14 Qxb7 Nb4 Black threatens a perpetual attack on the white

queen, and 15 cxb4? (15 Bb1 Reb8 16 Qd7 Rd8 with a draw) 15...Qxd4+
16 Rf2 Qxd3 is much better for Black.

13...c5 14 d5 (D)

Or:
a) 14 Bxf5 cxd4 is good for Black.

b) After 14 Nxf5 Bf8 15 dxc5 Bxc5+ 16 Kh1 Nc6 White has to be careful
because of his undeveloped queenside. Black has enough
compensation.

Black to play
14...Na6

There’s a reason it’s recommended to connect the rooks. 14...h5? is the

move we would like to play, but there is a problem with the unprotected
rook on e8: 15 Nxf5 c4 16 Nxd6 Bxd3?? 17 Nxe8 and White wins.

15 Bd2 h5

There are many possibilities here, but I will limit myself to one sample
line. Black’s bishop is about to reach h5 and Black has many interesting
ideas.
Even the obscure 15...c4 is given as close to equal by the engines, which
says something about the potential in this position.

16 Bxf5 c4 17 Bxg6 Qxg6 18 Qxg6 fxg6 19 Rae1 Nc5

Black has enough compensation for the pawn.

3B: 9 0-0

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3)

6...Bd6 7 Bd3 0-0 8 Ne2 Re8 9 0-0 (D)

Black to play

By simply castling, White is not aiming to refute Black’s set-up, but

instead going for more modest play. In most lines White is looking to
exchange bishops with Bf4, and we shall see how to deal with this.

9...Nd7

We can go for our usual set-up, but if you want to experiment a little, there
is an interesting alternative in 9...Na6:

a) The key point is that 10 Bxa6 bxa6 is not dangerous as Black’s

bishops and open files for the rooks compensate for the damaged
structure.
b) 10 Ng3 g6 11 Qf3 f5!? (11...Nc7 = is safer) 12 Bf4 (12 Bxa6 bxa6 13
Qxc6 Rb8 gives Black good play for the pawn; e.g., 14 Bg5 Qd7!)
12...Nc7 13 Rfe1 Be6 = Rabiega-Svane, Berlin blitz 2018.

c) 10 Bf4 Nc7 11 Bxd6 Qxd6 12 Ng3 g6 13 Ne4 Qd8 14 Re1 f5 15 Nd2 Be6
= Janik-Svane, European Ch, Batumi

2018. We now return to 9...Nd7 (D):

White to play
10 Bf4

White’s strategy of exchanging pieces is a common theme in the Korchnoi

Variation, but it is not a real attempt for an opening advantage. Other
moves:

a) 10 a4 is a somewhat surprising move, but makes sense as White has a

queenside majority. 10...Nf8 11 Ng3 and instead of 11...Be6, as in
Sutovsky- Asrian, European Ch, Istanbul 2003, I like the standard reaction
to a knight on g3: 11...g6 12 Ne4 Bc7 and Black will continue with ...Ne6
to cover g5, and then push the white knight away with ...f5. Black is doing
fine.
b) 10 Ng3 should be met by 10...g6:

b1) After 11 Qf3, as in Vardan-Sedlak, Bački Petrovac 2018, Black can

continue to restrict the knight on g3 with 11...f5!?. An important point is that
12 Nxf5? is bad because of 12...gxf5 13 Qxf5 Nf8. The knight on f8 is often
a key defender in the Korchnoi Variation.

b2) 11 Ne4 has been used in two games, both won by White, who was
the higher-rated player. However, Black is OK after 11...Be7:

b21) 12 f4?! is a positional mistake. After 12...f5, as in Djokić-Djorić, Veliko

Gradište 2016, Black is better due to White’s weakened e3- and e4-
squares.

b22) After 12 Bf4 Nf8 13 Qc2 Ne6 14 Bd2?! f5 15 Ng3 (as in T.Smirnov-
Shabunova, Moscow Team Ch 2016), White’s knight is poorly placed, and
Black takes the advantage with 15...f4! 16 Ne4 (16 Nh1? c5!) 16...f5! 17
Nc5 Bxc5 18 dxc5 Nxc5 –/+.

10...Nf8 (D)

White to play
11 Qd2

White can also take on d6 immediately. 11 Bxd6 Qxd6 12 Ng3 (12 Qd2
Ng6 transposes to the main line) 12...g6 13 Ne4 Qd8, and now:

a) After 14 Qf3 f5 15 Nd2 Be6 16 Bc4 Qg5 17 Rad1, Black played the
aggressive 17...h5 in Alsina Leal-Spraggett, Catalonian Team Ch 2014. I
would have preferred to complete development with 17...Rad8. Either
way, the position is equal.

b) 14 Re1 Be6 15 Qf3 f5 16 Nc5 Bd5 17 Rxe8 Qxe8 = Lubbe-

Feuerstack, Lüneburg 2019.

11...Ng6 12 Bxd6

The Londonesque 12 Bg3 has never been tested. A possible

continuation is 12...Be6 13 Rfe1 Qd7, when I see no problems for Black.

12...Qxd6 (D)

White to play
It can be argued that White has achieved his opening goal.

13 Rfe1

After 13 Ng3, 13...Qf4, as in Tasdogen-Annaberdiyev, Turkish Team Ch,

Konya 2019, is by no means necessary. I would have considered 13...Nf4
14 Rfe1 Be6 with a more complex position.

13...Be6

Black’s position is very safe, and 13...Bd7, as in V.Onishchuk-Agrest,

Riga blitz 2019, should also equalize.

14 Ng3

This position has occurred several times, often with strong

grandmasters playing Black. Other moves:

a) 14 Bc2 can be met by 14...Rad8. Having finished development, Black

is ready for ...c5, with an equal game.

b) After 14 b3 Rad8, the position is equal, but the following line shows
how things can easily go wrong for White: 15 c4 Nh4! 16 c5?! (this is bad
both positionally and tactically, but 16 Qc3 Bg4 looks fine for Black)
16...Qd5 17 Nf4?? Qxg2+! 18 Nxg2 Nf3+ –+.

We now return to 14 Ng3 (D):

 Black to play

14...b6

Or:

a) 14...Rad8, developing the last piece and waiting to react to White’s

move, could be worth trying.

b) 14...Nf4 and now:

b1) 15 Ne4 led to a relatively undramatic draw in Movsesian-Riazantsev,

German Cup, Wolfhagen 2019 after 15...Qc7 16 Nc5 Nxd3 17 Nxe6 fxe6
18 Qxd3 e5 19 dxe5 Rxe5 20 Rxe5 Qxe5 21 Qc4+ Kh8 22 Qf7 Re8 23 h3
Qe7 24 Qxe7 Rxe7 25 Rd1 Kg8 26 Kf1 ½-½.

b2) More testing is 15 Bc2, but here Black has an interesting tactical
resource. White intends to chase the knight away with Re4, which is
why 15...f5! is a strong move. If White doesn’t take the pawn, Black will
block out White’s minor pieces with ...g6 with an advantage to Black.
Alternatively, Black equalizes after 16 Nxf5 Bxf5 17 Bxf5 Nxg2!. Then 18
Bxh7+ Kxh7 19 Kxg2 is best but White’s damaged structure gives Black
sufficient compensation for the pawn.

15 Qc2 (D)

Or 15 Nf5 Qd7 16 Qc2 Rad8 = Cridland-R.Harris, corr. 2011.

Black to play
Black still has some leeway
here:

a) I am pretty sure my choice in a game would have been 15...Rad8 =.

b) 15...c5 16 Ne4 Qc7 17 dxc5 bxc5 18 Ng3 Rab8 19 c4 a5 =

Inarkiev- Tomashevsky, Sochi blitz 2014.

c) 15...Nf8 is a rather surprising retreat, which worked out perfectly in

Arjun- Maghsoodloo, Abu Dhabi 2017 as Black grabbed the initiative after
16 Bf5 g6 17 Bxe6 Nxe6 18 Ne4 Qe7 19 Re3 f5. The position is still
roughly equal, but from here on it gradually went downhill for White.

Lesson 4: Introduction to the Hansen Variation

By: TRH

Have you ever seen a chess article without a brilliant example of the
author’s own play? ‘Silly question,’ you will say. Quite.

Yuri Razuvaev

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-

0 8 Ne2 Re8 9 Qc2 h5 (D)

White to play

This move was first played by Andrei Chigvintsev in 2014, but the variation
didn’t come to real attention until Irina Krush played it in 2017. Neither of
these players has continued to play it on a regular basis. When I saw the
Krush game, I immediately started analysing and discovered that it’s not
only a good move, but it also leads to fascinating play. I have used it in
approximately 20 over-the-board tournament games, with only two losses.

Now White’s main decision is which way to castle. Here we shall focus on
10 0-0. See Lesson 5 for 10 Be3 followed by 11 0-0-0, and some other
lines
where White castles queenside.

Model Game 4

Kwiatkowski – T.R. Hansen

British League (4NCL) online 2020

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6

When playing this game, I realized it probably was my last game before
publishing the book. Despite being a game in the Four Nations Chess
League, it was played online due to the COVID-19 situation. I still felt a bit
nervous about the risk of ruining my excellent statistics.

6 c3 Bd6 7 Bd3 0-0 8 Ne2

We have seen the drawbacks of a knight development to f3, and

traditionally any combination of Qc2 and Ne2 used to lead to the clear
main line.
However, in Lesson 5 we shall investigate some possibilities that are
only available to White from the 8 Qc2 move-order.

8...Re8 9 Qc2 (D)

 Black to play

I had this position a few times in my youth. I tried each of the then-
standard ways to save the h7-pawn, but somehow the move 9...h5 never
occurred to me. To understand the motivation for it, you should know this
about the alternatives:

9...g6 used to be a main line but 10 h4! gives White a dangerous

attack. 9...h6 is solid, but a bit passive.
9...Kh8 is a decent choice which prepares ...Nd7-f8 without making
any committal kingside pawn moves. Unfortunately, if nothing else, it
allows White a couple of forced draws.
9...f5 is a relatively new and surprising pawn sacrifice. Although
it works reasonably well, I’m not entirely happy with the resulting
positions.

9...h5

Even if you have decided to take up this variation with the black pieces, it’s
a good idea to have some thoughts on how your opponents will react. After
all, Black’s last move looks a bit ugly, at least from a traditional point of
view. A
natural reaction is to try to punish our last move, but no simple path to
an advantage is to be found.

10 0-0 (D)

Black to play

Although the plan with queenside castling is more aggressive, be aware

that this move is the engines’ first choice.

10...h4

To me, this is clearly the most logical continuation. I want to use the h-
pawn to soften up White’s kingside. However, there is a decent alternative
in 10...f5, exploiting the fact that the knight on e2 would be left unprotected
if White captures on f5. After 11 Re1 g6 12 Bf4, 12...Na6 was played in
Suarez Real-J.Houska, British League (4NCL) 2017/18. Black got a good
position from the opening, but I’m not convinced, as White might end up in
a middlegame with a stronger minor piece: 13 Bxa6 bxa6 14 Bxd6 Qxd6
15 Qd2 looks better for White. If you want to try this line, I believe 12...Be6
is easier and fine for Black.
Some players, including Vachier-Lagrave, have tested 10...Nd7. Most of
these games were played in late 2019 or 2020, and I naturally took an
interest in this. However, the line 11 Ng3 Qc7? 12 Nxh5! Bxh2+ 13 Kh1
Bd6 (Firat- Arutinian, European Ch, Skopje 2019) 14 Bh6! is really scary,
and serves as an argument for Black to prefer 10...h4.

We now return to 10...h4 (D):

White to play
11 h3

White can possibly make do without this move, but I don’t see why he
should give Black the option of playing ...h3. To illustrate my point, 11 Re1
h3 (Black can also play safe with 11...Nd7, leaving ...h3 hanging over
White) 12 g3?! Bxg3 13 fxg3?? Qd5 is winning for Black. This was far from
White’s best play, but with a black pawn on h3, White is always tactically
exposed.

11 Bf4 is a tricky continuation that a few well-prepared players have tried.

Then the naïve 11...Rxe2?? allows White to show his tactical point: 12
Qxe2 Bxf4 13 Qe4 left White with a winning position in Jakić-Tica, Zadar
2018.
11...h3 is interesting but may be premature as after 12 g3 Na6 13 Bxd6 Qxd6
14 Nf4 White threatens to pick up the h-pawn with Bf5. In Arjun-
Bernadsky, Budapest 2019, White met 14...g5 with 15 Bc4! f5 16 Nxh3
Qh6 17 Rfe1, when White’s extra pawn may be enough to give him a slight
edge. Instead 11...Na6! looks right, when there isn’t anything special going
on. For a more detailed coverage of this position and a few 11th-move
alternatives, see 4A.

11...Nd7 (D)

White to play

Once again, I prefer sending the knight to f8 as quickly as possible.

The positions resulting from this variation tend to be of a strategic nature

with few forced variations, and I believe this is one of the advantages of the
opening. It is hard for the opponent to prepare, and even if they do, you are
still on steady ground. By playing it frequently you will become familiar with
the strategic questions. This game is far from perfect, and I failed to find all
the right moves. However, after analysing, I found the solutions to the
problems very instructive.

12 Bf4!?
This move contains a trap related to the one pointed out in the notes to 11
h3. At first sight it seems that Black can win material, but taking on e2
would be a huge mistake. For 12 Be3 and a few rare moves, see 4B, while
the main move, 12 Bd2, is covered in 4C and 4D.

12...Nf8

With the h7-square now protected, Black is threatening to take on e2.

The immediate 12...Rxe2?? 13 Qxe2 Bxf4 14 Qe4 is winning for White;
e.g., 14...Bh6 15 Qh7+ Kf8 16 Rfe1 with a crushing attack.

13 Bxd6

Practically the only option. Moving a piece for the second time with 13
Be3 would be enough for Black to claim equality. Much the same can be
said about 13 Qd2. In Mihok-Bogner, Mitropa Cup, Radenci 2019, Black
gradually equalized after 13...Ne6 (exploiting White’s last move with
13...Ng6 14 Bxd6 Qxd6 followed by ...Be6 is probably even simpler) 14
Bxd6 Qxd6 15 Rfe1 Bd7 16 Bc4 (16 Rad1 transposes to the main game)
16...Rad8 17 Rad1 Bc8 =.

13...Qxd6 (D)
 White to play
14 Rfe1

This is a major decision by White. By putting this rook on e1, he leaves

d1 free for the other one. In my eyes, this means that White is aiming for
breaking through with d5. By choosing the other rook, White would
signal kingside play with f4.

Against 14 Rae1 I would also have chosen 14...Bd7, when it’s not easy for
White to make progress. For instance, 15 f4? is a mistake due to 15...Re3
followed by ...Rae8. In general, White should be careful about playing f4,
as it weakens important squares on the e-file.

14...Bd7

I believe this is the strongest. I want to keep the e6-square vacant for
the knight.

15 Rad1 Ne6

Advancing the pawn to h4 provoked White to create a target on h3. I am

toying with the idea of playing ...Ng5, hoping to sacrifice successfully on h3.
However, when analysing this game, I started to wonder if I should have
been more patient.

The engine says 15...Re7 is completely equal. However, I believe the

downside to this is more of a practical nature. After 16 Nc1 White would
exchange at least one pair of rooks. I want to play for a win, and I didn’t
see the need to provoke more exchanges.

15...b6!? is flexible and later in the game you will see why this is a
useful move for numerous reasons. It’s not obvious how White should
make progress, and Black can react to White’s choice.

16 Qd2 Rad8 (D)

16...g6? is refuted by 17 Qh6!.

White to play

In this variation in general, I like to complete development before thinking

about an attack. Having the rook on d8 makes d5 harder to achieve, and at
the same time, sometimes Black can find tactical resources on the d-file.
17 Bc2

White’s idea is to play Qd3.

17 Bc4 transposes to the note to White’s 13th move above.

After 17 c4, 17...g6 is the move I would like to play. This intends ...Kg7,
when Black will have neutralized most of White attacking potential on the
kingside. With that out of the way, I can start to build up my own kingside
attack, with ...Ng5 as a follow-up; e.g., 18 Qh6 Nxd4 19 Bxg6 Nxe2+ 20
Kh1 Ng3+ 21 fxg3 Rxe1+ 22 Rxe1 fxg6 23 Qxg6+ with a perpetual check.
The alternative is 17...Ng5 to keep the tension, and avoid forcing lines; I
believe Black can be reasonably happy with his chances.

17...Bc8

A moment of hesitation, and played with a bad feeling. Although it is a

little passive, I didn’t want the bishop to block the d8-rook. In hindsight, I
should have preferred 17...b6. As you will see in numerous lines, guarding
the c5- square is useful. This move also gives Black the possibility of
playing ...c5 and ...Bc6. Additionally, it keeps the option of playing ...Bc8-
a6 to fight for the f4-square.

18 Qd3

White has conducted a strong regrouping, questioning why Black hurried

with ...Ne6. After all, from f8, the knight safely covered h7. However,
there is a downside to the move as well: the queen can no longer access
the h6- square after a future ...g6.

18...Ng5 (D)

18...g6? is strongly met by 19 Qe4, picking up the h4-pawn.

 White to play
19
Nc1?!

After 19 f4 I was intending 19...Ne4. I must admit I believed Black

was slightly better. However, White has a strong resource in 20 Ng3!.
After 20...hxg3 21 Rxe4 Rxe4 22 Qxe4 Qe6 it is White who has a
small plus, owing to Black’s weak pawn on g3.

19...g6

At this point, I considered my position slightly more comfortable. An obvious

possibility for Black is ...Bf5, when I can slowly improve my kingside with
...Kg7. White has to be careful about allowing a sacrifice on h3, and ...c5
is also a potential idea.

20 Qd2?

This is a provocative move. I suspected the sacrifice on h3 was strong,

but I couldn’t find anything decisive.

20...Kg7?!
However, the engine shows Black is much better after 20...Bxh3! 21 f4
Bxg2!; e.g., 22 Qxg2 (22 Kxg2? Qd5+ –+; 22 fxg5? Bf3 –+) 22...Qxf4
23
Rf1 Qd6, which I find hard to evaluate, but the engine shows no
mercy: Black is well on top.

21 f4

This move weakens some important squares. However, as you will see,
the way to exploit it is not trivial.

21...Ne6 22 Nd3 (D)

22 f5 Nf4 23 fxg6 fxg6 is slightly better for Black due to his strong knight on
f4.

Black to play

White’s idea is to go after the h-pawn. This was a critical moment of

the game, and I learned an important lesson when analysing the
position.

22...b6?!
Best play is 22...Nc7 23 Qf2 Rxe1+ 24 Rxe1 Rh8, with some
advantage. Black’s moves may seem weird, but the reality is that
White’s initiative is neutralized. Black is ready to play ...Bf5 and ...b6
and slowly improve his position, while White’s weaknesses will not go
away.

23 Qf2

23 f5 is well met by 23...Ng5 24 fxg6 Qg3 with a strong attack.

23...Ba6

This move didn’t go as planned. This is another instructive moment. Also

here 23...Nc7 is a strong move, even if it sacrifices the h4-pawn. After 24
Qxh4 Rxe1+ 25 Rxe1 Bf5 with ideas like ...Ne6, ...Nd5 and ...c5 Black
surely has a lot of compensation. I believe this resource, and Black’s
possibility on the previous move, prove why White should be careful about
playing f4 in this variation.

24 f5 Bxd3 25 Bxd3 Nf4 26 fxg6 fxg6 27 Qxh4 Rxe1+?!

Black’s last chance to prove sufficient compensation was 27...Rh8! 28

Qg4 Rh5! 29 Qf3 Rg5, with a lot of kingside pressure.

28 Qxe1 (D)
 Black to play

When playing 23...Ba6, I had overlooked that White could recapture on

e1 with the queen. White is a pawn up, and I have insufficient
compensation.

28...Rd7 29 Bf1 g5 30 Qf2 c5

Seeing things heading the wrong way, I decided to complicate the

position and was immediately rewarded.

31 Re1?!

This lets the advantage slip away. 31 g3 +/–.

31...cxd4 32 Qxd4?!

One inaccuracy follows the other. This one allows a tactic and lets Black
take over. I expected 32 cxd4 Qxd4 33 Re7+ Kg6 34 Qxd4 Rxd4 35 Rxa7
Rd2 and with such active pieces Black should have no problems drawing.

32...Nxh3+ 33 Kh1
33 gxh3?? loses to 33...Qg3+.

33...Qc6

33...Qxd4 34 cxd4 Nf4 is a safe continuation, leaving Black slightly

better due to White’s isolated pawn and Black’s more centralized king.

34 Qb4 Nf4 35 Re7+ Kg6 36 Rxd7 Qxd7 37 Qe4+ f5 38 Qa8 Qd1?!

This gives White a chance to draw right away. After 38...g4 39 Qg8+ Kf6
40 Qh8+ Kg5 White must defend grimly.

39 Kg1?!

39 Qe8+ leads to perpetual check. It’s important not to allow the black
king to h5 and g4.

39...Qe1

Now the game can go on.

40 Qc6+ Qe6 41 Qxe6+ Nxe6 (D)

 White to play

Black’s kingside majority and more active king give him the more
pleasant endgame. However, with correct play, White should be able to
hold.

42 g3?!

42 Kf2 was called for. After 42...Nc5, 43 Bb5 stops ...Na4, and, compared
to the game, the check is harmless.

42...Kf6 43 Kf2 Nc5

The threat of ...Na4 is now hard to meet.

44 c4?

After 44 Bb5 a6 45 Bc6 Nd3+ the b2-pawn is falling, though the

computer tells us that 46 Ke3 Nxb2 47 Kd4 should hold with precise
defence.

44...Ne4+

44...Ke5 also wins.

45 Kg2 Ke5 46 Kf3 Kd4

The pawn endgame is winning too: 46...Nd2+ 47 Ke2 Nxf1 48 Kxf1 Kd4
49 b3 Kc3 50 Ke2 a5 followed by ...Kb2 is decisive.

47 b3 a5 48 Be2 Nc3 49 a3 Nxe2 50 Kxe2 Kc3 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 0-0 h4 (11 h3 Nd7 12 Bd2 Nf8)

4A: (11 Bf4)

4B: (12 Be3)

4C: (13 Rfe1)

4D: 13 Rae1

We are nearing the core of a Korchnoi repertoire and if you are an

ambitious player, you should spend an evening on each of these
magnifiers. If you are a more casual player or trust that your opponents are
not well prepared, you should still play through the main moves and have
an extra look at any sideline that catches your attention. Fortunately, these
are entertaining lines where Black does the attacking, so you should have
a good time doing whatever preparation you find appropriate!

4A: 11 Bf4 and Rare 11th Moves

(1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 0-0 h4)

11 Bf4

White is aiming to swap pieces to get one step closer to the endgame. A
new idea is 11 Nf4 (D), when Black must be careful:
 Black to play

a) After 11...Bc7?!, the surprising 12 Qd1!? is an interesting response:

a1) After 12...Qd6? 13 Qh5 White is already winning. Kanarek-Krstulović,

Warsaw 2019 continued 13...g5 14 g3 Kf8 15 Bc4 Be6 16 Bxe6 fxe6 17
Qh7 Re7 18 Ng6+ Ke8 19 Qh8+ Kd7 20 Qxf6 1-0. A disaster for Black,
and I suspect more players would like to utilize this idea.

a2) 12...Nd7 is essential, as 13 Qh5 Nf8 is OK for Black, though he will

need to play ...h3 as a pawn sacrifice.

b) Luckily, the best way to fight White’s idea is our standard knight
manoeuvre: 11...Nd7 12 Qd1 Nf8. One of the reasons why a knight on f8 is
such a brilliant defender is that it’s almost impossible for White to get rid of
it. The position after 13 Qh5 Qc7 14 g3 has been reached twice in recent
blitz games. In both Sychev-Le Quang, Internet blitz 2020 and Khanin-
J.van Foreest, Internet blitz 2020, Black chose 14...Bd7. However,
14...Qd7 is preferable. Black’s plan is to play on the light squares. One
idea is ...Qg4, but Black may also play more ambitiously. One possible line
is 15 Bd2 c5 16 Rad1 (16 dxc5?? Bxf4 costs White a piece as the d3-
bishop is unprotected)
16...Qg4 17 Qxg4 Bxg4 18 f3 Bd7, with equal chances.

11...Na6! (D)

White to play
12 h3

Stopping ...h3 after all. White has also tried the following alternatives:

a) 12 Rae1 Nc7 13 Qd2 Be6 14 b3 was played in Arjun-R.Buhmann,

Karlsruhe 2019, and this is the right time for Black to play 14...h3! 15 g3
Bf8 as he is threatening 16...Qd5. After 16 c4 Na6, threatening 17...Bb4,
it’s clear that Black has the initiative.

b) A couple of months later, Buhmann reached the same position again,

this time in a blitz game: B.Krause-R.Buhmann, Magdeburg blitz 2019
continued 12 Bxd6 Qxd6 13 Rae1 Nc7 14 Qd2 Be6 15 a3 a5 (Black should
bring the last piece into the game with 15...Rad8 16 Qf4 h3 17 Qxd6 Rxd6;
this way Black manages to damage White’s structure as 18 g3? c5 is
obviously good for Black) 16 Qf4 Qxf4 17 Nxf4 Kf8 18 f3 b5 19 Nxe6+
Rxe6 20 Bf5 Rxe1 21 Rxe1 Nd5 22 Be4, with an equal ending as White
can’t make use of his
queenside pawn-majority.

c) 12 Rfe1 Nc7 13 Qd2 Ne6 14 Bxd6 Qxd6 15 Bc2 Bd7 16 f4?! (quite
often, as here, this move weakens too many squares on the e-file)
16...Nc5 17 Nc1 h3 18 Nb3 and in Bacrot-Villegas, French Team Ch 2019
Black accepted a draw in a slightly better position.

12...Nc7 13 Bxd6 Qxd6 14 Rae1

Bringing the other rook to e1 isn’t dangerous either: 14 Rfe1 Qd5 (14...Ne6
is also roughly equal, Handke-Grandelius, European Clubs Cup, Porto
Carras 2018) 15 c4 Qg5 and without dark-squared bishops, the black
queen has found a dominating position; e.g., 16 Kh1 Bd7 17 Rad1 Rad8
and Black has no reason to complain.

14...Bd7

The queen manoeuvre seen in the previous line doesn’t work with a rook
on f1: 14...Qd5?! 15 c4 Qg5 and whenever White feels the need, the
queen can be pushed away with f4.

15 Qd2 Rad8 16 Qf4 (D)

 Black to play

White plays for the pawn-majority in the endgame, but it’s really nothing to
fear. After 16...Qxf4 17 Nxf4 Ne6 18 Ne2 g6 19 Rd1 the higher-rated
player won in Bacrot-Blohberger, Bundesliga 2018/19, but this position
shouldn’t worry Black. I would have started with 19...f5 to blunt White’s
light-squared bishop and activate my king. Then after 20 Rfe1 Kg7 21 Kf1
Kf6, if anyone is better, it’s Black. He is ready to advance further on the
kingside and is more active while White is a long way from exploiting his
queenside majority.

4B: 12 Be3 and Rare 12th Moves

(1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 0-0 h4)

11 h3 Nd7 (D)
 White to play
12 Be3

This looks natural and is almost as popular as 12 Bd2, but it should be

noted that placing the bishop on e3 means it’s no longer possible to
advance the f- pawn, as the bishop would be hanging. Less important
moves include:

a) After 12 c4 Nf8, 13 Be3 brings us to note ‘a’ to White’s 13th move.

Alternatively 13 Bd2 transposes to note ‘b’ to White’s 13th move in
4C.

b) With 12 Kh1 White is planning to regroup the knight to g1 and f3.

Before completing development, such an operation is risky, and 12...Nf8
13 Ng1 Bc7 offers Black promising play:

b1) After 14 Nf3 Qd5 (threatening ...Bxh3) 15 c4 Qh5 (once again

threatening ...Bxh3) 16 Qd1 Black is obviously doing fine, and is
probably better after 16...Ng6, but I would be very tempted to play
16...g5 with the primitive idea of ...g4. I believe Black has a promising
attack.

b2) 14 f4 Ne6 and now:

b21) A move like 15 Be3 is best met by 15...g6 16 Rae1 (16 Bxg6?? fxg6
17 Qxg6+ Ng7 is simply a piece up for Black; thanks to the threat of
18...Bf5, White does not have a trace of compensation) 16...Ng7, leaving
Black with a better position. This is a typical way of exploiting White’s
weaknesses after he plays f4.

b22) After 15 Bc4?! b5 Black played a model game in Venkatesh-

Mahalakshmi, Abu Dhabi 2019: 16 Bb3 Qd6 17 Qf2 g5 18 Qc2? Ng7 19
Nf3 Bf5 20 Qf2 Nh5 (beautiful play by Black) 21 Nxh4 gxh4 22 Qxh4 Ng7
23 Rf3 Bg6 24 Rg3 Re1+ 25 Kh2 Rxc1 26 Bxf7+ Kxf7 27 Qh6 Rxa1 28
Qxg6+
Ke6 29 Qe4+ Kd7 30 Rxg7+ Kc8 31 Qf5+ Kb8 0-1.

12...Nf8 (D)

White to play
13 Rad1

This was played in a recent game with a rising star playing Black. I must
admit that this gave me a headache, as finding Black’s best set-up was
not easy. Other moves:
a) After 13 c4 Be6 one of my students managed to score an easy win
against a higher-rated player who played too slowly: 14 a3 Qd7 15 Qd2?
Bxh3! – a typical sacrifice for this line – 16 gxh3 Qxh3 17 Nf4 Qg4+ 18
Kh2 Ne6 0-1 G.Lund-Sidali, Fornebu 2017.

b) 13 Qd2 intends to exchange dark-squared bishops. However, this

takes time, and Black is not obliged to cooperate. 13...Be6 14 Rfe1
Qd7 15 Bf4 Be7 16 Be3 Rad8 gave Black a nice position in M.Pap-
Sedlak, Croatian Team Ch, Biograd na Moru 2018.

c) 13 Rfd1 is a rather strange move, as the other rook looks more natural
on this square. 13...Bc7 14 c4 and now the tactics resulting from
14...Qd6?! 15 Bf4 Rxe2? 16 Bxd6 Rxc2 should have proved fatal for
Black in Gavrilin- E.Ufimtsev, Russian Team Ch, Sochi 2019 since 17
Bxc7! Rxb2 18 Ba5! traps the black rook. I therefore suggest 14...g5 15
c5 Bb8 16 Bd2 Kg7, preparing ...Ng6 and ...g4. It’s about to get very
sharp, but Black’s chances are not worse.

d) 13 Rfe1 is another recent try by White:

d1) N.Vignesh-Vaibhav, Xingtai 2019 continued 13...g5. As mentioned

elsewhere, I have doubts about making this aggressive move as it
weakens a lot of squares. After 14 c4 Bc7 15 c5 Bb8 the position was very
complicated and probably equal; then White’s 16 d5 made the game
quickly explode, and both players had chances to get the advantage.

d2) I prefer the flexible 13...Bc7, when Black can adapt to White’s
next move.

We now return to 13 Rad1 (D):

 Black to play
13...Bd7

This flexible move looks a bit slow but gives Black time to organize
his forces optimally. Otherwise:

a) White’s idea was that after 13...Bc7?! he has the typical defensive idea
14 c4 Qd6 15 Nf4 g5? 16 c5!, when Black has failed miserably.

b) In the game B.Haldorsen-Artemiev, European Ch, Skopje 2019,

Black won after 13...Ne6 but White had the better position.

c) 13...Be6, although not my recommendation, led to some interesting

moments in Burke-A.Liang, USA Junior Ch, Saint Louis 2019 after 14 c4
Rc8 (White got the better game after 14...Qd7 15 Nf4 in Firouzja-
Harutjunyan, Internet blitz 2020) 15 Qa4?!. The way I see it, this is a typical
mistake. With the queen far away on the queenside, Black should try to
speed things up on the kingside:

c1) The game continued 15...g5? but experience has taught me that
advancing the g-pawn has serious positional drawbacks. It’s better to
attack with your
pieces – after all, the weakness on h3 is already created. After 16 Qc2
Black might have regretted having his pawn on g5. White will go Bc1 and
f4.

c2) 15...Ng6!, preventing both Nf4 and Bf4, is better. After 16 Qxa7?
Qd7 Black is close to winning, as the sacrifice on h3 cannot be
prevented.

14 c4

Alternatively:

a) 14 Qd2 planning 15 Bf4 is best met by 14...Ng6 15 c4 (15 Nf4 Ne7!

intending ...Nf5) 15...f5, when Black has a fine game; e.g., 16 Bg5 Be7 17
Bxe7 Qxe7 18 Rfe1 Qf6 and Black will continue with ...Rad8 to put
pressure on White’s centre. Chances are equal.

b) 14 Rfe1 can be met in the same fashion as the text-move: 14...Rc8.

14...Rc8

Black is intending ...Bb8, introducing the possibility of ...Qc7.

15 Nc3 Bb8 (D)

 White to play

It hurts to write this, but just before finishing this book, I lost a game
from this position. It was a rather poor performance from my side, and
the only good thing about it is that I remembered my analysis up to this
point. But from here on it was bleak.

16 Bc1

The idea is to play f4 without losing the bishop on e3. Black’s slow attack
can be annoying for White; for instance, 16 Rfe1 Qc7 17 f4 Rxe3!? 18
Rxe3 Qxf4 and Black’s control of the dark squares should give sufficient
compensation.

16...Qc7?!

It’s tempting to line up the queen and bishop on this diagonal, but this is
the first inaccuracy. If given another shot, I would go for 16...b5, which
makes use of the rook on c8; e.g., 17 c5 (or 17 b3 Be6 18 Ne2 Ng6 19 f4
f5 and I believe this tense position gives equal chances) 17...b4 18 Ne2
a5 and ...Be6 is next up, when the bishop aims for an excellent square on
d5. Black has nothing to complain about.

17 f4 Ne6?

More or less a blunder. Better is 17...Rcd8, with the point that after 18
Qf2 Ng6 Black is doing reasonably well. An instructive continuation is 19
Bxg6 fxg6 20 Qxh4 b5 followed by ...g5, when it’s clear that Black’s
strong bishop-pair provides excellent compensation.

18 Qf2

There is no adequate way to protect the pawn on h4.

18...g6?! 19 Ne4 Kg7

My opponent played forcefully from here on and scored a deserved

victory: 20 d5 cxd5 21 cxd5 Nc5 22 Nxf6 Kxf6 23 Qd4+ Ke7 24 Rfe1+ Ne6
25 dxe6 fxe6 26 Bb5 1-0 Suarez Real-T.R.Hansen, British League (4NCL)
online
2020.

4C: Introducing 12 Bd2

(1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 0-0 h4 11 h3 Nd7)

12 Bd2

This looks rather modest but is considered the main line for good
reasons, most of all because it allows White to fight directly for the open
e-file.

12...Nf8 (D)

White to play

I have seen many cases where Black misplays positions like this. Players
tend to be stressed by the fact that White has a queenside majority, but
Black’s central pressure provides enough compensation for this strategic
asset.

13 Rfe1
As Black often follows up with ...Bc7 and ...Qd6, White usually chooses to
keep the option of playing f4, so having the rooks on f1 and e1 makes
sense. That’s why 13 Rae1, which we examine in 4D, is the main move.
White has also tried:

a) After 13 Rad1 I find 13...Be6 or 13...Bc7 most logical, but 13...f5

illustrates a tactical idea that may be useful: the d3-bishop is overloaded,
and Black is able to block out White’s light-squared bishop by playing ...g6
next. Usually Black plays ...g6 followed by ...f5 to achieve this set-up, but
here the idea can be played in reverse order. After 14 Rfe1 g6 15 Qc1 Ne6
16 Nf4 Bxf4 (16...Qc7?, as in Ammon-V.Spitzl, Bamberg 2019, is bad
because of 17 d5!) 17 Bxf4 Nxf4 18 Rxe8+ Qxe8 19 Qxf4 Be6 followed by
...Qe7 the position is roughly equal.

b) After 13 c4 Bc7 14 Rfe1 a tempting option for Black is 14...Ne6 (D):

White to play

b1) 15 Bc3?! Qd6 and White must be careful to survive:

b11) 16 Kf1? (V.Dzjuba-I.Popov, Riga rapid 2019) 16...Qh2 gives Black a

decisive advantage. 17 d5 Nf4 is game over at once, while after 17 Ng1
Bd7
Black threatens ...Nf4; e.g., 18 Bh7+ Kh8 19 Be4 f5 20 Bf3 Ng5 and
things have gone horribly wrong for White. One idea for Black is to play
...Qh1 followed by ...Nxf3.

b12) 16 f4 is the only way to hold on. Nevertheless, Black has a

pleasant choice:

b121) 16...g6 (intending 17...Ng7) is a standard reaction to f4 – we want a

knight on f5! 17 Bxg6? (17 c5! followed by f5 keeps White in the game)
17...fxg6 18 Qxg6+ Ng7 19 Ng3 looks very dangerous, but Black has an
excellent defensive move that kills White’s attack in 19...Bg4!; for
instance, 20 Ne4 Rxe4 21 Qxe4 Be6 and Black is much better.

b122) 16...b5 is another attractive option. Surprisingly the engine says

Black is close to winning after 17 cxb5? (17 c5 is necessary here too)
17...Bb7!.
After connecting the rooks, Black has no back-rank issues and
the weaknesses in White’s position will start to tell.

b2) 15 Rad1 looks like the right move, but after 15...Qd6 16 Nf4 (16 f4 is
best met by 16...f5, when the position is complicated with equal chances)
16...Kf8 (now as the rook is protected, White has to do something with the
knight on f4) 17 c5 (17 Qa4 leads to a position where Black has three minor
pieces for the queen after 17...Nxf4 18 Bb4 Rxe1+ 19 Rxe1 Nxd3 20 Bxd6+
Bxd6 21 Re3 Nf4 which I wouldn’t mind playing with the black pieces)
17...Qd8 18 Ne2 (18 Nxe6+ Bxe6 shouldn’t worry Black thanks to his
control of the d5-square) 18...Nxd4 19 Nxd4 Rxe1+ 20 Rxe1 Qxd4 21 Re4
Qd5 22 Rxh4 Rb8 Black is about to untangle, and the position is roughly
equal.

We now return to 13 Rfe1 (D):

 Black to play

13...Bc7 14 Rad1

Or 14 Nf4 Be6, and now:

a) Taking the bishop is a positional mistake: after 15 Nxe6? Nxe6 –/+

Black threatens ...Qd6.

b) 15 Qd1 g6 and Black is ready to play ...Qd6 and ...f5 with a good position.

c) After 15 Be3 Qd6 Black is threatening ...g5, so White needs to kick

Black’s queen away; e.g., 16 c4 Rad8!? (Rambaldi) 17 Rad1 Bb8 (in case
of White playing c5, it’s useful to have the c7-square available for the
queen) intending ...Bc8 followed by ...Ne6. I believe Black’s position is
easier to play.

14...Be6 15 c4

Rambaldi gives two other moves that shouldn’t bother us: 15 Nf4 Qd6
=+ and 15 Bf4 g6 16 Bxc7 Qxc7 17 c4 Rad8 18 Qc3 Kg7 =.
15...Ng6 (D)

White to play

Exploiting the fact that the white queen will be overloaded. Black plans
to continue with either 16...Qd7, toying with the idea of sacrificing on h3,
or 16...Qd6 to provoke f4. The naïve 16 Bxg6?! fxg6 17 Qxg6? Bxc4
leaves Black much better.

16 f4

This is a typical move for this variation. It is obviously critical, but also a
very committal decision as it weakens a number of squares.

16...Ne7 17 f5

17 Be3 plans Bf2, but leaves White’s position tactically exposed. After
17...Bf5 18 Bf2 (18 Bxf5? Nxf5 19 Qxf5 Rxe3 leaves White’s position in
ruins) 18...Qd7 19 Bxh4 White has won the pawn, but 19...b5 means
it’s Black’s turn to have all the fun:

a) 20 cxb5 allows Black an excellent square on d5; e.g., 20...cxb5 21 Bf2

Bxd3 22 Qxd3 Rad8 with good compensation.

b) 20 b3 bxc4 21 bxc4 Bxd3 22 Qxd3 Ng6 23 Bg3 f5 gives Black

enough compensation for the pawn due to White’s weaknesses.

17...Bd7

It’s hard to find a reasonable plan for White, and I guess that
explains Rambaldi’s suggestion:

18 Kh1

This has the possible idea of playing Ng1-f3, but it takes time.

18...b5 (D)

White to play
I believe Black is doing fine:

a) Here is one line to illustrate that White’s knight manoeuvre

isn’t dangerous: 19 b3 a5 20 Ng1?! a4 21 Nf3?! Bg3 –/+.
b) After 19 c5 Nd5, the knight is a monster and has to be challenged; e.g.,
20 Be4 Rc8 when Black is intending ...Bb8 and ...Qc7:

b1) 21 Qd3 Bb8 22 Bxd5 cxd5 23 Nf4 Qc7 leaves Black a bit cramped,
but the idea of ...Re4 provides counterplay.

b2) 21 Bxd5 cxd5 22 Nf4 Bxf4 23 Bxf4 Qa5 24 Rxe8+ Rxe8 25 c6 Bc8
is very sharp with chances for both sides.

4D: The 12 Bd2 Main Line

(1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 0-0 h4 11 h3 Nd7 12 Bd2 Nf8)

13 Rae1

The e- and f-files are where the white rooks most likely belong.

13...Bc7 (D)

White to play
14 b3

Such a move can be considered a sign that White has failed to get an
opening advantage. However, none of the alternatives change the fact that
Black’s position is perfectly fine.

a) A rather strange draw by repetition occurred in Hraček-Adhiban, Batumi

Olympiad 2018: 14 Kh1 Qd6 15 Bf4 Qd8 16 Bd2 Qd6 17 Bf4 Qd8 ½-½.
The players were clearly keen on a short day at the office, but an
advantage is nowhere to be found for White.

b) 14 c4 looks more direct but gives Black no problems. The engine says
both 14...b6 and 14...g6 are fine, but I suggest developing the queenside
with 14...Be6 – it’s good that following healthy opening principles is a
decent alternative to memorizing long lines – 15 f4 Qd6 16 b3 Rad8 and I
have a feeling White’s space advantage is no strength; instead I see
vulnerable pawns.

c) 14 Nf4 should be met by 14...Be6 intending ...Qd6:

c1) 15 Be3 Qd6 16 c4 Rad8 17 c5 Qd7 is a risky way to play for White.
The d4-pawn is weak, and the d5-square might turn out useful for Black.
Next up is ...Bb8 followed by ...Qc7, and Black is at least equal.

c2) 15 Nxe6 Nxe6 and now:

c21) 16 f4 is best met by 16...Qd6 ideally followed up with ...g6 and

...f5. When White commits to f4, the counterattack with ...c5 is often
effective.

c22) 16 Qd1 plans Qh5, but Black has a standard reply: 16...g6! 17 Qg4
Qd6 18 f4 Kg7 threatening ...Nxd4, and if White defends against it with 19
Kh1, 19...c5 is slightly more comfortable for Black.

c23) 16 Bh7+ Kh8 17 Bf5 Qd6 18 f4 was played in Ostrovsky-Antal,

Charlotte 2018. Here I suggest the calm 18...Kg8, asking White to come
up with a useful plan, and at the same time taking the sting out of Qd1-h5.
I see no plan for White. With moves like ...Rad8 and ...c5 on the agenda,
Black is doing fine.
d) After 14 f4!? the standard reaction would be 14...g6. However, I find
White’s attack after 15 f5 g5 16 Qd1 rather uncomfortable, especially due
to Nf4 ideas. Therefore I recommend 14...Qd6. After 15 c4 I want to
introduce an instructive pawn sacrifice: 15...f5 (intending ...g6, after which
Black’s knight can find an excellent square on e6 or slowly creep towards
the e4- square) 16 Bxf5 Bxf5 17 Qxf5 Re6 (Black wants to double rooks
on the e- file) 18 d5 cxd5 19 cxd5 Ree8 and I feel Black has good
compensation thanks to White’s weaknesses created by f4.

We now return to 14 b3 (D):

Black to play
14...Ne6

The idea is to cover the f4-square in preparation for 15...Qd6. Instead,

Black played 14...g6 in Siugirov-Bai Jinshi, Qinhuangdao tt 2018, which is
also perfectly viable, but I’m not sure if entering the following forcing line is
the best practical choice: 15 Kh1 Qd6!? 16 Bf4 Rxe2 17 Bxd6 Rxc2 18
Bxc7 and after 18...Rd2? 19 Re8! +/– the game went sour for Black.
18...Rxa2! 19 Re8 Kg7 intending ...Ne6 or ...Nd7-b6 keeps the game close
to equal.
15 f4 Qd6 16 Bh7+?!

16 Rd1 is an engine suggestion. The idea is to go for the h4-pawn, but

after 16...Bd7 17 Be1 Black can play 17...g5 with a sharp game where he
is no worse.

16...Kf8 17 Be4?! (D)

Black to play

As White has played f4, the e3- and g3-squares are weakened. Tekeev-
Esipenko, Yaroslavl 2018 continued 17...Bd7?!, but in positions like this, I
am on the lookout for the manoeuvre 17...g6 followed by ...Ng7 and ...Bf5
to exchange the light-squared bishops. Black is already better, especially if
he ends up with a knight on f5. 18 f5? would be the critical test, but it loses
to 18...Qh2+ 19 Kf2 Ng5 20 Bxg5 fxg5 21 fxg6 f5! 22 Bxf5 Bg3+ 23 Nxg3
Qxg3+ and White drops the rook on e1.

Lesson 5: Hansen Variation: Lines with 0-0-0

By: TRH
Chess you don’t learn, chess you understand.

Viktor Korchnoi

When discussing the Hansen Variation with a world-class player, his

reasoning for not fully trusting Black’s position was a feeling that White’s
attack after castling queenside could be dangerous. I had also
suspected at first that this aggressive approach could be problematic for
Black, but it didn’t take long before I was starting to hope that White
would castle queenside.

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-

0 8 Qc2 Re8+ 9 Ne2

After 9 Be3 h5, 10 Ne2 is just another move-order. For White’s other
10th move follow-ups, see 5A.

9...h5 10 Be3 (D)

A different move with the same idea is 10 Bd2. Black should then aim for
variations where the bishop is worse placed on d2 than on e3. 10...Nd7 11
0- 0-0 has occurred in quite a few games, the only difference from the
main line being the placement of the bishop. Luckily, the recipe is simply to
follow Black’s play in the main line. After 11...Nf8 12 Kb1 Be6 13 c4 Rc8
14 Nc3 Bb4 15 Ne4 Be7 the d-pawn is a lot weaker with the bishop on d2
instead of e3. White’s best will soon be to play Be3, essentially losing a
tempo compared to the main line.
 Black to play
10...Nd7

I have always favoured placing the knight on f8, but there is an alternative
set-up in 10...Na6. White can then damage Black’s structure with 11 Bxa6,
but the bishop-pair and, especially, the control of the light squares give
Black enough compensation. If instead 11 0-0-0 Nc7, the knight can go to
d5 or e6, and Black is ready to push his queenside pawns. You may wish
to take a closer look, but in this book I will stick to my main weapon.

11 0-0-0

I’m delighted to see that world-class player and highly respected

theoretician Maxime Vachier-Lagrave has taken up this variation as Black.
His opponent, Karthikeyan, played 11 0-0 in a couple of 2020 blitz games.
Then 11...h4 12 h3 would take us back to our repertoire (4B). However,
MVL decided to play without this move altogether, leaving it outside the
scope of this book.

A hot move right now is 11 Ng3, which somewhat paradoxically seems

to encourage ...h4. For the details, see 5B.
11...Nf8

You will see this knight manoeuvre over and over again. The main idea is
to give extra protection to the king, which means we can use the other
pieces more aggressively on the queenside. Check out 5C for the
interesting, although riskier, alternative 11...b5!?.

12 Kb1

Curiously, having reached this position, all my opponents have chosen

this prophylactic king move. It’s fascinating that everyone thinks the same
way. Still, when faced with the constant threat of capturing the pawn on
a2, White can’t delay this move for long.

12...Be6 (D)

White to play

This position is central to this lesson. Black’s light-squared bishop is one

of the key pieces in Black’s set-up, as it supports a queenside attack.
Additionally, with the pawns mostly on dark squares, White is
potentially weak on the light squares.
 Model Game 5

R. Cumming – T.R. Hansen

London 2017
1 e4 c6 2 d4 d5 3
Nd2

There is no need to stop and think after this move. White avoids a couple
of rare lines, but these are not relevant to our repertoire.

3...dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8 Ne2 Re8 9 Qc2 h5!
(D)

White to play

After this move White must make an important decision. How do you
best punish a move like 9...h5? I have always had the feeling that if
there is a problem with this move, it must be in the lines where White
castles queenside, but honestly I think Black is doing fine.

10 Be3
This move normally indicates that White wants to castle queenside and
prove that the h5-pawn will be a serious weakness in a position with
opposite castling.

In a few games of mine, White has chosen 10 Bd2. The games become
similar to what you are about to see, but the bishop would be better placed
on e3.

10 h3 Nd7 11 g4 is risky without first bringing the king to safety. After

11...h4 (a good reaction) 12 Be3, I suggest 12...b5! (Pranesh-Chitlange,
Abu Dhabi 2019 got back to a position similar to the main line after 12...Nf8
13 0- 0-0 Be6 14 c4 Rc8 15 Kb1 b5; this should be fine for Black, but I like
the text-move even more), when 13 0-0-0 Nb6 gives Black a very quick
attack.
For instance, 14 Kb1 Be6 15 Rhg1 Nd5 16 Bc1, and here, to
accelerate Black’s attack, I suggest the typical pawn sacrifice
16...b4!? 17 c4 b3:

a) 18 axb3? Nb4 19 Qc3 Bd5! (threatening both ...Nxd3 and ...Bf3) 20

cxd5 Qa5 21 Be3 cxd5 leaves White’s king in huge trouble after a black
rook appears on c8.

b) 18 Qxb3 Rb8 19 Qa4 Nb4 gives Black sufficient compensation. The

first threat White has to parry is ...Nxd3 followed by ...Rb4. After 20 Nf4
Bd7 21 c5 Bxf4 22 Bxf4 Nxd3 23 Rxd3 Rb5 Black threatens to bring his
rook to e2 and his bishop to e6. With opposite-coloured bishops in the
middlegame, attacking is more important than ever.

10...Nd7 11 0-0-0 (D)

 Black to play
11...Nf8

The knight on f8 gives the king enough protection, but I’m fascinated by
the complications that occur after 11...b5!?. In 5C you can find analysis of
this incredibly complicated variation. For a long time I was the only one
who had tried it, but that changed dramatically in late 2020!

12 Kb1

White has a few rare but dangerous alternatives here that we give a
closer examination in 5D.

12...Be6

This is a critical position. More than once I have returned to analyse

it properly to make sure White doesn’t have anything.

13 c4

This is the main move but White has a few interesting alternatives, which are
analysed in 5E. However, a couple of brief points about them are
worth making here.

That 13 Nf4? is a strategic mistake is essential for understanding the

whole variation. Black replies 13...Bxf4 14 Bxf4 b5, after which he has nice
control of the light squares and moves like ...a5-a4 and ...Qd5 are easy to
play. Black already has the advantage – see 5E for more information.

13 h3, with the idea of playing g4, looks critical but 13...Bd5 14 Rhg1 b5
15 g4 is not as dangerous as it looks. You can find analysis in 5E.

We now return to 13 c4 (D):

Black to play

I was the first to reach this position in 2017 – twice, actually. On the first
occasion, I failed to respond properly. However, after analysing the
situation, I realized that Black is doing fine. One of the key things to know
about this position is that Black is usually happy to exchange his dark-
squared bishop for the white knight.

13...Rc8!
An important move. Black wants to play ...b5 to obtain control of the d5-
square. The rook move is prophylaxis against the d5 advance and at the
same time prepares ...b5. Fighting for the d5-square is essential. 13...b5?
was my initial thought when I first encountered this position, but after 14
d5! Black is in trouble.

14 Nc3

With this move, White aims to prevent 14...b5. Other ideas:

a) After 14 Ng3, as in Pichot-Perez Ponsa, Buenos Aires 2017, the

simple 14...g6 gives Black a good game.

b) 14 a3 prepares Nc3 without having to face ...Bb4. The immediate

14...b5?! is best met by 15 c5! (Suarez Uriel-Stefanova, Spanish Team Ch,
Linares 2020), which is probably good for White. Instead, I suggest 14...a5
with the idea 15 Nc3 b5 16 c5? (after 16 cxb5 cxb5 17 Bxb5 Re7 the open
files on the queenside will give Black attacking chances) 16...b4!, when
Black is better.

c) 14 h3 b5 (D) and now:

White to play
c1) 15 Rhe1?! is strategically suspect as it allows lines to be opened
towards White’s king. After 15...bxc4 16 Bxc4 Black has many tempting
possibilities. After 16...Bd5 (16...Bxc4 17 Qxc4 Nd7, with the idea of
manoeuvring the knight to d5 via b6, is also fine) 17 Bxd5 cxd5 18 Qa4?!
Rc4 19 Qb5 (Tejaswini-P.Neuman, Prague 2019) 19...Qa8!, protecting the
d-pawn and making room for ...Rb8, Black is clearly better.

c2) 15 c5 and now:

c21) At first, I wanted to recommend 15...Bc7 16 g4 Bd5 17 Rhg1 hxg4

18 hxg4 Bf3 but then I realized White has a very powerful idea: 19 d5!.
The more I analyse these positions, the clearer it becomes that this is a
move we should always have under control. After 19...cxd5 20 Nd4 Bxd1
21 Qxd1 White is an exchange and a pawn down, but I believe he is
completely dominating.

c22) 15...Bb8 16 g4 h4 17 Rhg1 and now 17...g6?! 18 Nc3 gave White

the advantage in Terletsky-Paszewski, Lvov 2019. Instead 17...b4! to
prevent Nc3 is an important improvement:

c221) 18 g5 Bd5 with a typically sharp position with level chances.

c222) After 18 Qa4 a5 19 Ba6 Bd5 it’s Black’s turn to sacrifice the
exchange: 20 Bxc8 Qxc8 and Black’s bishop on d5 dominates the whole
board.

We now return to 14 Nc3 (D):

 Black to play
14...Bb4

When playing this line I have realized that it takes experience to understand
this type of position. For every time I played it, I felt I understood more and
more. Not surprisingly, one of my opponents said it was difficult to
understand the strategic concepts. For instance, knowing that it’s often a
good idea to exchange the dark-squared bishop has helped me a lot, which
is why I played my last move.

The aggressive 14...b5 is possible. 15 c5?!, as played in Boixader Arbo-

T.R.Hansen, Barcelona 2017, is not a refutation because of 15...b4!,
which gives Black the advantage. Instead, White seems to be for choice
after 15 cxb5!, although Black has some compensation.

15 Ne4

Otherwise:

a) 15 Bd2? is a mistake for numerous reasons. Black can play

15...Bxc3 followed by 16...b5, or even capture a free pawn on d4 with
15...Qxd4. In
Santeramo-E.Cvitan, Mitropa Cup (Women), Radenci 2019 Black chose
a third option: 15...b5 16 d5 (16 cxb5 cxb5 17 Bxb5 Bxc3 18 Bxc3 Qd5
is dangerous for White, and after 19 Qa4 Nd7 Black has more than
sufficient compensation) 16...cxd5 17 Nxb5 Bxd2 18 Rxd2 Qb6, when
Black was clearly better and won the game.

b) Black gets everything he wants after 15 a3 Bxc3 16 Qxc3 b5 17 d5 cxd5

18 c5 Qd7 19 Ka1 a6 20 Bd4, with an extra pawn with a fine position. Here
Black should play 20...Qc7, possibly followed with ...Bd7 and ...Ne6, to
challenge White’s blockade. Instead in Tabatabaei-Svane, Internet blitz
2020, Black allowed a tactical resource with 20...Bf5? 21 c6!, when Black’s
queen was overloaded and White was suddenly much better.

15...Be7 (D)

White to play

This renews the threat of ...b5. I remember fearing that White would
retreat the knight with a silent draw offer. However, Black is of course not
obliged to allow a repetition.

16 Rhe1
Now Black gets what he wants. Otherwise:

a) 16 h3 b5! and now:

a1) 17 c5 f5 18 Nc3 b4 19 Ne2 Bg5! =.

a2) After 17 Nc5 Rambaldi gives 17...bxc4 18 Bxc4 Bd5 = as

an improvement on K.Griffith-Rambaldi, Concord 2020.

b) 16 Nc5 is a logical continuation, but Black should be fine if after

16...Bxc5 17 dxc5 he plays 17...Qc7! keeping pieces centralized (rather
than 17...Qa5?!, as in Gukesh-Solozhenkina, Caleta 2019, when the queen
lost contact with the other side of the board and after 18 a3 Rcd8 19 h3 g6
20 Ka1 Bc8? 21 g4 White was better). Now the consistent 18 h3 is
harmless:

b1) 18...Ng6 equalizes. After 19 Bxg6 fxg6 20 Qxg6 Bxc4 21 Qxh5

Bf7 Black’s light-square control is sufficient compensation.

b2) 18...Rcd8 19 g4 Ng6 20 gxh5 (after 20 Bxg6? fxg6 21 Qxg6 Bxc4

Black’s bishop is much better than White’s; e.g., 22 gxh5 Bd5 and all the
threats force White to sacrifice the exchange for insufficient compensation)
20...Nf4 21 Bxf4? Qxf4 and White has weak pawns everywhere, while
Black has the idea of playing ...Rxd3 followed by ...Bf5.

We now return to 16 Rhe1 (D):

 Black to play
16...b5 17
c5

Unfortunately for White, 17 d5? is now out of the question because

of 17...cxd5, which shows the benefit of putting the rook on the c-
file!

At this point, I chose the most direct and aggressive continuation, but
in hindsight a slower approach would have been better.

17...f5?!

There was no rush. First covering the f5-square with 17...g6 is preferable,
and gives Black at least equal chances.

18 Nc3 b4

Continuing to fight for the d5-square.

19 Ne2

The knight is ready to reach the f4-square, from where it could fight for the
light squares.

19...Bg5 (D)

White to play
20 Bxg5

Once again, 20 Nf4?! Bxf4 is a highly favourable exchange for Black.

The game has sharpened, and at this point accurate calculation was
required. 20 Bxf5 Bxf5 21 Qxf5 Bxe3 22 fxe3 Rxe3 is fine for Black; then
23 Rf1 (23 Qxh5? Qe7 is a winning pin) 23...Qe8 gives Black the
advantage.

20...Qxg5 21 g3 Rb8

Threatening 22...b3.

22 Qa4?

The first step in the wrong direction. One thing I have learned through
painful experience is the importance of keeping the pieces centralized.
Better
is 22 Bc4 Bxc4 23 Qxc4 Qf6 with equal chances.

22...Qf6

Indirectly protecting the c6-pawn.

23 b3?

Continuing down the wrong path. White prevents ...b3, but now the queen
is cut off from the rest of the board. Even worse is 23 Qxc6?? Bxa2+,
when Black wins the queen.

23...Bd5 24 Qxa7 Ne6 (D)

White to play

A dream set-up for Black. From the moment White castled queenside,
occupying the d5-square was the main goal. Also, the knight has taken up
an aggressive post. Black has a decisive advantage, despite being a pawn
down.

25 Bc2 Bf3 26 Rc1 Ra8 27 Qb7 Reb8 28 Qd7 Bxe2 29 Rxe2 Nxd4 30 Re3
Rd8 0-1
White resigned because after 31 Qe7 (31 Qb7 Nb5) 31...Nxc2! 32 Qxf6
Na3+ Black ends up with an extra knight.

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ (9 Ne2 h5 10 Be3 Nd7 11 0-0-0 Nf8 12 Kb1 Be6)

5A: 9 Be3

5B: 11 Ng3

5C: 11...b5!?

5D: (12 h3)

5E: (13 h3)

These magnifiers focus on some of the sharpest lines of the entire

Korchnoi Variation. For the competitive player these are lines that should
be studied well in order to gain an understanding of what’s going on,
probably in company with your chess engine and an updated database.
Quite possibly you want to take on the challenges of 5D first, just to ensure
that I haven’t overlooked anything scary in my analysis. For the sake of full
understanding, you probably shouldn’t skip 5C either, even if that’s an
additional option for Black – studying the more direct attacking ideas will
improve your general understanding of the Korchnoi Variation.

If you have a less ambitious approach to theoretical preparation, the

solution is the standard one: play through the main moves, stop at
anything that catches your interest and test the lines in informal games
when you get the chance. Most players are more inclined to study a move
that they have met over the board.

5A: 9 Be3
(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-
0 8 Qc2 Re8+)

9 Be3 (D)

Black to play

This was Svidler’s choice against Carlsen, and immediately after the
game I started to investigate the subtleties of this move-order.

9...h5(!)

The world champion played the modest 9...h6, but we will stick to our
main guns!

10 Nf3!?

Delaying the development of the g1-knight by 10 0-0-0 Be6 11 c4 is a

very ambitious idea. However, playing without the knight also means
White is playing without the rook on h1. In the only practical test, Black
played very directly in the next phase: 11...Na6 12 a3 b5 13 c5 Bf8 14
Ne2 Nc7 15 Nf4 Nd5 16 Nxe6 Rxe6 17 Bh7+ Kh8 18 Qf5 Rxe3 (Suković-
Kalezić,
Montenegrin Ch, Podgorica 2019) 19 Qxh5 Nf4 20 Qxf7 Kxh7 21 fxe3 Qd5
leads to a very complex endgame which is probably slightly better for
White. I believe one preparatory move would have given Black at least
equal chances: 11...a5 12 Ne2 Na6 13 a3 b5 14 c5 Bc7 could be an
improvement.
From this square, the bishop is ready to chop off a knight landing on f4,
and Black is ready to play ...b4.

We now return to 10 Nf3!? (D):

Black to play

In the main line, this knight is developed to e2, a difference the players
have to adjust to.

10...Be6!

An important move! The World Champion was right to sense danger, as

after the standard 10...Nd7?! Svidler went for 11 h3 in a later game. Now,
after 11...Nf8 White has two continuations, both of which are scary:

a) My analysis after the Carlsen game went 12 0-0-0 Be6 13 Nh4 and
by quickly pushing g4, White is likely to get the upper hand.
b) However, Svidler launched his attack with the immediate 12 g4, when
12...Bf4?! 13 0-0-0 Bxe3+ 14 fxe3 Qd5?! 15 e4 +/– followed in Svidler-
Villegas, Internet blitz 2020. In my opinion, 12...hxg4 13 hxg4 Bxg4 is
the critical test. Then 14 Bh7+ Nxh7 15 Qxh7+ Kf8 surely looks
dangerous to Black, but I see no knockout.

11 0-0

This is safer than 11 0-0-0 Na6!. You may be surprised to see how often
the knight can be developed to this square. 12 Kb1 Nc7 13 h3 b5 (D) and
now:

White to play

a) After 14 Nh4? Rb8 15 g4 Nd5 Black’s attack is very dangerous. I have

looked closely at this position, and it’s an excellent starting-point for home
analysis. I will give one sample line, and challenge you to do more
investigation on your own. 16 gxh5? (probably a defensive move like 16
Bd2 is necessary, but Black is already for choice) 16...b4! 17 c4 b3! (you
will see more of this motif later!) 18 axb3 Nb4 19 Qc3 Qa5 and Black’s
attack is decisive.

b) 14 g4 is best: White needs to play with the utmost vigour to avoid a

disadvantage. After 14...Bd5 15 Nh4 deep computer analysis leads to a
“0.00” assessment. There are several ways the game can end in a
repetition, such as 15...g5 16 Nf5 Bxh1 17 Rxh1 h4 18 Nh6+ Kg7 19 Nf5+
Kf8 20 Nh6 Kg7 21 Nf5+ or even 15...f5 16 Nxf5 Bxh1 17 Rxh1 h4 18 Qd2
Nd5 19 Bg5 Qc7 20 Nxg7 Kxg7 21 Bh6+ Kh8 22 Qg5 f6 23 Qh5 Re7 24
Bg7++, but there are many ways for both sides to deviate, while 15...Bxh1
16 Rxh1 Nd5 17 gxh5 Bf4 18 Rg1 is also dynamically equal, but with no
forced draws in sight.

11...Nd7 (D)

White to play
Now:

a) After 12 Rfe1 Nf8 13 Rad1 Black can play 13...Bc7 with the idea of
freeing d6 for the queen, provoking a weakness in front of the white
king. Black should have no problems.

b) 12 Nh4 Nf8 13 Nf5 Bxf5 14 Bxf5 g6 15 Bd3 f5 is fine for Black. The mass
of pawns blunts White’s light-squared bishop, while Black’s development
flows easily.
c) 12 c4 has been tested. After the logical 12...Nf8 13 Bf5, instead of
13...Qd7, as in Külaots-Shtembuliak, Moscow 2020, I prefer 13...g6 14
Bxe6 Nxe6. Next up is ...Qc7 and ...Rad8. Later Black can improve the
knight by going to g7 and eventually to f5.

5B: 11 Ng3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-

0 8 Qc2 Re8+)

9 Ne2 h5 10 Be3 Nd7 11 Ng3

This move has been popular lately. Here it’s important for Black to play
a little differently from the way we are used to.

11...Nb6 (D)

White to play

With White’s king still in the centre, Black wants to play ...Nd5.

12 0-0
Or:

a) After 12 Nxh5 White must soon return his pawn; e.g., 12...Nd5 13 0-0
Bg4 14 Ng3 Bxg3 15 hxg3 Nxe3 16 fxe3 Rxe3 17 Rae1 Qe7 18 Qd2 Re8
19 Kf2 Rxe1 20 Rxe1 Qxe1+ 21 Qxe1 Rxe1 22 Kxe1 was equal and soon
drawn in J.Bjerre-Thybo, Århus 2018.

b) 12 c4 Bf4 13 0-0-0 Bxe3+ 14 fxe3 h4 (a clever move as after 14...Rxe3,

White should avoid 15 Nxh5? Bg4 –+, but 15 Qf2 offers him good
compensation for the pawn) 15 Nf1 (15 Nf5? Bxf5 16 Bxf5 Rxe3 17 Bh7+
Kf8 18 Qf2 Nxc4 leaves Black two pawns up, and he soon won in
Charmeteau-Hera, French Team Ch 2019) 15...Bg4 16 Rd2 (16 Re1 c5!
with the idea of 17 d5 Nd7, reaching the e5-square with at least equality)
16...Nc8!. Now that White’s attack has been slowed down, Black takes
time to regroup. The plan is ...Be6 and ...b5, leaving him slightly better.

c) 12 0-0-0 Nd5 (D) (with a knight on d5, Black’s attack is usually

much stronger than when it’s on f8) and now:

White to play

c1) 13 Rde1 is not dangerous for Black. 13...Be6 14 Kb1 Qa5 (White is
already facing a dangerous attack) 15 a3 Rad8 16 Ne4 Nxe3 17 fxe3 c5 18
c4 cxd4 19 Nxd6? Rxd6 20 e4 Rc8 21 Qe2 Rb6 22 Ka1 Rb3 23 Qc2
Rxa3+! 24 bxa3 Qxa3+ 25 Kb1 Rc6! 26 c5 Rxc5 0-1 Bernotas-Berkes,
European Ch, Batumi 2018.

c2) 13 Bd2 Nf4 14 Bxf4 Bxf4+ 15 Kb1 and then:

c21) 15...Qa5 16 Bh7+?! Kf8 17 Bf5 Bxf5 18 Nxf5 g6 19 Nh4 (Abdumalik-

Lenderman, Saint Louis 2017) 19...Qd5 =+; e.g., 20 Rhe1 b5.

c22) Another interesting idea is to provoke weaknesses with 15...Qd5 16

f3 (16 c4 Qxg2 17 Nxh5 Qf3 is complicated with equal chances) 16...Be6
17 c4 Qa5 and again, the position is roughly equal.

We now return to 12 0-0 (D):

Black to play

12...Nd5 13 Bd2

Or 13 Ne4:
a) If you feel like proving that the endings in the Korchnoi Variation are
not as bad as some people think, 13...Bf4 is the move for you. The line
14 Bxf4 Nxf4 15 Rfe1 Nxd3 16 Qxd3 Bf5 more or less guarantees Black a
draw. A logical continuation is 17 Qf3 Bxe4 18 Rxe4 Qd5 19 Rxe8+ Rxe8
20 Qxd5 cxd5 21 Kf1 Re6 and we have the rook endgame discussed in
the Introduction. The only difference is that the pawn is on d5 instead of
c6, which makes things a lot easier for Black.

b) 13...Bc7 14 Bd2 was tested in Imnadze-E.Cvitan, Austrian Women’s

Team Ch 2018/19. Here I wouldn’t mind grabbing the bishop-pair with
14...Nf4 15 Bxf4 Bxf4. The position is equal, but Black has long-term
chances due to his bishops.

13...h4 14 Ne4 (D)

Black to play

This was played in Bruzon-A.Liang, Greensboro 2019. Here, instead of

inserting 14...h3 15 g3, Black should have tried to make use of the knight
on d5 and exploit the f4-square with 14...Nf4 15 Nxd6 Qxd6 16 Rae1 Bd7
=.
5C: 11...b5!?

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-

0 8 Qc2 Re8+ 9 Ne2 h5 10 Be3 Nd7)

11 0-0-0

There is nothing wrong with 11...Nf8, but for variety’s sake and to avoid
preparation, it’s useful to search for opening surprises within your field
of interest. I came well prepared to one of my games, and played a
novelty:

11...b5!? (D)

White to play

Rather than placing the knight defensively on f8, the idea is to

play ambitiously and head for b6 and d5.

This move was probably not on White’s radar, but I am happy to see that it
is now getting some worldwide attention. Jorden van Foreest gave it a shot
in an online blitz game, but much more importantly, Magnus Carlsen has
been
attracted by Black’s chances. The day before sending the manuscript of
this book to the publisher, I was positively surprised to see that none other
than the World Champion played it against Duda in Norway Chess 2020,
which I consider a major stamp of approval.

12 d5

This is the engine’s first choice and the move chosen by Duda and my
opponent, Ivić. After the game my opponent said that he had briefly
checked it in his preparations and saw that the engine liked White. The
best opening discoveries are when you can prove the engine evaluation
wrong, which I believe is the case here. The reason why I’m attracted by
this line is that any hesitation can easily turn out to be fatal for White.
White has a few alternatives:

a) In Azarov-J.van Foreest, Internet blitz 2020 there followed 12 Kb1?! Nb6

13 Bc1 (Black’s main idea can be seen after 13 h3 Nd5 14 Bc1 b4! 15 c4
b3!, when I wouldn’t be surprised if Black is already winning). Here Black
chose 13...Be6, but my preference is 13...a5, when Black seems to have a
strong attack and the better position.

b) A student of mine suggested the crazy-looking 12 g4!?. I would have

been shocked if faced with such a move in a real game, but now I know it
must be taken seriously. Probably the sensible choice is to play 12...h4
(12...hxg4 13 h3 g3 14 Rhg1 gxf2 15 Bxf2 Nb6 16 Qd2 Kf8! looks
incredibly risky for Black, but the engines say there’s no problem) 13 d5 c5
and it’s a messy position, but Black should be doing fine.

Both Magnus’s and my game followed my analysis for a number of moves:

12...c5 13 Bxb5 Rb8 14 c4 a6 15 Ba4 (D)

 Black to play

At this point Magnus took a big think, and indeed it’s a critical position.
Even the overly drastic 15...Rxe3 is possible, when Black enjoys long-term
compensation after 16 fxe3 Nb6. However, I believe in a slower approach.

15...Re5

Black wants to play ...Bf5. I had been sure this was the strongest move,
and originally gave it an exclamation mark. I had looked at the alternatives,
and although they are very interesting, I liked my choice more. However,
it’s hard to argue against a world champion’s choice.

Magnus chose 15...Re7, planning to swing the rook over to the b-file. This
is also very interesting, and can lead to slower-burning play. Then:

a) I have a feeling 16 Rhe1 is a good move; e.g., 16...Ne5?! 17 Bd2 and

it’s not obvious how Black continues; actually, in many lines after 15...Re7,
Bd2 makes a lot of sense as it stops ...Rb4. But the calm 16...g6 looks a
lot more playable, as ...Bf5 will soon be useful; e.g., 17 Bd2 Nb6 18 Ng3
h4 19 Ne4 Bf5 20 Bb3 (20 Nxf6+ Kh8!) 20...Be5 21 Ba5 Kg7 and Black
retains positional compensation.
b) Duda, unfamiliar with the complexity in this line, replied 16 Ng3 (D) and
the position was extremely complicated as on basically every move the
players have interesting tries:

Black to play

b1) The game continued 16...Ne5?! 17 Ne4?! (in my opinion 17 Bd2! is

stronger; in the game Magnus got use of the b4-square) 17...Reb7 and
again there are several options:

b11) 18 Nxd6? Rxb2 19 Qxb2 Rxb2 20 Kxb2 Qxd6 leaves Black much
better due to White’s exposed king.

b12) 18 Bb3?! is strongly met by 18...Bf5 intending ...a5-a4, which is

more than awkward for White.

b13) 18 b3 was the game continuation. Magnus was happy to sacrifice an

exchange to get at White’s king: 18...Rb4 19 Bd2 Rxa4 20 bxa4 Bf5 21
Rde1 and in Duda-Carlsen, Stavanger 2020 the World Champion played
21...h4? and eventually lost, ending his unbeaten streak of 125 classical
games.
However, after 21...Ng4 the engine’s verdict is 0.00, although I see a lot
of fun. I believe this game will attract more players to this complicated
and
practically unexplored variation.

b2) 16...Nb6!? 17 Bc6 f5 (after the less patient 17...Rxe3 18 fxe3 h4 19 Nf1
Bg4 Black has compensation, and White has to return some material) 18
Bg5 f6 19 Bd2 h4 20 Nf1 f4 leads to a strange situation where Black is a
pawn down but White’s pieces are uncoordinated and out of play. Just
imagine Black getting to play ...Re5 and ...Bf5.

b3) Here too 16...g6!? is worth considering and can transpose to the 16
Rhe1 g6 line; e.g., 17 Bd2 Nb6 (D) and then:

White to play

b31) 18 Bb3 h4 19 Ne4 Bf5 20 Rde1 (20 Rhe1 transposes to line ‘a’)
20...Be5 21 g4 (21 Ba5 Qf8!? with ideas including ...Na4 and ...Nxd5)
21...Bxg4 22 Rhg1 Bf5 23 Nxf6+ Bxf6 (23...Kh8?! is less convincing after
24 Ne4 Na4 25 f3) 24 Qxf5 Rxe1+ 25 Bxe1 offers Black a choice between
25...Nd7 intending ...a5, and 25...a5 26 Bxa5 Bxb2+. There is a fair degree
of choice here, and very rich possibilities.

b32) 18 Bc6 h4 19 Ne4 Bf5 20 Rde1 Be5 21 b3 Nc8 22 g4 Bxg4 23

Rhg1 Bf5 24 Nxf6+ and again (like in line ‘a’) the clever 24...Kh8! gives
Black
compensation after 25 Ne4 Nd6 26 f3 f6! (with some ...g5 ideas and to
meet Bc3 with ...Kg7). White is a little tied up and Black’s pieces have
good scope.

16 Nc3 Nb6 17 Bb3 Bf5 (D)

White to play
18 Qd2

The only place I can find a line where the engines give White a plus is after
18 Qe2 Bg4 19 f3 Qc7 20 Rd2 Rbe8 21 Nd1. Finding these moves is far
from easy, but even so, I’m sure this isn’t as glamorous as the engines
claim.
Imagine 21...Bc8 followed by ...a5-a4-a3. Black surely has compensation.

18...a5 19 a3 Nxd5 20 Nxd5 Rxb3 21 Nc3 Rb6 22 Nb5

All of this happened in Ivić-T.R.Hansen, European Clubs Cup, Porto

Carras 2018. When sitting at the board I remembered having looked at
this, and I knew the engine’s evaluation was 0.00. The problem was that I
had never checked how the line went. I spent almost all my time and
failed to find the right continuation, and instead got lost in the
complications. Here is the
amazing draw:

22...Re4 23 b3 Qe8 24 Nxd6 Rxb3 25 Qd5 Rexe3 26 Nxe8 Re2

Despite being a full queen up, White must play 27 Qxf5 Rc3+, when it’s a
draw by perpetual check. It surely is a powerful line, but to play this way
you must have done your homework!

5D: 12th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 Be3 Nd7 11 0-0-0)

11...Nf8 (D)

White to play

12 h3

Other moves:

a) 12 c4 Be6 13 Nc3 (13 Kb1 transposes to Model Game 5) 13...Bb4! and

just like in the main line, Black is eager to exchange his dark-squared
bishop for the white knight; e.g., 14 Ne4 b5 (with the king still on c1, this is
stronger than ever) 15 cxb5 (15 c5 Bxa2 is a pawn for free, as 16 b3?
loses to 16...Qa5) 15...cxb5 16 a3 and both 16...Rc8 and 16...Bd6 are
obviously fine for Black.

b) 12 Ng3?! is an attempt to attack the h-pawn immediately. Bear in

mind that it’s not necessary to defend it, as Black has a counterattack:

b1) After 12...Be6!?, White’s best is probably 13 Kb1, transposing to note

‘b’ to White’s 13th move in 5E, as 13 Nxh5 Qa5 hits both a2 and h5.

b2) 12...g6 13 Bh6?! (as I have also faced this plan, I guess this move
comes naturally to players who are unfamiliar with this line) 13...Be6, and
Black is already better. After 14 Kb1 (14 Bxf8? Kxf8 15 Bxg6?! fxg6 16
Qxg6 Bf4+ 17 Kb1 Bf7 –+ gives Black an extra piece) Black has a
pleasant choice:

b21) After 14...b5 White got desperate and tried 15 d5? Bxd5 16 Rhg1? in
De Jonge-Elgersma, Dieren 2019, when 16...f5 –+ would have left White
with no compensation at all. The position was bad, but not that bad: 15 Ne4
f5 16 Nxd6 Qxd6 is the type of position Black is aiming for.

b22) 14...f5 15 Bc1 b5 16 Ne2 a5 17 f3 a4 18 g4 fxg4 19 fxg4 hxg4 20

Nf4?! (time and again I see White make this strategic error, but in this case
Black already had a clear advantage) 20...Bxf4 21 Bxf4 Qd5 was the
dream scenario for Black from T.Dahl-T.R.Hansen, Norwegian Team Ch
2017 that we saw in the introduction to Lesson 3.

12...Be6 (D)
 White to play
13 c4

13 g4?! is best met by 13...Bd5! as the bishop will often find a beautiful
square on f3; e.g.: 14 Rhg1 (14 gxh5 Bxh1 15 Rxh1 Qa5 gives White
insufficient compensation for the sacrificed exchange) 14...h4 15 g5?!
Bh2! 16 gxf6 Qxf6 17 Rg4 (White has to do something as otherwise
Black will launch his usual attack, starting with ...b5) 17...Bf3 –/+.

13...b5 14 c5

14 cxb5 gives Black a pleasant choice between 14...cxb5, 14...Rc8,

14...Bxa2 and 14...Qa5 – all looking good!

14...Bc7

Black has gained control of the d5-square. Next up is depriving White of

the means to fight for this square. For instance, 15 Rhg1 b4 16 g4 h4 17
g5 Bd5 18 Nf4 Bxf4 19 Bxf4 g6, threatening ...Ne6, with an attack against
both the bishop on f4 and the pawn on g5. The engine says all moves are
equal, and I wouldn’t mind having the black pieces.
5E: 13th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 6 c3 Bd6 7 Bd3 0-0 8

Qc2 Re8+ 9 Ne2 h5 10 Be3 Nd7 11 0-0-0 Nf8)

12 Kb1 Be6 (D)

White to play
13 h3

I have never faced this move, but I consider it so critical that I check the
lines regularly. White’s idea is to use the h5-pawn as a hook and play g4.
Other alternatives to 13 c4 include:

a) 13 Nf4? is another version of this typical mistake. By now, we know

that Black is happy to exchange his dark-squared bishop for the knight.
After 13...Bxf4! 14 Bxf4 b5 Black is about to take control of the light
squares:

a1) 15 Be4 Bd5 16 Rhe1 Bxe4 17 Rxe4 Ne6 and in Abrahamyan-Krush,

USA Women’s Ch, Saint Louis 2017 (the game that made me aware of
the
whole variation) Black was better. Next up is ...Qd5 and ...a5 with an
easy game.

a2) 15 Rhe1 a5 16 Be4?! Qd7 17 f3?! a4 18 h3 f5 19 Bd3 (White has been

playing too slowly, and Black is already winning) 19...Qd5 20 Rxe6 Nxe6
0- 1 Knüdel-Pubantz, Lüneburg 2019.

b) For some reason, I have often faced 13 Ng3?! g6, and now:

b1) 14 Bh6?! transposes to note ‘b2’ to White’s 12th move in 5D,

where White gets punished for hesitating.

b2) 14 Ne4 can be met by 14...f5!? 15 Nxd6 (15 Bg5? fxe4 16 Bxd8
exd3 gives Black three well-coordinated pieces for a queen, and he is
simply winning) 15...Qxd6 is a another version of the incorrect exchange
of minor pieces by White. Next up is ...b5 and ...Qd5 with a comfortable
position.

13...Bd5 (D)

This is logical as it immediately pins the g-pawn.

13...b5 was played in Gutenev-Indjić, Sochi rapid 2018, but I see no reason
to opt for this move, as after 14 g4! Black has no time to play 14...Bd5? as
the exchange sacrifice after 15 gxh5 is strong.
 White to play
14 Rhg1

It seems logical to force through the g4 advance. Other moves:

a) White cannot play the direct 14 g4? as 14...Bxh1 15 Rxh1 h4 keeps

the position closed, so White is left with insufficient compensation.

b) It’s fascinating how many strong players misplay these positions. 14

Rdg1?! b5 15 Nf4?! is a common mistake. Black is happy to give up the
dark-squared bishop to get rid of White’s knight and obtain control of the
light squares. After 15...Bxf4 16 Bxf4 Ne6 Black was already for choice
in Khanin-Oparin, Russian Team Ch, Sochi 2018.

14...b5 15 g4 h4

If you really want to play with fire, the line 15...hxg4 16 hxg4 Bh2 17 Rh1
Qc7 is 0.00 according to the engine, but that doesn’t mean it’s boring.

16 g5
This direct attack involves an exchange sacrifice.

16...Bh2 17 Rh1 Bxh1 18 Rxh1

This computer line has been in my files for a long time. Black is probably
slightly better, but it’s extremely complicated and newer programs may
come up with improvements. Stockfish prefers 18...Bg3!? but 18...Bd6 and
18...Bc7 also seem possible.
2: Classical: Capablanca Variation

Lesson 6: Rare Fifth Moves

By: SJ

A good player is always lucky.

Jose Raul Capablanca

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 (D)

White to play

This is most commonly known as the Classical Variation but that’s what we
call 3 Nc3. Therefore we’ll call it the Capablanca Variation, which is also a
widely recognized name. Not only was this line a frequent choice of the
great Cuban, it also has several classical qualities in common with him; it’s
solid but reasonably active and not too theoretical. Nevertheless it has over
the
years been played by many of the game’s greatest and has a
considerably more detailed body of theory than the Korchnoi Variation of
the previous lessons. However, we hope that you will find this first lesson
a relatively comfortable introduction as there are relatively few GM games
in these lines and therefore fewer lines that lend themselves to
memorization.

Now White’s clear main line is 5 Ng3, which is the subject of the next
six lessons. However – as we shall see – White’s knight is often
misplaced on g3, so the following rare options have at least some logic
behind them:

5 Nc5 and the really rare 5th-move alternatives are covered in

Model Game 6.
The gambit 5 Bd3 is analysed in
6A. 5 Qf3 is the subject of 6B.

Model Game 6

T. Burg – Hamblok

Belgian Team Ch 2012/13

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Nc5

White’s alternatives to 5 Ng3 are so rare that many experienced

tournament players have never taken the time to study them in detail. The
idea of a white knight attacking an unprotected black pawn on b7 on the
5th move somehow feels alien.

The even rarer 5 Nc3 loses time. Black should be equal after 5...Nf6, which
could also be reached by the move-order 4...Nf6 5 Nc3 Bf5 (see 1B). After,
e.g., 6 Nf3 e6, a typical example of play is 7 Be2 Bd6 8 Bg5 Nbd7 9 0-0
Qc7 10 Bh4 0-0-0 11 Bg3 Bxg3 12 hxg3 Nc5 13 Qc1 Nce4 = Thorfinnsson-
T.R.Hansen, Nordic Junior Team Ch, Östersund 1992 (Fun fact: this game
by the 12-year-old Torbjørn is his oldest to be found in the databases).

We now return to 5 Nc5 (D):

 Black to play
5...e5!

5...Nd7 is a good alternative but this is the most ambitious approach and
we’ll only offer alternatives for Black when we see clear benefits in doing
so.

6 Nxb7

The only consistent move. 6 Bd3?! Bxc5 7 dxc5 (7 Bxf5 may be better
but 7...Ne7 still looks like an edge to Black) 7...Bxd3 8 Qxd3 Qxd3 9 cxd3
Na6 10 Nf3 0-0-0 –/+ Cosain-Abdel Aziem, Kuwait 2013.

6...Qb6 7 Nc5 exd4 (D)

7...Bxc5 also seems perfectly fine but once again Black doesn’t
need alternatives.
 White to play

White’s slightly sounder pawn-structure is amply compensated by

Black’s greater piece activity.

8 Nb3

A rare but interesting alternative is 8 Na4 as Black’s c-pawn is a more

real weakness after 8...Qa5+ 9 c3 dxc3 10 Nxc3 when the c-file is half-
open.
Willow-Sidorika, Internet Under-18 rapid 2020 continued 10...Nf6 11 Nf3
Ne4 12 Bd2 Nxd2 13 Qxd2, and now 13...Bc5 would have led to an
equal and interesting position.

8...Bb4+

This is more common and scores better than 8...c5.

9 Bd2

Undeveloping with 9 Nd2?! looks too unnatural even if the knight could
have a bright future on c4.
9...Bxd2+!?

9...Nf6 is more standard but this looks just as strong.

10 Qxd2 c5 11 0-0-0?! (D)

11 Bd3 Bxd3 12 cxd3 may be best, with rough equality. After 12...Ne7,
Varga-Korpa, Budapest 2015 went 13 Nf3 0-0 14 0-0 Nd7 15 Qg5 Rfe8 16
Rfe1 h6 17 Qh5 Qd6 18 Nfd2 Qd5 19 Qxd5 ½-½. However, with that
move- order, White could possibly have created some challenges with 13
Qg5, so 12...Nd7 13 Nf3 Ne7 14 0-0 0-0 may be a more accurate way to
reach the same position.

Black to play

Castling queenside looks risky. The knight is vulnerable on b3 and the

half- open b-file aids Black’s attack. Black probably has an advantage
already – or at least his position is easier to play for most humans (who
tend to attack better than they defend).

11...Ne7 12 Qg5?!
This is far too greedy. 12 Qa5 0-0 13 Nf3 Rc8 14 Bc4, as in Narciso
Dublan- Arjun, Køge 2013, isn’t much better as 14...Qf6 15 Rhe1 Nd7
gives Black an obvious plus.

After a developing move like 12 Nf3 it’s also difficult for White, but nothing
is clear.

12...Nbc6 13 Bd3?

This normal-looking move simply loses. Suggesting an improvement is

hard but development with 13 Nf3 could be best.

13...Bxd3 14 cxd3 Nb4!

This has to be played with a specific idea in mind, as Black ignores his
own development and his king suddenly looks vulnerable on the open e-
file.

15 Kb1 (D)

Black to play

Do you see how Black should proceed?

15...Nxa2!

Black has to keep up the initiative.

16 Re1

There is no hope after 16 Kxa2 Rb8 17 Ka1 Qxb3 18 Rd2 Nd5; e.g., 19
Re2+ Kf8 and now 20 Nf3 allows 20...Qa4+ 21 Kb1 Nc3+ 22 Kc1 Qa1+ 23
Kd2 Rxb2#.

16...0-0 17 Nd2

White is completely helpless after 17 Qxe7 Qxb3. Black will simply

double on the b-file and then play either ...Nc3+ or ...Nb4.

17...Nd5 18 Kxa2

Black is a piece down but on the queenside his numerical superiority

is decisive. Now he has a number of winning continuations but finds
the quickest forced mate:

18...Nc3+! 19 bxc3 Rab8

There is no defence against the double mate threats.

20 Nc4 Qb3+ 21 Ka1 Qa4+ 22 Na3 Qxa3# (0-1)

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 (5 Qf3)

6A: 5 Bd3

6B: 5 Qf3

These magnifiers were included mostly for completeness. In his book on

the Caro-Kann, Bologan has the following to say about the moves in
question:
“During work on this book, I was surprised to discover that in this position,
the moves 5 Qf3 and 5 Bd3 have been played, and by some quite good
players. What can one say? I have played the Caro-Kann for more than
15 years and won several tournaments with its help (including Dortmund
2003) but knew nothing about these moves.”

If Bologan could wait for 15 years, you can probably do the same –
unless one of your clubmates happens to specialize in one of these
lines.

6A: 5 Bd3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5)

5 Bd3 (D)

Black to play

This developing move is very natural except for the fact that it offers
an important centre pawn.

5...Qxd4
Probably best, but White gets some play for the pawn. A reasonable
alternative is 5...Bxe4 6 Bxe4 Nf6 7 Bf3 and now one idea is 7...Nbd7
with the point 8 Ne2 e5! =.

6 Nf3 Qd8

Not quite forced, but the queen is probably more vulnerable on other squares.

7 Qe2

7 0-0? Bxe4 8 Bxe4 Qxd1 9 Rxd1 Nf6 10 Bd3 Nbd7 11 c4 e6 12 Bf4 Be7
gave White virtually nothing for the pawn in Voiculescu-Y.Reich, corr. 2006.

After the text-move White probably has approximate compensation for

a pawn. That’s another way to say that chances are equal in an
interesting position, which isn’t a bad starting point for Black at all.

7...Bxe4

This gives up the bishop for a knight in order to gain some time for
development. Black should obviously avoid 7...Nd7?? 8 Nd6#, but
7...Nf6!? could actually be Black’s best: 8 Nd6+ (8 Nxf6+ gxf6 9 Bxf5
Qa5+ 10 Bd2 Qxf5 11 0-0-0 =+) 8...Qxd6 9 Bxf5 g6 10 Bh3 Bg7 11 0-0 0-
0 =+.

8 Bxe4 Nf6 9 Bd3 e6 10 0-0 (D)

Alternatively, 10 Bg5 Nbd7 11 0-0-0 Qc7 12 Bh4 Bd6 13 Bc4 0-0-0

=+ Milner-Barry – Regedzinski, Buenos Aires Olympiad 1939.
 Black to play

Black is solid and has an extra centre pawn but White’s bishop-pair and
freer piece-play provide some compensation. Following 10...Be7 11 Rd1
(11 Bd2, planning c4, Bc3 and Rad1 could be a better try to keep up some
play) 11...Nbd7 12 c4 Qc7 13 a3, 13...0-0-0?! 14 b4 left the position
unclear in Mamoshin-Slugin, Gagarin 2009. Instead, 13...0-0 looks safer
and should be at least a little better for Black.

6B: 5 Qf3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5)

5 Qf3 (D)
 Black to play

This may look primitive but attacking Black’s bishop cannot be that bad,
and the move has been played by Tal and Bronstein among others.

5...e6

Black protects his bishop with a developing move that is more useful
than White’s queen sally. Now he threatens to capture the d-pawn.

6 Ne2

This develops a new piece while protecting the d-pawn. The downside is
that for a moment White shuts in his light-squared bishop and that the
knight really doesn’t have a bright view from e2. However, the alternatives
aren’t really better:

a) 6 c3 is quite harmless; e.g., 6...Nd7 7 Bf4 Qb6 8 0-0-0 Ngf6 9 Nxf6+

Nxf6 10 Bd3 Bg4 11 Qg3 Qa5 12 f3 Bf5 13 Bxf5 Qxf5 = Djurić-Šolak,
Serbian Team Ch, Vrnjačka Banja 2010.

b) Arguably 6 Bd3?! is critical but it seems Black can safely snatch the pawn:
6...Qxd4 7 Ne2 Bxe4 8 Bxe4 Qd8 9 Qb3 Nf6!? (9...Qb6 –/+ requires less
calculation) 10 Qxb7?! Nxe4 11 f3 (11 Qxa8?! Bc5 12 0-0 Qb6 and Black
is more or less winning thanks to the trapped queen) was played in
Bensiek- Von Weizsaecker, corr. 2000, and now 11...Nd7! 12 fxe4 Rb8 13
Qxc6 Bc5! leaves White in desperate trouble.

6...Nd7 7 N2g3

White attacks Black’s light-squared bishop. His problem is that this knight
will hardly have better prospects on g3 than the other knight has in the
main lines after 5 Ng3. Chances are equal; e.g., 7...Bg6 8 h4 Ngf6 9 h5
Bxe4 10 Nxe4 Nxe4 11 Qxe4 Nf6 12 Qf4 h6 13 c3 Qd5 = Nasuta-
Ducarmon, Internet rapid 2020.

Lesson 7: Rare Sixth Moves

By: SJ

Ninety percent of the book variations have no great value, because either
they contain mistakes or they are based on fallacious assumptions; just
forget about the openings and spend all that time on the endings.

Jose Raul Capablanca

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 (D)

 White to play

In this lesson and the next, we shall examine White’s alternatives to the
main line with h4 and Nf3. The lines are quite varied but most involve
bringing a knight to f4, either to exchange it for a bishop or to gain a tempo
by attacking it, forcing a retreat to h7.

There is a multitude of move-orders and possible transpositions within

each lesson and also between the two lessons. However, most lines with
an early h4 are covered in the next lesson, which starts out 6 h4 h6 and
only now something else than 7 Nf3, while lines where White plays his
knight to f4 before advancing with h4 are covered here. I shall recommend
a set-up where Black fights for the b8-h2 diagonal and in particular the f4-
square with the moves ...e6, ...Bd6 and ...Qc7 or ...Qh4. The following
position (with White to move after 6 N1e2/Nh3 e6 7 Nf4 Bd6 8 h4 Qc7) is
quite typical.
 White to play

Black will give up both his bishops for knights (9 Nxg6 hxg6 10 Ne4 Nf6 11
Nxd6+ Qxd6), in exchange for a lead in development and play against the
white king, which will not be safe on the kingside thanks to his h-pawn
advance and Black’s half-open h-file.

Sometimes this priority will mean that the natural developing move ...Nf6 is
delayed. This way it will matter to us little whether White’s knight goes to f4
via h3 or e2. What does make a difference is the move Bc4, because we
want a consistent repertoire and cannot avoid the position of Lesson 8
arising after the moves 6 h4 h6 7 Nh3 e6 8 Nf4 Bh7 9 Bc4 Nf6 anyway.
Therefore I shall frequently recommend simply meeting Bc4 with ...Nf6,
transposing to Lesson 8.

Model Game 7

T. Vos – Turov

Haarlem 2017
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3

White’s knight is surprisingly often misplaced here, and later in the

game White frequently struggles to get it back to e4.

5...Bg6

Both sides have had to retreat their sole developed minor piece. Black’s
bishop is considerably more active than White’s knight, but also
somewhat exposed.

6 N1e2

White’s most popular move here is 6 h4, which is the subject of Lessons 8 to
12. Second in popularity is 6 Nf3, which will usually lead to the same
lessons as White normally follows up with 7 h4 (other moves after 6...e6
are covered in 7A and 7B).

6 Nh3 (known as the Flohr Variation) is the ‘twin’ of the text-move and will
normally transpose after 6...e6 7 Nf4. White either blocks his light-squared
bishop or his h-pawn, and there are some independent lines depending on
which route the knight takes, but in our repertoire they generally come to
the same thing.

Rarer alternatives include 6 f4, which you find in 7C, and 6 Bc4,
which unless it transposes to Lesson 8 is covered in 7D.

We now return to 6 N1e2 (D):

 Black to play
6...e
6

The lines based on 6...e5?! (intending either a queen check on a5 or a

queen exchange and then piece-play against the exposed e-pawn) are
less popular than they used to be (7 h4!? may be the most critical reply),
and in some ways less practical as then there are significant differences
depending on whether White chose the h3 or e2 path for his knight. It is
also worth mentioning that trying to keep the bishop-pair with 6...h6?! is
too slow when White hasn’t spent time on h4. A typical line is 7 Nf4 Bh7 8
Bc4 Nf6 9 0-0 e6?! 10 Re1 Be7 and now the standard sacrifice 11 Bxe6!
fxe6 12 Nxe6 should be winning; e.g., 12...Qd7 13 Nxg7+ Kf7 14 Ne6 Na6
15 Bxh6 +–.

7 Nf4

Now we are back in the Flohr Variation (where the knight reaches f4 via
h3). Instead 7 h4 h6 8 Nf4 leads to Lesson 8.

7...Bd6 (D)
 White to play
8 h4

This is the main move. Instead, 8 Bc4 Nf6 transposes to note ‘b’ to White’s
8th move in 7D. White’s rarer moves, including 8 Nxg6, 8 Ngh5 and 8
Nfh5, are covered in 7E, and the main alternative, 8 c3, in 7F.

8...Qc7 9 Nxg6

White has also tried 9 h5?! Bxc2!, but Black has the better of the
forcing lines:

a) 10 Qg4? Nd7! –/+; e.g., 11 Qxg7? 0-0-0 12 Qxh8 Qa5+ 13 Bd2 Bb4 –+.

b) 10 Qxc2?! Bxf4 11 Bxf4 (11 Ne4?! Nd7 12 Bxf4 Qxf4 13 g3 Qc7 14 0-0-0
Ngf6 15 h6 g6 16 Kb1 Nxe4 17 Qxe4 0-0-0 –/+ Bellaiche-Machlik, European
Under-18 Girls Ch, Poreč 2015) 11...Qxf4 12 Qb3 Nd7 13 Qxb7 Rb8 14
Qxc6 and in Gheng-H.Messner, Böblingen 2003, Black should have
simply played 14...Qxd4; e.g., 15 Rd1 Qb4+ 16 Rd2 Nf6 =+.

c) 10 Nxe6 fxe6 11 Qxc2 Bxg3 12 fxg3 Qxg3+ 13 Qf2 Qxf2+ 14 Kxf2 Ne7
=+ M.Perez-Labollita, Argentine Team Ch, Villa Martelli 2014.

9...hxg6 10 Ne4

Or 10 Qf3 Nf6 11 Ne4 Nxe4 12 Qxe4 Nd7 13 g3 0-0-0 14 Be3 Nf6 15 Qf3
Nd5 16 Bd2 Nf6 ½-½ Pierrot-Peralta, Argentine Ch, Buenos Aires 2010.

10...Nf6!?

If you cannot make yourself give up the full bishop-pair, 10...Bf4 11 Bxf4
Qxf4 is fully playable. After 12 Ng5 Nf6 13 g3 Qd6 14 c3 Qd5 15 Rg1 Nbd7
16 Bg2 White had somewhat the better chances in Duda-Navara, Wroclaw
rapid 2014 but with White having lost kingside castling rights, Black will
always have counterplay.

11 Nxd6+ Qxd6 (D)

White to play

When I first considered this position, my inner voice protested loudly: “It’s
the bishop-pair against two knights – forget it!” However, for the moment
the bishops aren’t doing much, Black has a very clear lead in development
and
that h-pawn on the half-open file must after all be a positional deficit.

12 Be2

12 g3 Qd5 should be fine for Black. The critical line may be 12 c3 Nbd7
13 Bg5, which is discussed in 7G.

12...Nbd7 13 g3

13 c3 should probably be met by 13...e5! (13...0-0-0 14 Qa4 e5!? seems

unnecessarily complicated), when 14 Be3 Nd5 15 dxe5 Qxe5 16 Bd4 Qe4
is fine for Black: 17 Qd3 (17 Bxg7 Rh7 18 Bd4 0-0-0 19 Qd3 Qe7 20 Qd2
Re8
21 Kf1 Rxh4 =) 17...Qxd3 18 Bxd3 c5 19 Be4 cxd4 20 Bxd5 dxc3 21
bxc3 Rc8 =.

13...e5

13...0-0-0 may be playable, but after 14 0-0, Black should certainly

avoid 14...Nh5?! 15 Kg2 Ndf6 16 c3 Ne4? 17 Qe1! f5?! (17...Qd5 18
Bf3 f5 19
Bg5) 18 b4 e5 19 Bxh5 Rxh5 20 f3 +– T.Vos-J.van Foreest, Vlissingen 2016.

14 dxe5 Qxe5 15 0-0 (D)

 Black to play

White’s kingside looks a little airy but he has the bishop-pair to cover his
weaknesses and I struggle to understand how it can go downhill so quickly
as it did in this game, even with a 400 rating-point difference.

15...0-0

Instinctively, 15...0-0-0?! seems more natural, seeking to make use of

the half-open h-file, but this presents White with a target, and having
seen the course of the game I tend to think Turov made the right
decision!

16 c3 Rfe8 17 Re1 Qf5 18 Bg5 Rad8

Chances are equal.

19 Qc1

If White wants to play Qf4, offering a queen exchange, 19 Qd4 Nb6 20

Qf4 is more accurate but 20...Qc2 21 Bxf6 Nd5 is OK for Black too; e.g.,
22 Qc1 Qxc1 23 Raxc1 gxf6 =.
19...Re4 (D)

This prevents White from offering the queen exchange but 19...Ne4 may
be more precise, and leads to an edge for Black. One of White’s problems
is that he will be worse if he has to give up his light-squared bishop for a
knight.

White to play
20 Bf1?

What could be more natural than seeking exchanges while moving the
bishop over to protect the light-square weaknesses on the kingside?
Actually it seems that White has to play fairly accurately to survive. 20
Qc2?! Rde8 21 f3 is possible but the kingside isn’t any safer and after
21...Qc5+ Black still has the edge and a much easier position to play.
Stockfish offers the tactical solution 20 Bd3! Rxe1+ 21 Qxe1 Qxd3 22 Rd1
Qf3 23 Bxf6 gxf6 24 Rxd7 Rxd7 25 Qe8+ Kg7 26 Qxd7 with equality. But
although not impossible to calculate it’s easy to miss. Except for that, the
only path to equality seems to be 20 Qd2 Rde8 21 Bxf6 gxf6 22 Kg2 =.

20...Ne5
Now Black’s advantage is very clear.

21 Rxe4?

The line 21 Bg2 Nd3 22 Qd2 Nxe1 23 Qxd8+ Re8 24 Qd2 Nxg2
illustrates how a black knight is stronger than White’s dark-squared
bishop: 25 Bxf6 Qf3 26 Be7 Qe4 27 Qd4 Qxd4 28 cxd4 Rxe7 29 Kxg2
Re2 –/+ or 25 Kxg2 Qe4+ 26 Kg1 Ng4 27 Bf4 Qe2 –/+.

21...Nxe4

21...Nf3+! 22 Kh1 Qxe4 23 Bg2 Ng4! is even more effective.

22 Be3?! (D)

22 Qe3 is somewhat more stubborn but 22...Rd2 23 Bg2 Nxg5 24 hxg5

Rxb2 should be winning for Black.

Black to play
22...Rd2!
Black finds the thematic move, notes that it is winning, and plays it.
The computer of course points out that there are even more efficient
ways to finish off White, including 22...Nxf2! 23 Bxf2 Rd2!.

23 Bg2 0-1

It seems White resigned while waiting for Black’s reply, quite

understandably as the calm 23...Re2 is just as lethal as the more
forcing 23...Nf3+ 24 Bxf3 Qxf3 25 Qf1, when Black again may choose
between a number of quiet moves and the powerful 25...Nxg3!.

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 (6 Nh3 e6 7 Nf4 Bd6 8

h4 Qc7 9 Nxg6 hxg6 10 Ne4 Nf6!? 11 Nxd6+ Qxd6 12 c3)

7A: 6 Nf3 e6 (7 Bd3)

7B: 6 Nf3 e6 7 Ne5

7C: 6 f4

7D: 6 Bc4

7E: (8 Nxg6)

7F: 8 c3

7G: 12 c3

This collection of magnifiers is a mixed bunch. If you tend to meet

well- prepared opponents, you should definitely take a close look at
7G. It will deepen your understanding of the main line.

7A and 7B consider lines that were popular when the Capablanca

Variation was young. There is little reason to commit anything more than
the absolute main lines to memory, but knowing the historical context
may add to your
appreciation of why White advances his h-pawn in the modern main lines.

7D, 7E and 7F were included mainly for completeness and may safely
wait until you want to look up a move after facing it over the board. In
contrast, I would make sure to remember at least up to move 8 in 7C – it
may save you from handling a very difficult practical task over the board.

7A: 6 Nf3 e6: Rare 7th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6)

6 Nf3 e6 (D)

White to play

This is the Bareev move against this move-order. Now 7 h4 h6 takes us

to Lesson 9.

7 Bd3

Other possibilities include:

a) 7 Bf4!? hasn’t been seriously tested but may be seen as preparation for
Ne5 followed by h4. Black shouldn’t become too optimistic as he still
needs to equalize. 7...Qb6!? 8 Ne5 Qxb2 9 Rb1 Qc3+ 10 Qd2 Qxd2+ 11
Kxd2 isn’t clear at all, but 7...Nd7 is probably the simplest solution. After 8
Bd3 Ngf6 9 Bxg6 hxg6 10 0-0 c5 11 dxc5 Nxc5 12 Qe2 Be7 Black seems
very close to equality in an interesting position.

b) With 7 Bc4 White risks his bishop being slightly out of play after
7...Nd7 8 0-0 Ngf6:

b1) After 9 Ng5 h6 10 Nh3 Bd6 11 Nf4 (Tal-Petrosian, USSR Team Ch,
Moscow 1961) Black’s most promising path is probably 11...Bh7 12 Re1
(12 Bxe6?! fxe6 13 Nxe6 Qa5 14 Nxg7+ Kf7 –/+) 12...Qb8, when the
aggressive 13 Nxe6?! fxe6 14 Bxe6 Qc7 =+ offers White little.

b2) 9 Re1 Be7 10 c3 Qc7 11 Nh4 (11 Qe2 Nd5 12 Bd3 Bxd3 13 Qxd3 0-0 14
Bd2 h6 15 Rad1 Rad8 16 Nh5 N7f6 17 Nxf6+ Nxf6 18 Ne5 Nd7 19 Qg3
Nxe5 20 dxe5 Kh8 = I.Marshall-W.Lautenbach, corr. 2011) and now
rather than the inaccurate 11...Nd5?! 12 Nxg6 hxg6 13 Qf3 N7f6 14 Bb3
b5?! 15 Bg5 Nh5 16 Bxe7 Nxg3? (Gallagher-B.Vučković, European Ch,
Budva 2009) 17 Bc5! Ne2+ 18 Kf1 Nef4 19 g3 +–, Black should play
11...Bd6!?, when after 12 Qf3 0-0 Black intends ...c5 or ...b5 and the
knight on g3 is poorly placed.

We now return to 7 Bd3 (D):

 Black to play

This move is extremely solid. Black will be able to complete his

development and castle kingside in comfort, but don’t assume that he will
easily obtain complete equality.

7...Bxd3 8 Qxd3 Nf6 9 0-0

9 Bg5 Nbd7 10 0-0 Bd6 transposes.

9...Bd6 10 Bg5

White has a little more space and a small lead in development but Black
just needs a few more moves to castle and free his position with ...c5.
Other sensible moves include:

a) 10 b3 Nbd7 11 c4 0-0 12 Bb2 Qc7 13 Ne4 Nxe4 14 Qxe4 Nf6 ½-½

Kengis-Podzielny, Bonn 1995.

b) 10 Ne4 Nxe4 11 Qxe4 and now Black must avoid 11...0-0??, when 12
Ng5! g6 13 Qh4 h5 14 g4! suddenly gives White a decisive attack.
Playing 11...Nd7 before castling is safe and natural, and leaves Black
with few
problems.

10...Nbd7 (D)

White to play
Now:

a) The fact that 11 Ne4 Be7 12 Bf4 leads to a position that could also
arise from the move-order 9...Be7 10 Bf4 Nbd7 11 Ne4 isn’t really
troubling as Black is quite comfortable. He should castle and free his
position with a quick ...c5.

b) 11 Rfe1 is more demanding:

b1) 11...0-0 may be slightly premature: 12 c4 Qc7 13 Ne4 Nxe4 14 Rxe4

and now Black should avoid 14...c5?! (A.Muzychuk-E.Atalik, Women’s
Bundesliga 2017/18) 15 b4! +/–.

b2) After 11...h6 12 Bd2 Black should sidestep 12...Qc7?! 13 Nf5, but
12...0- 0 13 Ne4 Nxe4 14 Rxe4 Qf6 15 Rh4 Rfe8, as in G.Joo-J.Nemeth,
Hungarian Team Ch 2006/7, seems equal.
7B: 6 Nf3 e6 7 Ne5

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nf3 e6)

7 Ne5 (D)

Black to play

White invests some time to gain the bishop-pair.

7...Nd7

The knight should be confronted before White can create additional

threats with his h-pawn.

8 Nxg6

After 8 f4?! Nxe5 9 fxe5 Qd5 Black was at least equal in Mieses-
Tartakower, Teplitz-Schönau 1922.

8...hxg6 9 Ne4
Or:

a) 9 Be3 Ngf6 10 Bc4 Nd5 =+.

b) 9 Bf4?! Ngf6 10 c4 Qa5+ 11 Bd2 Bb4 12 a3 Bxd2+ 13 Qxd2 Qxd2+

14 Kxd2 c5 15 dxc5 Nxc5 16 Re1 =+ C.Yip-Ippolito, Internet blitz 2020.

c) Against the slow 9 c3, the thematic 9...Qh4!? seems to work perfectly;
e.g., 10 Qb3 0-0-0 11 Be3 Bd6 12 Bd3 Ngf6 13 0-0-0 Bf4 and in
E.Pogorelow- Baskin, Neustadt an der Weinstrasse 2018 Black would have
been better even if White had avoided 14 Kb1?! Bxe3 15 fxe3 Ng4 –/+.

d) Against 9 Bd3, the immediate 9...c5!? seems logical but in the absence
of practical tests I instead offer 9...Ngf6 10 Qe2 c5! 11 dxc5 Bxc5 12 Bd2
Qb6 with excellent chances for Black, Borriss-Tischbierek, Bundesliga
1996/7.

9...Ngf6 (D)

White to play

Black is already at least equal. His main plan is to castle queenside and
open the position with either ...c5 or ...e5.
10 Bd3

Or:

a) 10 Ng5?! c5! 11 c3 Qb6 12 dxc5? Bxc5 13 Qe2 Rd8 14 h3 0-0 15 g3

(A.Vovk-Margvelashvili, Lvov 2008) and now 15...Ne5! wins thanks to the
double threats to d3 and f3 and the line 16 Qxe5 Bxf2+ 17 Ke2 Nh5 18
Ne4 Rd5 –+.

b) 10 Nxf6+ gxf6 11 c3 and now 11...Qc7 12 g3 0-0-0 13 Bg2 f5 14 0-0

Bh6 15 b4 was very double-edged, with a race between two fairly
promising attacks, in Panarin-I.Popov, St Petersburg rapid 2015. But why
encourage White’s standard fianchetto set-up? 11...Qa5, planning to meet
12 g3?! (and possibly some other moves too) with 12...Qd5! looks
interesting. Maybe White’s best reply is 12 Bc4, but then I like 12...0-0-0
as 13 b4?? is premature because of 13...Bxb4!.

10...Nxe4 11 Bxe4 (D)

Black to play

Despite his bishop-pair, White must show great care not to be worse thanks
to Black’s half-open h-file and active pieces:

a) 11...e5 12 dxe5 Qa5+ 13 Bd2 Qxe5 14 Qe2 0-0-0 15 0-0-0 Ba3 16 c3 Nc5
17 Rhe1 Nxe4 18 Qxe4 Qxe4 19 Rxe4 Bc5 = Purdy-L.Steiner, Sydney

1945. b) 11...f5!? 12 Bd3 Qf6 13 Be3 e5 14 dxe5 Nxe5 =.

7C: 6 f4

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6)

6 f4

This move has some surprise value, and can be quite dangerous if Black
isn’t up to the task and allows the f-pawn to advance further up the board.
Downsides include a certain weakening of e4 and the blocking of
White’s dark-squared bishop. When preparing for it with Black you
should bear possible transpositions to the line 6 h4 h6 7 f4 (8B) in
mind.

6...e6 (D)

This is the correct move-order as the knight belongs on e7 (preventing f5).

 White to play
7 Nf3

For 7 h4 h6, see 8B. Alternatives include:

a) 7 Bd3!? actually could be one of the better gambits of the d4-pawn:

a1) 7...Qxd4 8 f5 looks a little dangerous; e.g., 8...exf5 (8...Qe5+!?) 9

Qe2+ Be7 10 Nf3 Qd6?! 11 0-0 and it is hard to evacuate the black king.

a2) 7...Bd6 8 Nh3 (8 Nf3 Qc7! 9 Ne5 c5 looks fine for Black) 8...Ne7 9 0-
0 Nd7 10 Nf2 Nf6 and Black should be at least equal.

b) Against the somewhat mysterious 7 Nh3!?, I like the equally mysterious

7...Nh6!?, keeping the path open for the dark-squared bishop while
watching the key square f5.

c) 7 Bc4 Bd6 8 N1e2?! Ne7 (8...Qc7!?) 9 0-0 Nd7 10 Bd3 Qc7 and now:

c1) After 11 c4, 11...c5?! is unnecessarily risky as it invites White to

blast open lines to his benefit by 12 b4! (Santagati-Atalik, Agrigento
2014).
Instead 11...Rd8! is an excellent move, after which White is hard-pressed
to demonstrate equality. 12 c5?? is obviously impossible, while 12 f5 (12
Kh1 Nf6 =+) 12...Nxf5 13 Bxf5 Bxf5 14 Nxf5 exf5! (keeping ...Bxh2+ in
reserve) is at least OK for Black.

c2) 11 Ne4 Bxe4 12 Bxe4 Nf6 13 Bf3 c5 14 c3 0-0 =+ Savitsky-Ovechkin,

Khanty-Mansiisk 2007.

We now return to 7 Nf3 (D):

Black to play
7...Bd6

The knight is going to e7, so it’s important to get the bishop out first. As a
matter of fact, 7...Nd7?! is Black’s most common choice but after 8 Bd3
Bxd3 9 Qxd3 Ngf6?! 10 f5 (or 10 0-0, planning f5) Black is fighting an uphill
battle (and scores miserably).

8 Bd3

For 8 h4!? h6, see 8B. Alternatives include:

a) 8 Ne5 Bxe5 9 fxe5 c5 and Black is at least equal.

b) 8 Be2 Ne7 9 Ne5 Bxe5 10 fxe5 c5 11 dxc5 Qxd1+ 12 Bxd1 Na6 13 0-0
Nxc5 14 Bf3 Nc6 15 Bxc6+ bxc6 16 b3 Nd7 17 c4 a5 = J.Polgar-
Khalifman, FIDE Knockout, Las Vegas 1999.

c) 8 Bc4 Ne7 9 c3 (9 0-0 Nd7 10 c3 Nf6 11 Ne5 Qc7 12 Qf3 0-0 =) 9...Nd7
and now:

c1) 10 h4?! h6 11 h5 Bh7 12 Bd3 Bxd3 13 Qxd3 Qc7 14 Ne5 Rd8 15 Qe2 c5
–/+ Papastavropoulos-Prohaszka, Greek Team Ch, Achaea 2012.

c2) 10 Ne5 looks better, but 10...Nf6 11 h4 h6 12 Nxg6 Nxg6

demonstrates another reason for Black to put his knight on e7. Probably
chances after 13 Nh5 0-0 are roughly equal, but White must be careful
with his king in the centre as ...c5 will not be hard for Black to arrange.

8...Ne7 (D)

White to play
Now:
a) 9 c3 Nd7 10 Ne5 Bxe5 11 fxe5? Nxe5 12 Bxg6 N5xg6 13 Qf3 Qd5 –
/+ M.Lewitt-Caro, Berlin 1905.

b) 9 Qe2 Nd7 10 Bd2 Bxd3 11 Qxd3 Qc7 12 Ne5?! (12 0-0 =) 12...Rd8 13
Qe2 c5 =+ Kofidis-Atalik, Panormo Zonal 1998.

c) 9 0-0 Nd7 and now:

c1) 10 Kh1 Qc7 11 Ne5 Rd8 12 Qe2 Bxd3 13 Nxd3 0-0 14 Bd2?! c5 15 Ne4
Nf5 16 dxc5 Nxc5 17 Ndxc5 Bxc5 18 Bc3 Bd4 =+ Marshall-
Capablanca, New York 1927.

c2) 10 Ne5 0-0 11 Nxg6 hxg6 12 Ne4 Nf6 13 Qf3 Nxe4 14 Qxe4 Qc7 15
Bd2 Rad8 16 b4 b6 17 a4 Nf5 (17...c5 =+) 18 Bc3? (18 c3 c5 =+) 18...c5 –
/+
Pötsch-Gledura, Budapest 2013.

7D: 6 Bc4

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6)

6 Bc4

This seemingly active bishop development can easily transpose to

lines covered in Lesson 8.

6...e6 (D)
 White to play

With f4 still possible, we should keep the ...Bd6 and ...Ne7 option open.

7 N1e2

We already had a look at 7 Nf3 under the move-order 6 Nf3 e6 7 Bc4 in

note ‘b’ to White’s 7th move in 7A, so the following remain:

a) 7 f4 is harmless if Black takes standard precautions against the f5

advance with 7...Bd6 8 Nf3 Ne7:

a1) 9 0-0 0-0 10 Qe2 Qc7 11 Ne5 Nd7 12 Ne4 Bxe4 13 Qxe4 Nf6 14 Qe2 c5
15 dxc5 Bxc5+ =+ Berndtsson Kullberg-Ståhlberg, Gothenburg 1931.

a2) 9 c3 Qc7 10 0-0 Nd7 (10...Bxf4?! is too greedy: 11 Bxf4 Qxf4 12 Qe2)
11 Ne5 Nf6 12 Qe2 (12 Bb3 Rd8 13 Nc4 0-0 14 Nxd6 Qxd6 =+ B.Bui-
P.Heuser, Berlin 2017) 12...0-0 13 Be3 c5 (13...Nfd5 =+) 14 dxc5 Bxc5
15 Bxc5 Qxc5+ = Stoltz-Flohr, Bled 1931.

b) 7 Nh3 Nf6 8 0-0 (for 8 Nf4 see note ‘b’ to White’s 8th move) 8...Bd6 9
f4!? has been discussed in some fairly high-level games but they have
all
followed the same course up to 9...Qc7 10 Kh1 0-0 11 f5 exf5 12 Nxf5
Nbd7 13 Nxd6 Qxd6, when White’s bishop-pair doesn’t quite seem to
outweigh Black’s more active pieces and control of d5; e.g., 14 Bf4 (14 Nf4
Be4 doesn’t improve White’s chances) 14...Qb4 15 Bb3 and Black has
several moves giving at least equal chances; e.g., 15...Rfe8 (15...a5 has
also been tested and seems fine) 16 Ng5 (16 c3 Qb6 17 Bg3 a5 18 Nf4
Be4 doesn’t change the evaluation) 16...a5 17 c3 Qb6 18 Qd2 a4 19 Bc4
h6 20 Nf3 Ne4 and Black probably had an edge in Ponomariov-Vidit,
Spanish Team Ch, Linares 2017.

7...Nf6 (D)

White to play

There is little point delaying this move, as White can more or less force
a transposition to Lesson 8 anyway.

8 0-0

The only critical continuation is 8 h4 h6 9 Nf4 Bh7, leading to Lesson

8. Otherwise:
a) 8 c3 Nbd7 9 Nf4 (9 0-0 Bd6 10 Bf4 Nb6 11 Bb3 Qc7 = Manik-Szamos,
Presov 2001) 9...Bd6 10 Qf3 Qc7 11 Nxg6 hxg6 12 Bd2 c5 13 Bb3 0-0-0
14 0-0 Kb8 = Andriasian-Tukhaev, Kavala 2010.

b) 8 Nf4 Bd6 and now:

b1) 9 h4 Qc7 10 h5 (10 Nxg6 hxg6 11 Qf3 Nbd7 12 Bg5?! c5 13 dxc5 Qxc5
14 Bxf6 Nxf6 15 Bb3 0-0-0 –/+ Tiviakov-Dreev, Ubeda 1999) 10...Bxc2 11
Nxe6 Bxd1 12 Nxc7+ Bxc7 13 Kxd1 0-0 = Rozentalis-Shengelia, Austrian
Team Ch 2007/8.

b2) 9 Bb3 Nbd7 with a further split:

b21) 10 Qf3 a5 11 c3 a4 12 Nxg6 hxg6 13 Bc2 e5! 14 0-0 exd4 15 Rd1

(15 Re1+ Kf8 =+) 15...Qe7 16 cxd4 Nf8 =+.

b22) 10 h4 Qc7 11 Qf3 e5 12 Nxg6 hxg6 13 Ne4 exd4 =+

Makropoulou- Morchiashvili, Athens 2007.

b23) 10 0-0 Nd5 11 Ngh5 0-0 12 c4 Nxf4 13 Nxf4 Qc7 14 Qf3 (Guliyev-
Riemersma, Dutch Team Ch 2015/16) 14...c5 =+.

b3) 9 0-0 Qc7 10 Qf3 Nbd7 11 Bb3 e5 12 Nxg6 hxg6 13 c3 exd4 14

Re1+ Kf8 15 cxd4 Rh4 = V.Knox-M.Freeman, British Ch, Ayr 1978.

8...Bd6 9 f4

9 Nf4 Qc7 10 Nxg6 hxg6 11 a4 Nbd7 12 c3 Bf4 13 Qf3 Bxc1 14 Raxc1 0-0
15 Rfe1 c5 16 Bb5 ½-½ Movsesian-Leko, FIDE Knockout, Las Vegas
1999.

9...Qc7 10 f5

10 Bd3?! 0-0 11 Kh1 Nbd7 12 f5 exf5 13 Bxf5?! (Krivovitsin-Kurkin,

Lipetsk 2007) 13...Rae8 14 Bf4 Nd5 15 Bxd6 Qxd6 16 Qd2 Ne3 –/+.

10...exf5 11 Nxf5 (D)

11 Bf4 is playable but 11...Nbd7 equalizes comfortably.

 Black to play
11...Bxf5

Picking up the h2-pawn with check is actually too risky. That’s no

problem as simple play is good.

12 Rxf5 Nbd7

Black is at least OK:

a) 13 Bf4 g6 14 Bxd6 Qxd6 15 Rf3 0-0-0 16 Qc1 Rhe8 =+

Ravichandran- I.Popov, Hyderabad 2013.

b) 13 Ng3 0-0-0 14 Bxf7 (14 Qf1? h5 –/+ Fichtl-Golombek, Munich

Olympiad 1958) 14...Nc5!? 15 Qf3 (15 Bc4 Rhe8 16 Bg5 Bxg3 17
hxg3 Nce4 leaves Black at least equal as 18 Bf4? is met with 18...Qb6
–/+) 15...Qxf7 16 dxc5 Bxg3 17 hxg3 Qc4 (Fressinet-Kallai, French
Team Ch 2001) 18 c3 Rhe8 and Black’s activity compensates for the
pawn.

7E: Rare 8th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6)

6 Nh3 e6 7 Nf4 Bd6 (D)

White to play
8 Nxg6

This is a rather straightforward attempt to achieve a small plus.

In contrast, the two knight moves to h5 are complicated. They are

obviously similar but the differences may be quite confusing. 8...Kf8!?
seems fine against both but I suggest capturing the knight in both cases,
reducing your workload for two relatively rare lines:

a) 8 Nfh5 Bxh5 and now:

a1) For 9 Nxh5, see line ‘b2’.

a2) 9 Qxh5!? is an additional option. Black should refuse the pawn on

offer as 9...Bxg3?! 10 hxg3 Qxd4 11 Rh4! is too risky. Instead 9...Nf6 10
Qh4 Nbd7 11 Be2 Qc7 looks fine for Black:
a21) 12 0-0 may be a little premature as 12...h6 13 Nh5 Nxh5 14 Qxh5 Nf6
15 Qh3 0-0-0 is not easy to evaluate but Black’s kingside attack seems
easier to play.

a22) 12 c3 h6 13 Ne4 should be equal; e.g., 13...Nxe4 14 Qxe4 Nf6 15

Qf3 Nd5 =.

b) 8 Ngh5 Bxh5 and now:

b1) 9 Qxh5?! Bxf4 10 Bxf4 Qxd4 11 Be5 Qe4+ 12 Qe2 (12 Be2? Nd7 13
Bxg7 Qxg2 14 Bxh8 Qxh1+ –/+) 12...Qxe2+ 13 Bxe2 Nf6 and White
might just have enough for the pawn.

b2) 9 Nxh5 g6 10 Ng3 Nf6 (a simpler solution than 10...Qh4!?) 11 c3

Nbd7 12 Bc4 Qc7 13 Qf3 0-0 14 Bg5 Kg7 15 0-0 h6 and now:

b21) 16 Be3 c5 (16...Nd5 also looks promising) 17 Bb3?! (Stockfish

thinks 17 Rac1 is equal) and in Andriasian-Navara, European Ch,
Erevan 2014, 17...c4 probably would have led to a plus for Black.

b22) 16 Bd2 should be sufficient for equality; e.g., 16...Nd5 17 Rfe1 Bf4
18 Bxf4 Nxf4 =.

8...hxg6 9 Ne4 Be7 (D)

I don’t think Black should allow two bishops vs two knights positions when
there isn’t an exposed pawn on h4.
 White to play

While White’s bishop-pair theoretically might give him a small plus, it’s hard
to demonstrate analytically.

10 g3

Or 10 c3 Nf6, and now:

a) 11 Ng5 Nbd7 12 g3 e5 13 Nf3 exd4 14 Nxd4 Ne5 15 Bg2 (15 Bf4?! Qd5
=+ Collutiis-Dreev, Cento 2011) 15...c5 16 Nf3 Qxd1+ 17 Kxd1 Nd3 =.

b) 11 Bd3 Nxe4 12 Bxe4 Nd7 and

now: b1) For 13 g3, see the main line.

b2) 13 Qe2 Nf6 14 Bf3 (Vorobiov-Gavrilov, Pardubice 2001) 14...Qa5!?

has the point that 15 0-0 Bd6 16 g3 Qf5 17 Bg2 g5 offers Black promising
kingside counterplay.

10...Nf6 11 Bg2 Nxe4 12 Bxe4 Nd7 13 c3 Nf6

Now:

a) 14 Bg2?! (Godena-Borgo, Italian Ch, Saint Vincent 1998)

14...e5! equalizes easily; e.g., 15 dxe5 Qxd1+ 16 Kxd1 Ng4 =.

b) After 14 Bf3! I suggest 14...Qa5!? with counterplay:

b1) 15 0-0 g5! offers Black enough play on the h-file for him to view
the future with confidence.

b2) 15 Bd2 g5 16 h3 Qf5 looks OK for Black after both 17 Qe2 Rxh3 18
Rxh3 Qxh3 19 Bxg5 and 17 Bg2 g4 18 hxg4 Rxh1+ 19 Bxh1 Nxg4,
though White’s bishop-pair should not be underestimated.

b3) 15 h4 0-0 (or 15...e5!?); e.g., 16 Qb3?! (as this doesn’t prevent the
freeing break, 16 Qe2 or 16 Kf1 intending Kg2 is preferable) 16...c5 17
dxc5 Bxc5 18 0-0 e5 and Black should not be worse as taking on b7 with
the queen doesn’t promise White anything.

7F: 8 c3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nh3 e6 7 Nf4 Bd6)

8 c3 (D)
 Black to play

This move looks too sedate to create any real problems for Black.
However, it’s flexible and generally useful as White’s queen is now free to
move without leaving the c-pawn en prise.

8...Qh4

8...Nf6 is a playable alternative but after 9 h4 full equality is still

rather elusive:

a) 9...Bxf4 10 Bxf4 h6 11 a4 0-0 12 a5 Nd5 13 Bd2 Nd7 14 c4 N5f6 15 h5

Bh7 16 Bc3 c5 17 Be2 Qc7 18 0-0 += Ivanchuk-Artemiev, Huaian rapid
2017.

b) 9...Qc7 10 Qf3 Nbd7 11 h5 Bc2 12 h6 g6 13 Bd3 Bxd3 14 Nxd3 0-0?! 15

0-0 Qa5 16 c4 Bxg3 17 Qxg3 Qf5 18 Re1 += V.Onishchuk-Nabaty, Riga
2017.

9 Ngh5

9 Qf3 Nf6 10 Nxg6 hxg6 11 Bd3 Nbd7 12 Ne4 Nxe4 13 Qxe4 Nf6 14 Qxh4
Rxh4 ½-½ Tukmakov-M.Vukić, Odessa tt 1975.

9...Bxh5 10 Qxh5 Qxh5 11 Nxh5 g6

Now:

a) 12 Ng3 Nf6 13 Bh6 Nbd7 14 f3 Nd5 15 Ne4 Bf4 16 Bxf4 Nxf4 17 c4 f5

18 Nc3?! e5 19 dxe5 Nxe5 20 0-0-0 Ke7 =+ Stević-Michalik, Vienna
2012.

b) 12 Nf4 Nd7 13 g3 Ngf6 14 Bg2 0-0 brings Black close to

equality, Šarakauskas-J.Richardson, Crawley 2016.

c) 12 Bf4 Bxf4 13 Nxf4 Nf6 14 Nd3 Nbd7 15 g3 0-0 16 Bg2 Rfc8 17 a4

a5 18 Kd2 Kf8 19 Rhb1 += J.Polgar-Anand, Haifa rapid 1998.

7G: 12 c3

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nh3 e6 7 Nf4 Bd6)

8 h4 Qc7 9 Nxg6 hxg6 10 Ne4 Nf6!? 11 Nxd6+ Qxd6 12 c3

This sensible move may be the critical test of Black’s compensation for
the bishop-pair.

12...Nbd7 (D)

Now 12...Qd5?! is relatively pointless and in Tomashevsky-Navara,

Huaian rapid 2016 White had a clear plus after 13 Be2 Qe4 14 Bg5.
 White to play
13 Bg5

This looks active and could be best. Alternatively:

a) Black is fine after 13 g3 Qd5.

b) 13 Be2 is also an important option:

b1) 13...0-0-0 14 g3 e5 (Malik-St.Williams, Birmingham 2016) 15 0-0

seems like a small plus for White.

b2) 13...e5 14 dxe5 (14 Be3 Nd5 15 dxe5 Qxe5 comes to the same thing)
14...Qxe5 15 Be3 Nd5 16 Bd4 Qe4 appears fine for Black; e.g., 17 Bxg7
Rh7 18 Bd4 0-0-0 19 Qd3 Qe6 20 Qd2 (20 0-0-0?? Nf4) 20...Re8 21 Kf1
c5 and
Black has compensation.

13...Qd5

Except for the obvious fact that Black should be active and develop
quickly, it’s not quite clear how he should play.
14 Be2

14 Qb3 Qe4+ 15 Kd2 Nc5! =.

14...Ne4!

I think this improves over 14...0-0-0?!, as in Külaots-

P.H.Nielsen, Copenhagen 1996.

15 Be3 g5 16 Qd3 (D)

Black to play

Here Black has a number of ways to get an interesting game. I’ll just
offer two sample lines after 16...f5!?:

a) 17 Bf3 gxh4 18 0-0-0 0-0-0 19 c4 (19 Kb1?! Ne5!) 19...Qa5 20 Kb1 Ndf6
and I would take the black pieces.

b) 17 0-0-0 Qxa2 18 hxg5 Rxh1 19 Rxh1 0-0-0 seems quite likely to end in
a draw by repetition but it appears that it’s White who has to be careful.
Lesson 8: Rare Seventh Moves

By: SJ

Chess books should be used as we use glasses: to assist the sight,

although some players make use of them as if they thought they
conferred sight.

Jose Raul Capablanca

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4

This move is obviously slightly weakening and makes kingside castling

somewhat less tempting. However, theory and GM practice indicate that
the kingside space and activity it generates are more important factors. It
should be noted that 6 Nf3 usually leads to the same position as White
mostly follows up with 7 h4. However, in this lesson we shall focus on
lines where the knight is directed towards f4.

6...h6 (D)

The bishop was in danger of being trapped (OK, ...Bf5 may be possible as
long as ...Qa5 is check but why then retreat the bishop at all?) and
6...h5?! is risky, as the pawn is exposed to attack.
 White to play
7 Nh3

This is the main alternative to 7 Nf3 but there is a small family of related
moves as 7 N1e2 and 7 Bc4 will usually transpose after 7...e6. For 7 f4,
see 8B. And for some really rare 7th moves, see 8A.

After 7 h5 Bh7 White usually plays 8 Nf3, as in Lesson 9. Other moves are
at best harmless:

a) 8 f4 e6 9 Nf3 Nd7 10 Bd3 Bxd3 11 Qxd3 Ngf6 12 Ne5 c5 13 Be3 Qa5+

14 c3 Rd8 15 0-0 cxd4 16 Bxd4 Bc5 =+ Timman-Hübner, Nijmegen rapid
1979.

b) 8 Bd3 is one of several similar pawn sacrifices:

b1) 8...Bxd3 9 Qxd3 e6 should transpose to Model Game 9 after 10 Nf3.

Instead after 10 Bd2 Nf6 11 0-0-0 Be7, 12 c4?! (Augusta-Draganić,
Croatian Team Ch, Zagreb 2009) 12...b5 leaves Black better. 12 f4 is a
better try, and roughly equal.

b2) 8...Qxd4!? is the principled answer. After 9 Be3 Qd8 10 Bxh7 Qxd1+ 11
Rxd1 Rxh7 White has some compensation thanks to his better
coordinated pieces, M.Martins-Starostits, Mondariz 2006.

7...Nf6

It is possible to delay this for a few moves with 7...e6, but trying to do
without ...Nf6 is rather pointless as 8 Nf4 Bh7 9 Bc4 Bd6? 10 Ngh5!
leaves Black in deep trouble; e.g., 10...g6 11 Ng7+ Kd7 12 Ngxe6! fxe6
13 Bxe6+! Kc7 14 Bh3.

8 Nf4 Bh7 9 Bc4 e6 (D)

White to play

This position is the main subject of this lesson. It may arise from a number
of move-orders, including some we discussed in Lesson 7 after 6 Bc4 or 6
N1e2. Against some of these move-orders Black may avoid the diagram
position, but my advice is to steer for it. This is partly for consistency
reasons but mainly because I believe in Black’s chances here. White has
developed actively and is committed to following up aggressively as
otherwise both the pawn on h4 and the bishop on c4 may prove misplaced.
However, it seems to me that Black has excellent winning chances in the
critical line.
 Model Game 8

Brendel – M. Dziuba

Warsaw rapid 2011

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nh3 (D)

7 N1e2 Nf6 8 Nf4 and 7 Bc4 Nf6 8 Nh3 e6 9 Nf4 Bh7 are two
important move-orders that lead to the game.

Black to play
7...Nf6

7...e6 8 Nf4 Bh7 9 Bc4 Nf6 is just an alternative move-order.

8 Nf4

Consistent; where else should the knight go? 8 Bc4 is rare but likely
to transpose after 8...e6.
8...Bh7 9 Bc4

White is more or less committed to this bishop development. The bishop

risks being misplaced here but preparing to exchange it on d3 gives Black
easy play:

a) 9 Be3 e6 10 Bd3 Bxd3 11 Qxd3 Nbd7 12 0-0-0 Qa5 13 Kb1 Bd6 14 c4 0-

0-0 15 d5 (Kupreichik-K.Berg, Nørresundby 1992) and now Black’s best
is 15...Nb6 16 Ne4 Be5 17 Nxf6 Bxf6; e.g., 18 d6 Be5 19 Qc2 Rxd6 20
Nd3
Nd7 21 c5 Rd5 22 Nxe5 Nxe5 23 Rxd5 cxd5 24 Bd4 =.

b) 9 c3 e6 10 Bd3 (for 10 Bc4 see 8C) 10...Bxd3 11 Qxd3 Nbd7 and now:

b1) 12 Nfh5 Qa5 13 Nxf6+ Nxf6 14 Ne4 Nxe4 ½-½ Simonian-Burmakin,

St Petersburg 2009.

b2) 12 Ne4 Nxe4 13 Qxe4 Nf6 ½-½ Ljubojević-Karpov, Malta Olympiad

1980.

9...e6 (D)

White to play
10 0-0

This looks ugly as the advanced h-pawn doesn’t fit well with kingside
castling. But White is playing for a blitz attack and assumes that his own
king’s safety will not be an issue. We shall try to prove that assumption
wrong in 8E. White’s main alternatives are 10 Qe2 and 10 c3, which both
are covered in 8C.

10...Bd6 11 Nxe6?!

This is the consistent follow-up even if it doesn’t score very well and moves
the computer evaluation from equality to a small plus for Black. White
should avoid 11 Bxe6? Bxf4, which wins for Black, and 11 Qe2 0-0, which
is very comfortable for Black. In 8D we’ll take a closer look at 11 Re1 and
11 Ngh5, which both may come close to equalizing for White (and
consequently are objectively better).

11...fxe6 12 Bxe6 Na6!?

The knight is heading for d5. Defending is always harder than attacking,
but White’s slow-burning attack allows Black some leeway, and this is
probably at least as good as the more tested alternative 12...Nbd7 (and
certainly better than 12...Qc7?! 13 Nh5!).

13 Re1 (D)

Other moves are harmless at best:

a) 13 Qf3? Nc7 14 Bb3 Qe7 15 Nf5 Bxf5 16 Qxf5 Qd7 17 Qg6+ Kd8 –/+.

b) 13 Nh5? Kf8 14 Qf3 (14 Re1 Nc7 –+) 14...Be4 15 Qe2 Nc7 16 Bb3

Bf5
17 Nf4 Qd7 18 Bd2 Re8 –+ S.Das-Lalith, Commonwealth Ch, Nagpur 2008.
 Black to play
13...Kf8?!

It seems very natural to get out of all discovered checks. However, the
really interesting move here (and the reason I recommend 12...Na6) is the
cold- blooded 13...Nc7!, forcing White to cash in his investment
immediately.
Black’s point is that after 14 Bc8+ Kf7 15 Bxb7 Qd7! 16 Bxa8 Rxa8, Black
hasn’t only eliminated any trace of a white attack, but he is also ready to
launch a very dangerous (if slow-burning) counter-attack himself. Check
8E to find my analysis of this new move.

14 c3

14 Qf3 is probably not an improvement:

a) 14...Nc7 15 Bb3 a5! (after 15...Bg6?!, as in J.Christiansen-Melkumyan,

Riyadh blitz 2017, 16 Bd2 gives White interesting compensation) 16 a4
Bxg3 (16...Bg8!? may improve) 17 Qxg3 Ncd5 seems a safe option. If I
didn’t think that 13...Nc7! is the critical line, I would have focused my
analysis on lines like 18 Bd2 Kf7 19 c4 Ne7 20 Bc3 Re8 with interesting
play.
b) 14...Bxc2!? looks risky, but 15 Nh5 Bg6! 16 Nf4 Be8! is probably fine for
Black; e.g., 17 d5 cxd5 (17...Nc7!?) 18 Bd2!? (18 Be3?! Nc7 –/+)
18...Bxf4! 19 Qxf4 Bf7 and Black clearly isn’t worse.

14...Nc7

14...Bxg3 15 fxg3 Nc7 16 Bh3 Bg6 =+.

15 Bb3?

15 Bh3 Bxg3 16 fxg3 Bg6 =+.

15...Qd7

15...a5 is also good.

16 Qf3 Bg6 17 Bf4 Bxf4 18 Qxf4 Nd5 (D)

White to play
Black has a very clear
plus.
19 Bxd5 cxd5 20 Re5 Kg8 21 Rae1 Re8 22 h5 Nxh5 23 Nxh5 Bxh5

Black is winning.

24 Qh4 Rxe5 25 Rxe5 Bg6 26 Re7 Qd8 27 g3 Kf8?

A poor liquidation. 27...Qb6 keeps a huge advantage.

28 Rxb7 Qxh4?! 29 gxh4 a5 30 Ra7 a4 31 Rxa4 Ke7 32 Ra6?! Be4 33 a4

g5 34 hxg5?? hxg5 35 Kf1 Bd3+ 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 (7 Nh3 e6 8 Nf4

Bh7 9 Bc4 Nf6 10 0-0 Bd6)

8A: (7 Bd3!?)

8B: 7 f4

8C: (10 c3)

8D: (11 Re1)

8E: 11 Nxe6?! fxe6 12 Bxe6 Na6 13 Re1 Nc7!

Here we have another mixed collection. In 8B it may be worthwhile to

memorize the moves up to 12...Qc7! and have a look at some of the
practical examples.

The lines in 8A, 8C and 8D shouldn’t be high on your list of priorities. You
will come a long way with common sense and some calculation. However,
I would recommend spending some time on 8E. The line is very sharp
and a mistake could cost dear.

8A: Rare 7th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6)

7 Bd3!?

This may be among White’s better versions of this pawn sacrifice.

Some other rare moves include:

a) For 7 N1e2 e6 8 Nf4, see 8C-8E.

b) For 7 Bc4 e6 8 N1e2 (or 8 Nh3 Nf6 9 Nf4) 8...Nf6 9 Nf4, see Model
Game 8.

c) 7 c3 protects d4 and makes sense only as an alternative preparation for

Bd3, so after 7...Nf6 8 Bd3 Bxd3 9 Qxd3 e6, 10 N1e2 seems consistent.
Then 10...Bd6 11 Bf4 Qc7 12 Bxd6 Qxd6 13 Ne4 Nxe4 14 Qxe4 Nd7 was
equal in Banas-Ferry, French Ch, Paris 1963 but preparation for ...c5 with
10...Be7,
...0-0 and ...Nbd7 seems more

logical. We now return to 7 Bd3!?

(D):
Black to play
7...Bxd3

Here 7...Qxd4 8 Nf3 Qd6 9 Bxg6 Qxg6 10 Qe2 seems to give White
roughly sufficient play for his pawn, but if you enjoy being a pawn up in a
complex position, this is probably fine. After 10...Nd7, White didn’t get
enough for his pawn in R.Marić-Susić, Yugoslavia 1966, which continued
11 h5 Qd6 12 Rh4 e6 13 Rd4 Qc7 14 Bf4 Qa5+ 15 Bd2 Qb5 =+.
Nevertheless, White’s rook activation is quite thematic and worth
remembering as an option in slightly different versions. Instead, 11 Bf4 e6
12 0-0-0 Ngf6 13 h5 Qh7 14 Ne5 seems to give roughly sufficient
compensation.

8 Qxd3 e6!

The natural 8...Nf6 surely must be playable but 9 f4! demonstrates

White’s secondary idea as the king’s knight probably belongs on e7
against f4 systems (as we know from 7C).

9 Bf4 (D)

9 Nf3 transposes to 9A, and after 9 f4 Nd7 10 f5 e5 Black has his share of
the play.
 Black to play
9...Nf6

This looks sensible now that f4 isn’t on the agenda. After 9...Qa5+ 10
Bd2 Qc7, 11 Nf3 Nf6 again leads to 9A, but maybe White should try 11
Ne4!?, when Black isn’t ready to exchange the knight.

10 0-0-0 Bd6

The patient 10...Nbd7 11 Nf3 Be7 should also be OK.

11 N1e2 Qc7

Black seems close to equality:

a) 12 Be5 Nbd7 13 f4 0-0-0 14 Qf3 +=.

b) 12 Qe3 Bxf4 13 Nxf4 Nbd7 14 Rhe1 0-0-0 (14...0-0!?) 15 Nd3 Kb8 16

Kb1 Rc8 17 Ne5 c5 18 Nxd7+ Nxd7 19 d5 exd5 20 Rxd5 Nf6 21 Rdd1 Rhe8
= Edighoffer-A.Braun, corr. 2009.

8B: 7 f4

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6)

7 f4

The combined advance of White’s d-, f- and h-pawns looks ugly but this
line actually has some bite.

7...e6 (D)
 White to play

This position may also arise from the move-order 6 f4 e6 7 h4 h6.

8 Nf3

8 h5 Bh7 9 Nf3 Bd6 leads to the same position.

8...Bd6!?

White probably shouldn’t be allowed to gain time by attacking our

bishop with Ne5. More common is 8...Nd7, which seems an excellent
move but doesn’t fit with my recommendation in 7C.

9 h5

This is clearly the most common move.

9 Ne5 may appear consistent, but after 9...Bxe5 10 fxe5 Ne7 11 h5 Bh7
12 c3, Bologan’s suggestion 12...0-0 (instead 12...c5?! 13 Qg4! Rg8 has
been played but is rather difficult to handle) 13 Qg4 Kh8 14 Ne4 Nd7 15
Be2 c5 seems at least equal for Black.
9...Bh7 10 Bd3 Bxd3 11 Qxd3 Nf6 (D)

White to play
12 Bd2

12 Qe2 Qc7 13 Ne5 c5 14 dxc5 Qxc5 (14...Bxc5 may be better as Black is

winning after 15 Nf5?? exf5 16 Ng6+ Ne4 17 Nxh8 Bf2+ 18 Kf1 Nc6 –+) 15
Nf5! 0-0 16 Be3 Qc7 17 Nxd6 Qxd6 18 g4 Qb4+ (after 18...Nc6 chances
seem balanced) 19 c3 Qe4 20 0-0-0 Nd5?! 21 Rhe1 += Marjanović-
Groszpeter, Zenica 1986.

12...Qc7! 13 Ne5 c5 14 0-0-0

14 c3 0-0 15 0-0 Nc6 16 Nxc6?! Qxc6 17 Rae1 cxd4 18 cxd4 Qd5 –

/+ T.Horvath-C.Klein, Germany tt 2013/14.

14...cxd4 15 Qxd4 Nc6 16 Nxc6 Qxc6

Now:

a) 17 Rhe1?! Bc5 (17...0-0, intending to meet 18 Ne4 with 18...Be7, looks at

least as strong) 18 Qd3 Rd8 19 Qe2 0-0 =+ Mamedyarov-Döttling,
French Team Ch 2007.

b) 17 Bb4 Bxb4 18 Qxb4 (Tomczak-Hera, Bundesliga 2014/15) 18...Rc8

19 Rd2 a5 20 Qxa5 0-0 21 Kb1 b5 =+.

8C: 10th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6)

7 Nh3 e6 8 Nf4 Bh7 9 Bc4 Nf6 (D)

White to play

10 c3

Or:

a) Unless you desperately need to win as Black, the knight moves to h5

are harmless: 10 Ngh5 (10 Nfh5 Nbd7 11 c3 Nxh5 12 Nxh5 transposes
to line ‘a2’) 10...Nbd7 and now:
a1) 11 Bb3 Nxh5 12 Nxh5 Nf6 13 c3 Nxh5 14 Qxh5 Bd6 15 Be3 Qf6 16 0-0-
0 and after Bologan’s recommended 16...a5, the outcome is open.

a2) 11 c3 Nxh5 12 Nxh5 Nf6 13 Bf4 Nxh5 14 Qxh5 Bd6 15 Bxd6 (15 Be5
Bg6 16 Qe2 Bxe5 17 Qxe5 Qf6 18 h5 Bh7 19 f4 Rd8 20 g4 Qxe5+ 21 fxe5 =
Ponomariov-Karpov, Moscow blitz 2008) 15...Qxd6 16 0-0 0-0 17 Rfe1
b5 18 Bf1 b4 19 Qa5 bxc3 20 bxc3 Qd8 = Korneev-Riazantsev,
Portuguese Team Ch, Evora 2007.

b) After 10 Qe2?! I believe Black should bravely pick off the pawn with
10...Qxd4!?:

b1) 11 0-0 Nbd7 12 Rd1 Qe5 13 Be3 Bc5 –/+ Bologan.

b2) 11 Bxe6 fxe6 12 Nxe6 Qg4 13 Nc7++ Kf7 14 Qxg4 Nxg4 15 Nxa8 Bd6
16 0-0 Bxc2 –/+.

b3) 11 Nxe6 fxe6 12 Qxe6+ Kd8 13 Be3 looks critical, but Bologan’s
13...Bb4+! 14 c3 Bxc3+ 15 Ke2 Qg4+ 16 Qxg4 Nxg4 17 bxc3 Nxe3 18
Kxe3 Nd7 =+ is convincing.

10...Bd6 (D)
 White to play
11 Ngh5

11 Nfh5 looks less logical as that knight is already centralized. The main
point is 11...0-0 12 Bg5!? but after 12...Nbd7, White hasn’t achieved
much: 13 Nxf6+ (13 Qf3? hxg5 14 hxg5 Nd5 15 Qg4 Qe7 16 Qh4 f6 –/+
Bologan)
13...Nxf6 14 Qf3 Bxg3 15 Qxg3 Ne4 16 Bxd8 Nxg3 17 Rh3 Rfxd8 18
Rxg3 b5 19 Be2 b4 20 0-0-0 a5 and both players had chances in L.Milov-
Battaglini, Calvi 2010.

11...0-0 12 g4

White must keep the pressure up. Otherwise:

a) 12 0-0 Nbd7 13 Re1 Nxh5 14 Qxh5 (Selbes-Kobalia, Internet blitz 2020)

is too slow as Black can seize a dangerous initiative with 14...Qf6! 15 Ne2
b5 16 Bb3 c5 –/+.

b) 12 Rh3?! Nbd7 13 Rg3 Nxh5 14 Qxh5 Kh8 –/+ Djurić-Dražić,

Formia 1995.
c) 12 Be3 Nbd7 13 Nxf6+ Nxf6 14 Nh5 Re8 15 g3 b5 16 Nxf6+ Qxf6 17
Be2 Rad8 18 0-0 e5 =+ Bologan.

12...Bxf4 13 Nxf4 e5!

This logical central counter-strike is Bologan’s recommendation. He gives

the further moves 14 Ne2 Nxg4 15 f3 Nf6 16 dxe5 Qxd1+ 17 Kxd1 Nh5
with equality. The sharp 18 e6 fxe6 19 Nd4 doesn’t improve White’s
chances but the computer line 19...Rd8 20 Bd2 Nd7 21 Nxe6 Ne5 22 Bb3
is complicated. After 22...Kh8 23 Nxd8 Rxd8 Black has full compensation if
not more.

8D: 11 Re1 and Minor Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nh3 e6 8

Nf4 Bh7 9 Bc4 Nf6)

10 0-0 Bd6 (D)

White to play
11 Re1
Or:

a) 11 Qe2?! 0-0 12 c3 Nd5 13 Bxd5 exd5 =+ Tartakower-Flohr,

Folkestone Olympiad 1933.

b) 11 Ngh5 could be White’s best try for equality:

b1) Black should avoid 11...Nxh5?! 12 Nxh5 Qxh4 13 g3 Qh3 14 Re1 0-0
15 Bf4 as 15...Bxf4?? loses to 16 Nxf4 Qf5 17 Nxe6!.

b2) After 11...0-0 12 Re1, 12...Nxh5 13 Nxh5 Qxh4 14 g3 Qe7 is safe;

e.g., 15 Qg4 g5 16 Bf4 Bf5 17 Bxd6 Qxd6 18 Qf3 Nd7 =+ as given by
Bologan.

11...0-0

Now the king is secure, and Black can be happy with his position.

12 c3

Not 12 h5?! Qc7 –/+.

12...c5! (D)
 White to play

Black isn’t really behind in development and now he challenges the

white centre. It’s not clear that White can fully equalize:

a) 13 Be3?! cxd4 14 cxd4 is a poor version of an IQP position. After

14...Qc7 15 Qc1 Nc6 16 d5 e5, Black is better.

b) 13 Qf3?! Nc6 14 Be3 cxd4 15 cxd4 e5 16 Nfh5 Nxh5 17 Nxh5 Qxh4 –

/+ Castrillon Gomez-Swapnil, Barcelona 2015.

c) 13 Ngh5 Nxh5 14 Nxh5 Bf5 and now 15 g4?? loses to 15...Qxh4 –+.

d) 13 dxc5 Bxc5 is at least easier to play for Black: 14 Ngh5 (14 Bd3 Bxd3
15 Nxd3 Be7 16 Qe2 Nc6 =+ Korostenski-Ballero, corr. 2012) 14...Nxh5
15 Qxh5 Bb6 (15...Qc7 could be even better) 16 Qg4 Bf5 17 Qg3 Bc7 and
in E.Barr-B.Mulligan, corr. 2012, White was close to equality.

8E: 13...Nc7!

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nh3 e6 8 Nf4

Bh7 9 Bc4 Nf6 10 0-0 Bd6)

11 Nxe6?! fxe6 12 Bxe6 Na6 13 Re1 Nc7! (D)

We examined the less critical but also playable 13...Kf8?! in Model Game
8, along with mentioning deviations for both sides on moves 12 and 13.
 White to play

This can be seen as an attempt to force White’s hand immediately. If

White cashes in the material, I believe that Black will have a dangerous
attack for only the material equivalent of a pawn.

14 Bc8+

Regaining some material seems best.

Otherwise: a) 14 Bb3+?! Kd7 15 c4 Kc8 –/+.

b) 14 Nf5?! Kf8 15 Nxd6 Qxd6 16 Bb3 (after 16 Bh3 Re8, Black’s material
advantage should prove decisive) 16...Re8 and Black’s extra piece
outweighs the two pawns.

c) After 14 Bg8+?! Kf8 15 Bxh7 Rxh7 16 Nf5 Kg8 17 h5 Qd7 18 Qf3 Bf8!
Black has consolidated his plus, as 19 Nxh6+? Kh8 only helps him
activate his rook.

14...Kf7 15 Bxb7 Qd7! 16 Bxa8 Rxa8 (D)

 White to play

Am I right to assess that Black has much the better chances here?
Stockfish considers Black’s plus rather moderate, but it’s a better defender
than any human. After ...Kg8 and ...Rf8, all the black pieces will soon be
aiming at the white king and the weakening effect of White’s 6th move will
be very visible. Some sample lines:

a) 17 Nh5 Ncd5 18 c4 Bg6 19 Nxf6 Nxf6 doesn’t ease White’s

defence much; e.g., 20 b3 Bh5 21 Qd3 Kg8 –/+.

b) 17 c4 Kg8 18 Qb3 (18 Nh5 Nxh5 19 Qxh5 Bb4 –/+) 18...Kh8 19 Be3 Rf8
–/+.

c) 17 Qf3 Bxc2 18 Bf4 (18 Nh5?! Nd5 19 Nxf6 Nxf6 20 Bf4 Bxf4 21 Qxf4
Kg8 –/+) 18...Bxf4 19 Qxf4 Ne6 20 Qe5 Nxd4 =+.

d) 17 Bd2 Kg8 18 Qe2 Kh8 19 c3 Rf8 20 Rac1 Ncd5 21 Qe6 Qc7 and
now: 22 Nf5?? Bh2+ 23 Kf1 Ng4 –+, 22 b3 Ba3 23 Rcd1 Bc2 –/+ or 22
Qe2 Nf4 23 Bxf4 Bxf4 –/+.
Lesson 9: Bareev Introduction and Campora Hybrid 8 h5

By: SJ

Chess was Capablanca’s mother tongue.

Richard Réti

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4

This is the most popular way to reach the Capablanca main lines. Almost
as common is 6 Nf3, which transposes after 6...e6!? 7 h4 h6.

6...h6 7 Nf3 (D)

This move-order gives White a few extra options compared to the

immediate 7 h5.

Black to play
7...e6!?

This relatively new move is known as the Bareev Variation and is our key
move against the Classical Variation. Arguably it’s currently the hottest
topic in the Capablanca Variation. During large parts of the previous
century, 7...Nd7 was considered more or less mandatory in order to
prevent Ne5.
Black’s provocative idea is to develop his queen’s knight to c6 in some
lines and only go to d7 in order to oppose a white knight that has spent
two moves getting to e5. A related and equally challenging idea is 7...Nf6,
which is known as the Campora Variation. As we shall soon see, there is
even a fairly common transposition.

8 h5

This move indicates that White hopes to transpose back to the old
Capablanca main line. If there is a problem with Black avoiding 7...Nd7, it
can hardly be demonstrated by this move. Actually only 8 Ne5 (Lessons
10-
12) aspires to this. Now most games continue as follows:

8...Bh7 9 Bd3 Bxd3 10 Qxd3 Nf6

This active developing move is very natural and arguably one of the main
ideas behind the Bareev Variation. It should be noted that here play
merges with the Campora move-order 7...Nf6 8 h5 Bh7 9 Bd3 Bxd3 10
Qxd3 e6.
Black’s challenge is to prove that this really improves over 10...Nd7,
which would take us right back to the traditional main lines.

Model Game 9

Zelić – I. Bender

Croatia Cup, Mali Lošinj 2017

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 Nf3

A somewhat more popular move-order leading to the same position is 6 h4

h6 7 Nf3 e6, as given in the introduction.

6...e6
Black heads for the Bareev Variation; the traditional main line is
6...Nd7, preventing Ne5.

7 h4

We already had a look at the solid but unambitious 7 Bd3 and other
rare options in 7A and 7B.

7...h6 (D)

White to play
8 h5

8 Bd3 invites Black to transpose with 8...Bxd3 9 Qxd3 Nd7 to the old line
6 h4 h6 7 Nf3 Nd7 8 Bd3 Bxd3 9 Qxd3 e6, where White has avoided
advancing his h-pawn to h5. For some ideas on how to improve on this,
see 9A and 9B.

In contrast, the seemingly more active bishop placement with 8 Bc4 is

again harmless as only one of White’s knights is in a position to threaten
the light
squares around the black king. After 8...Bd6, White has tried:

a) 9 h5 Bh7 10 Qe2 Nf6 11 Ne5 Qc7! and Black is taking over the play in
lines like 12 c3 a5 and 12 Bf4 Nd5. After 12 0-0 0-0 Black is slightly
better, partly thanks to his more compact king’s position. This was even
more visible in Stojčevski-Zelčić, Katowice blitz 2017 after the further
moves 13 f4 Nbd7 14 c3 c5 15 Be3 Rac8 =+.

b) 9 Ne2 reroutes the knight, but is slow. 9...Qc7 10 Be3 Nf6 (10...Bh5!?)
11 Bd3 Bxd3 12 Qxd3 Nbd7 13 Nd2 Nd5 14 Ne4 Nxe3 and Black is fine:
15 Nxd6+ (15 fxe3 Be7 = Grajeda Perez-A.Wong, Buenos Aires Olympiad
1978) 15...Qxd6 16 fxe3 Qb4+ 17 Qc3 Qxc3+ 18 Nxc3 Rd8 =.

8...Bh7 (D)

White to play

It actually makes some sense to group Capablanca Variation

positions according to where the kings are. The two main options
are:

Both castle queenside.

White castles queenside and Black kingside.
But in addition there are some rarer constellations, including those where
White castles kingside despite his advanced h-pawn and some where
Black keeps his king flexible for quite a long time.

9 Bd3

Black’s light-squared bishop is quite strong and has spent three moves
getting to h7, so this is clearly the main line. Instead 9 Bc4 Bd6 transposes
to note ‘a’ to White’s 8th move above.

9...Bxd3 10 Qxd3 Nf6 (D)

Consistent; surprisingly, 10...Nd7, immediately transposing to the

traditional 6...Nd7 lines, has been played by several very strong players. It
doesn’t make much sense to me, but if you know both lines well, perhaps
it’s a good idea to vary occasionally, making preparation harder for future
opponents.

White to play
11 Bd2

This seems the logical way to prepare queenside castling. As usual the main
alternative is 11 Bf4, but then the offer of a bishop exchange on d6 seems
to help Black free his game, while the checks on the a5-e1 diagonal may
lead to more complicated play. See 9D for details. For even rarer 11th
moves, see 9C.

11...Be7

Black prepares ...0-0 followed by ...c5 and active deployment of his

queen’s knight. Surprisingly, Black’s most popular move here is actually
11...Nbd7, which seems almost as inconsistent as on the previous move
(unless you consider having avoided some Bf4 ideas an achievement).

Second in popularity is 11...c5!?, which in contrast is very consistent

and seems to provide at least equality (Black has a heavy plus score,
but that seems mostly a result of Black being better prepared and
higher rated).
However, so does the game move, and in general it leads to more
complex positions where both kings are frequently under attack.

We now return to 11...Be7 (D):

White to play
After White’s h-pawn advance, queenside castling seems normal but
not everybody will feel comfortable in the race to the enemy kings that
will frequently result.

12 0-0-0

After 12 Ne4, Black could consider lines with ...Nbd7 but the
cleanest equalizer seems to be 12...Nxe4 13 Qxe4 Qd5!:

a) 14 Qxd5 cxd5 15 c3 Nc6 16 Bf4 (Aulia-J.Gonzales, Surakarta 2018)

and now the thematic 16...b5 at least equalizes.

b) 14 Qe2 b5 15 a4 Nd7 16 axb5 cxb5 17 Ra5 a6 = Abdusattorov-

Mareco, Sharjah 2019.

12...0-0

Not forced – 12...c5 is one of several interesting alternatives – but once

again consistent and interesting; with opposite castling we’ll soon see very
sharp play.

13 Qe2 (D)

Both players have nearly completed development and we are close to the
middlegame. The main reason that the position passes as ‘opening theory’
is the fact that hundreds of games have been contested from this position,
including quite a few by top grandmasters. White now must choose how to
improve his position. A list of useful moves includes Kb1, Ne4, Qe2, Ne5,
and (with some reservations) c4, so it’s not surprising that this is also a list
of White’s most popular 13th moves. Most of these moves may obviously
be combined, so transpositions abound and we may not be able to point
them all out. See 9F for analysis of 13 Kb1 and 9E for the rest of White’s
13th-move alternatives.
 Black to play
13...c5!

This move was Black’s reason to delay ...Nd7. Now the knight may go to
c6 instead (or even a6).

14 dxc5 Qc7 15 Ne4

White should free his g-pawn to advance and arguably the g3-knight is
White’s worst placed piece anyway. After 15 Ne5 Rd8 16 Bc3 (16 Nd3
Na6!? can hardly be bad for Black) 16...Nc6 17 Nxc6 Qxc6 18 Bxf6 Bxf6
19 Ne4 Be7 Black is comfortable:

a) 20 b4?! Rd5 21 Qc4? (21 Rxd5 is better; e.g., 21...exd5 22 Nc3 d4

23 Qxe7 dxc3 =) 21...b5! –+ Hraček-Epishin, Bundesliga 1998/9.

b) 20 Rxd8+ Rxd8 21 Rd1 Rxd1+ 22 Kxd1 Bxc5 23 Qc4 Qd7+ 24 Ke2 Bb6
=.

We now return to 15 Ne4 (D):

 Black to play
15...Nbd7

Now that White has released his central grip, this may be the best way to
develop the knight, as Black still has to pick up that c5-pawn (though
Black has a number of good alternatives, including 15...Na6 and
15...Rd8).

16 g4

The game enters a phase where each move is critical. It’s not really
surprising to see my computer indicating that 16 Nxf6+ Bxf6 17 g4 Nxc5 18
g5 hxg5 19 h6 should end with a perpetual or a drawn endgame after
19...g6 but the chances of two humans finding their way through the
complications seem rather low.

16...Nxe4 17 Qxe4 Nxc5 18 Qc4 b5!

This is the obvious choice for a strong player. The fact that some
engines might prefer 18...Rfc8 is irrelevant for practical chess.

19 Qxb5??
Totally misjudging the position. Pawns don’t count but open lines do.
Black also has the better chances after 19 Qe2 Rfc8 (19...Bf6!?), although
even there correct play may lead to a draw; e.g., 20 Kb1 Na4 21 Bc1
Rab8 22 g5! b4 23 Nd4! with Qg4 ideas.

19...Rab8 20 Qe2 (D)

After, e.g., 20 Qc4 Qb7, Black’s attack is decisive.

Black to play
20...Rxb2!

This is the thematic way, so it’s a pity that according to Stockfish the
prosaic moves 20...Na4! and 20...Bf6! are even stronger.

21 Bc3

Or 21 Kxb2 Na4+ 22 Ka1 (22 Kb1 Qb6+ –+) 22...Bf6+ 23 c3 Bxc3+ 24 Kb1
Rb8+ and Black wins.

21...Qf4+ 22 Kxb2?
A mistake but Black also wins after 22 Qe3 Qxe3+ 23 fxe3 Rxa2.

22...Na4+ 23 Ka1 Nxc3 24 Qe3 Qc4 25 Kb2

25 Qxa7 isn’t really any better: 25...Nxd1 26 Rxd1 Qxc2 27 Re1 Bb4 –+.

25...Rb8+ 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 (8 h5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nf6 11 Bd2 Be7 12 0-0-0 0-0 13 Kb1)

9A: 8 Bd3 Bxd3 9 Qxd3 Nf6 (10 Bd2 Qc7 11 0-0-0 Nbd7 12 Rhe1)

9B: 8 Bd3 Bxd3 9 Qxd3 Nf6 10 Bd2 Qc7 11 0-0-0 Nbd7 12 Kb1

9C: (11 Ne4!?)

9D: 11 Bf4

9E: (13 Ne5)

9F: 13 Kb1

The most important magnifier in this lesson is 9F, which examines the
main line.

9A and 9B are quite dangerous if somewhat out of fashion, so having a

look at them could well pay off. The same goes for 9D, which is quite
common at club level as it seems very natural for a player who is running
out of book moves.

9C and 9E have been included mainly for the sake of completeness.

9A: 8 Bd3 Introduction

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6)

8 Bd3

This old move shouldn’t be underestimated. White holds his h-pawn back
in order to gain some flexibility.

8...Bxd3 9 Qxd3 Nf6 (D)

White to play

Except for the white h-pawn, this is a very standard position.

Q: Why would anyone reject the free tempo that White gains by attacking
the g6-bishop?

Experience has indeed confirmed that advancing the h-pawn to h5

improves White’s chances. However, this is far from obvious, as White’s
kingside pawn-storm is actually stronger with the pawn on h4.

Q: How can that be so?

In order to open lines on the kingside, White needs to play g4-g5,
confronting the h6-pawn. That’s far easier if the pawn is supported by its
colleague on h4. Therefore, in these lines Black should be very careful, and
if White castles queenside, Black should usually do the same.

10 Bd2

After 10 Bf4 Qa5+ 11 Bd2, Black could well play 11...Qa6 12 Qxa6 Nxa6
13 Ne5 Nb4 = as in Nespor-Hurta, Orlova 2015. However, 11...Qc7,
transposing to the main line below (with a higher move number) is the
more pragmatic choice.

10...Qc7

Bologan prefers to delay this move somewhat, instead developing the

bishop to e7, but to me the queen move seems more flexible as there are
a few lines where the bishop should go to d6 and more where ...c5 is
important (and
...Bf8xc5 a nice tempo win).

11 0-0-0

11 c4 will at best transpose as 11...Bd6 12 Ne4 Nxe4 13 Qxe4 Nd7

equalizes immediately.

11...Nbd7 (D)

Black sensibly goes for (or at least is ready for) queenside castling. I tried
to make 11...Bd6!? work, hoping to develop the knight more actively
following
...c5 but it seems that after 12 Ne4 Nxe4 13 Qxe4, Black has nothing
better than 13...Nd7 anyway.
 White to play
12 Rhe1

12 Kb1 is covered in 9B. Otherwise White has the usual array of

sensible moves:

a) 12 Qe2 0-0-0 13 Ne5 Nb6 14 Ba5 Rd5 15 Bxb6 axb6 16 c4 Ra5 17 Kb1
Bd6 18 f4 Qe7 19 Ne4 Nxe4 20 Qxe4 Kc7 21 Rh3 Rha8 and Black
is comfortable, Golubović-Minasian, European Clubs Cup, Fügen
2006.

b) After 12 c4 0-0-0 13 Bc3 c5, White may return to main lines with 14
Kb1 or try 14 Qe2 Bd6 15 Ne4 Nxe4 16 Qxe4 cxd4 17 Bxd4, and now:

b1) 17...Qxc4+ 18 Kb1 Nf6 19 Rc1 Nxe4 20 Rxc4+ Kb8 21 Bxg7 Rhg8 22
Bxh6 Nxf2 23 Rf1 and rather than 23...Rxg2? 24 Rc2 +– Larsen-
K.Rasmussen, Danish Ch, Lyngby 1991, Black can stay in the game
with 23...Rg4.

b2) 17...Rhg8 18 Kb1 (18 Bxa7? Nf6 19 Qc2 b6 leaves the bishop
trapped) 18...Kb8 19 Bc3 Be7 20 Rhe1 Bf6 =.
c) In some ways 12 Ne4 is White’s most flexible move, as the knight clearly
needs a better square. 12...0-0-0 (D) looks right with White’s g-pawn ready
to advance in case of kingside castling.

White to play
Some important
lines:

c1) 13 Kb1 c5 14 Nxf6 Nxf6 15 Qc4 Ng4 16 Qe2 cxd4 17 Bc1 Bc5 18
Nxd4 Rxd4 19 Rxd4 Bxd4 20 Qxg4 Rd8 = Pilnik-Szabo, Mar del Plata
1962.

c2) 13 g3 Ng4!? and now:

c21) 14 Rhf1 f5 15 Nc3 Bd6 16 Ne2 Ndf6 17 Qb3 Rhe8 18 Nf4 Qb6 19 Nd3
Qxb3 20 axb3 Ne4 = Glaser-Thierry, corr. 2008.

c22) 14 Be1 Be7 15 Qe2 Rhe8 16 Bc3 Bf8 17 Ned2 Nb6 18 Nb3 Nd5 19 Bd2
e5 20 dxe5 Nxe5 21 Rhe1 Nxf3 22 Qxf3 Rxe1 23 Bxe1 Rd7 =
Wharrier- Westera, corr. 2010.

c23) 14 Be3 Ndf6 15 Nxf6 gxf6 16 Qe2 Nxe3 17 Qxe3 c5 18 Kb1 cxd4 19
Rxd4 Bc5 20 Rxd8+ Rxd8 21 Qxh6 f5 22 Ng5 Bxf2 = Glanze-Egorov,
corr.
2011.

c24) 14 Qe2 f5 15 Bf4 (15 Nc3 Bd6 16 Kb1 Rhe8 =) 15...Qa5 16 Ned2 Qxa2
is hard to evaluate but possibly

balanced. We now return to 12 Rhe1

(D):

Black to play
12...0-0-0

With White’s h-pawn still on h4, it’s generally safer to castle queenside.

13 Qe2

After 13 Qb3?! Ng4! Black is probably better already.

13...c5

Now White has tried:

a) 14 Kb1 c4 15 Ne5 (15 Ne4 Nxe4 16 Qxe4 Nf6 17 Qe2 b5 = Lisitsyn-

Makogonov, USSR Ch, Moscow 1940) 15...Nxe5 16 dxe5 Nd5 17 Ne4
Be7 18 Qf3 Rhf8 19 Qg3 += E.Moser-Petroczi, Austrian Team Ch 2003/4.

b) 14 Ne5 Nb6 15 dxc5 Bxc5 16 Kb1 Na4 (16...Rd4!?) 17 b3?! (Arakhamia

Grant-Kachiani Gersinska, European Women’s Team Ch, Leon 2001; 17
Nd3 is preferable) and now Black has 17...Rxd2! 18 Qxd2 Nd5 19 bxa4
Bb4 20 Qd3 Nc3+ 21 Kb2 Nxd1+ 22 Rxd1 Qxe5+ 23 c3 Ba5 =+.

9B: 8 Bd3 Main Line

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Bd3

Bxd3 9 Qxd3 Nf6 10 Bd2 Qc7 11 0-0-0 Nbd7)

12 Kb1 (D)

Black to play

Most games with this position are some years old, but with names like
Euwe, Smyslov and Spassky behind the white pieces, it shouldn’t be
underestimated.
12...0-0-0

This sensible-looking move is clearly most popular. However, the lightly

tested 12...c5!? may improve; e.g., 13 c4?! cxd4 14 Nxd4 Bc5! 15 Bc3 0-0
16 Ne4 Rfd8 and Black has an edge as White should avoid 17 Nxc5?!
Nxc5 18 Qe2 Nce4 19 Nb5 Qf4 –/+.

13 c4

Or:

a) 13 Bc1 c5 14 Qe2 c4 15 Ne5 Nb6 16 c3 Bd6 17 f4 Kb8 18 Rhf1 Nfd5

=+ Ljubojević-Miles, Dubai Olympiad 1986.

b) 13 Ne4 c5 14 Nxf6 Nxf6 15 Qa3 Kb8 16 Qa4 c4 17 Ne5 Rc8 18 Rhe1

Bd6 19 g4?! Nd5 =+ Spoelman-M.Pavlović, Hilversum 2009.

13...c5 14 Bc3

14 Qe2 Bd6 15 Ne4 Nxe4 16 Qxe4 Nf6 17 Qe2 a6 18 Bc3 cxd4 19 Bxd4 Bc5
20 Bxc5 Qxc5 21 Ne5 Qc7 22 Rhe1 Rxd1+ = Ivkov-Parma, Beverwijk 1964.

14...cxd4 15 Nxd4 a6 (D)

 White to play

This position looks perfectly OK for Black (although winning it against an

opponent of equal strength could prove challenging):

a) 16 Nb3 Be7 17 Qe2 Nc5 (or 16 Qe2 Nc5 17 Nb3 Be7) and now:

a1) 18 Be5 Qb6 19 Bd4 Qc7 20 Nh5 Nxb3 21 axb3 Qc6 22 Nxf6 Bxf6 23
Bxf6 gxf6 24 Qh5 Qxg2 25 Qxf7 Qe4+ 26 Ka2 Qf5 ½-½ Aronin-F.Olafsson,
Moscow 1961.

a2) 18 Nxc5 Qxc5 19 Be5 Rxd1+ 20 Rxd1 Rd8 21 Rxd8+ Bxd8

= Nievergelt-Larsen, Zurich 1959.

b) 16 Rhe1 Be7 (16...Ne5 17 Qe2 Nc6 18 Ne4 Be7 19 h5 Nxe4 20 Qxe4

Nxd4 21 Bxd4 +=) 17 Ngf5 Bf8 18 Nxe6 (18 Ng3 just repeats) 18...fxe6
19 Rxe6 gives White fair compensation.

c) 16 Nf3 Bc5 17 Qe2 Rhe8! (17...Bd6?! can be met by 18 c5!? or 18 Ne4

Be7 19 Nxf6 Bxf6 20 Bxf6 Nxf6 21 Ne5 +=) 18 Ne5 (18 b4 Bd6 19 Ne4
Nxe4 20 Qxe4 Nf6 21 Qe2 Qc6 =) 18...Nxe5 19 Bxe5 Qb6 20 f3 (M.Pavlov-
Neamtu, Romanian Ch, Bucharest 1965) and Black’s safest path to equality
is
20...Rxd1+ 21 Rxd1 Rd8 22 Rxd8+ Qxd8 =.

9C: Rare 11th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6)

8 h5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nf6 (D)

White to play
11
Ne4!?

This is a little tricky. Should Black accept the transposition to a not too
popular line usually arising after the main move 11 Bd2 or should he look
for something more active? White’s alternatives are more straightforward:

a) 11 Be3 hopes to save a move on lines where the bishop first goes to d2
and a few moves later to e3. The problem is that it invites checks on the
a5-e1 diagonal:

a1) 11...Qa5+ 12 Bd2 transposes to the note to Black’s 11th move in 9D.
a2) 11...Bb4+!? 12 c3 Bd6! is a thematic but untested idea. After 13 Ne4
Nxe4 14 Qxe4 Nd7 White must find a plan and his king needs to get out
of the centre. Perhaps 15 0-0-0 is best but then it seems that White’s c-
pawn would have been happier on c2.

b) 11 Ne5 Nbd7 12 f4 c5 (D) and now:

White to play

b1) 13 Be3 Qa5+ 14 Bd2 Qa4!? (or 14...Qc7!, transposing to line ‘b2’) 15
c4? cxd4 16 b3 Qa6 17 Qxd4 Bc5 18 Qc3 (S.Sokolov-Alavkin, Kazan
2008) and now 18...Rd8! is very strong, as it brings in ...Bd4 ideas.

b2) After 13 Bd2 Black should avoid 13...cxd4?! 14 Ng6! Rg8 (the point
is that 14...fxg6?? loses to 15 Qxg6+ Ke7 16 Bb4+) 15 Nxf8 Nxf8 16 0-0-
0 Qd5 17 Qa3 += Jakubowski-Nihal, Warsaw 2017. Instead 13...Qc7!
seems fine: 14 Nxd7 (14 c3?! cxd4 15 cxd4 Be7 16 Bc3 0-0 17 0-0-0
Nd5 –/+
Driza-V.Johansson, Bled Women’s Olympiad 2002) 14...Qxd7 15 dxc5
(15 Bc3?! 0-0-0 16 0-0-0, as in Escobar Dominguez-D.Fridman, Internet
blitz 2009, is best met by 16...Qc6 17 Qe2 Bd6 –/+) 15...Qxd3 16 cxd3
Bxc5 17 Ne4 Bd4 18 Rc1 Nd5 =.
11...Nxe4

Black more frequently plays 11...Nbd7, which allows 12 Bd2, transposing

to the line 11 Bd2 Nbd7 (not our repertoire move) 12 Ne4 (not the main
move). That wouldn’t be much of a problem, but the knight exchange
introduces another important idea.

12 Qxe4 Qd5 (D)

White to play

This trade offer frees Black’s position.

13 Qe2

Or:

a) 13 Qd3 Nd7 14 Bd2 b5 15 b3 Be7 16 c4 bxc4 17 bxc4 Qd6 18 Qe4 Nf6 19

Qf4 c5 20 Qxd6 Bxd6 21 Rb1 cxd4 22 Nxd4 0-0 =+ Klyuner-Bagirov,
Recklinghausen 1999.

b) 13 Qxd5 cxd5 and now:

b1) 14 Bf4 Nc6 15 0-0-0 Rc8 16 Kb1 Be7 17 Rhe1 (Hajnal-Velička,
Austrian Team Ch 2015/16) and now 17...b5 is an instant equalizer.

b2) 14 Ne5 Nc6 15 c3 Nxe5 16 dxe5 b5 (16...Bc5 may be safer) 17 a3 Bc5

18 Ke2 0-0 19 Be3 Rfc8 = K.Richter-Bashilin, Dortmund 2004.

13...Nd7

13...b5!? is a new move which could be worth investigating.

14 Bd2 (D)

Black to play

Unlike Black, White doesn’t have the option of giving his king a secure
home by castling kingside.

14...b5

This computer suggestion looks very logical as it allows the black queen
to retain its dominating position and makes queenside castling less
tempting. I don’t like 14...0-0-0?! 15 c4 Qf5, as in Savanović-Ratković,
Bosnian Team
Ch, Sarajevo 2017, since 16 Bc3 seems to keep an edge for White.

15 b3

White fights for control over c4 and d5. 15 a4 doesn’t threaten much so
Black may go ahead with 15...Be7 as in the main line or even the
somewhat more active-looking 15...Bd6.

15...Be7

One of White’s ideas is c4 followed by d5, so it seems sensible to block

the e-file.

16 c4

Provoking castling with 16 a4 doesn’t promise much after 16...0-0 17 c4

bxc4 18 bxc4 Qf5. White’s best try is the wild 19 g4! Qxg4 20 Bxh6! Bf6 21
Bxg7 but although it’s very complicated, Black is at least equal.

16...bxc4 17 bxc4 Qf5

This looks nice for Black but White can probably keep the balance
with active play.

18 Rd1! (D)
 Black to play
Now:

a) Black should avoid 18...0-0?! 19 g4! Qxg4 20 Bxh6! with dangerous play.

b) 18...Rb8 19 d5! is complicated and probably equal.

c) 18...Rd8 may be best. Giving meaningful lines in the absence of

practical examples is difficult but it should be pointed out that after 19
Nh4 Qf6 20 Ba5, Black has 20...Nb6 since 21 c5? loses to 21...Rxd4!.
Instead 21 Nf3 c5 is probably equal; e.g., 22 Bc3 0-0 23 0-0 cxd4 24
Nxd4 Qg5 =.

9D: 11 Bf4

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 h5 Bh7

9 Bd3 Bxd3 10 Qxd3 Nf6)

11 Bf4 (D)

This active developing move is second in popularity after the bishop

development to d2. Now Black may give checks on the a5-e1 diagonal to
see
whether White will retreat his bishop or weaken his queenside by
advancing his c-pawn.

Black to play
11...Bd6

I prefer this offer to exchange bishops over the checks mainly because it’s
so clear. Black frees his game and equalizes quite easily. That being said,
11...Qa5+ 12 Bd2 Qa6 (12...Bb4!? is more complicated but maybe even
stronger) is also a very simple solution if you enjoy semi-endgames:

a) 13 c4 Bd6 14 Ne5 c5 15 Bc3 cxd4 16 Bxd4 Qa5+ 17 Bc3 Bb4 18

Bxb4 Qxb4+ 19 Qd2 Qc5 =+ T.Ernst-Zelčić, Cutro 2004.

b) 13 Qxa6 Nxa6 and Black is fine: 14 Ke2 (14 0-0-0 Bd6 15 Rde1 Bxg3
16 fxg3 c5 17 dxc5 Nxc5 18 Bc3 0-0 19 Bxf6 gxf6 = Feuerstack-Ressler,
Kiel 2000) 14...Bd6 15 Rh4 Bxg3 16 fxg3 c5 17 dxc5 Nxc5 =+ Mendoza-
Arutinian, Turin Olympiad 2006.

12 Bxd6
After 12 Ne5 Bxe5 13 Bxe5 Qa5+ 14 c3 Nbd7 15 f4 0-0-0 White is
struggling to equalize as his king is exposed and his bishop also vulnerable
to
...f6 ideas; e.g., 16 Qe2 (16 Qc4?!, as in Petijević-Golubović, Dubrovnik
2008, should be met by 16...Nd5 –/+) 16...Ne8 17 Ne4 f6 18 Bd6 f5 19
Bb4 Qd5 20 Ng3 Nc7 21 0-0 and in Ermenkov-Filipenko, Arco seniors
2010, Black should have prepared ...g5 with 21...Nf6 22 Rfd1 Rhg8 23 c4
Qd7, when he has the better chances.

12...Qxd6 13 0-0-0 (D)

Black to play
13...Nbd7

13...0-0 is an alternative move-order as 14 Kb1 Nbd7 transposes, but

offers White 14 Ne5 as an additional option.

14 Kb1

Or 14 Qe2 0-0 15 Ne5 Qc7 16 Rhe1 Rad8 17 f4 c5 18 dxc5 Nxc5 19

Qc4?! Rc8 (19...b5!) 20 Kb1 Qb6 21 Qd4 Rfd8 22 Qf2 Nd5 23 Ne2 Qa5
=+
J.Polgar-Ivanchuk, Mexico City rapid 2010.
14...0-0

After 14...0-0-0, it seems that 15 Ne4 gives White some plus.

15 Ne4

I noticed that some engines like 15 Nd2 but 15...a5 16 Nge4 Qe7 gives
Black his fair share of the chances.

15...Nxe4 16 Qxe4 Nf6 17 Qe2 (D)

After the decentralizing 17 Qh4, 17...c5 looks fine for Black.

Black to play

This seems to be another position which is probably equal but must

be handled accurately because of the opposite castling:

a) 17...Rfd8 18 Ne5 c5 and now:

a1) 19 dxc5 Qxc5 20 g4 Rac8 21 c4 a6 22 f3 Qc7 gives Black

interesting play in an equal position, Marjanović-Campora, Bor 1983.
a2) 19 g4 cxd4 20 Rhg1 Nd7 21 Nf3 f6 (21...e5!? 22 g5 is also a little
better for White) 22 Nxd4 Nf8 23 c3 += Rebord-T.Carlsen, corr. 2007.

b) 17...Qd5 18 c4 Qe4+ (18...Qf5+!?) 19 Qxe4 Nxe4 20 Rh4 (Rotman-

Khenkin, Geneva 1996) 20...f5 probably gives Black a small plus.

c) 17...Rac8 seems to be the computer favourite. After 18 Ne5 c5 19 dxc5

Qxc5 20 g4 Rfd8 21 c3 b5 22 f4, Black has the simplifying idea 22...Nd5
23 Qe4 Qe3 24 Qxe3 Nxe3 25 Rde1 Nc4 =.

9E: Rare 13th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 h5 Bh7

9 Bd3 Bxd3 10 Qxd3 Nf6)

11 Bd2 Be7 12 0-0-0 0-0 (D)

White to play

13 Ne5

Or:
a) 13 Nf1?! prepares g4 but 13...c5 equalizes (or more) immediately: 14
Ne3 Nc6 15 dxc5?! (15 Bc3 cxd4 16 Nxd4 =) 15...Qxd3 16 cxd3 Bxc5 17
Bc3?! Rfd8 18 a3? (18 Rh4 Nd5 =+) 18...Bxe3+ 19 fxe3 Ng4 –/+ Polok-
Triapishko, Internet blitz 2020.

b) After 13 c4?! b5 (13...Nbd7 can transpose to ‘b1’ after 14 Kb1 b5)

Black may already be better:

b1) 14 Kb1 Nbd7 15 cxb5 cxb5 16 Qxb5 a5 17 Ne5?! Nxe5 18 dxe5?! Nd5
–/+ Ganguly-Yakovenko, Spanish Team Ch, Motril 2008.

b2) 14 cxb5 Qd5 15 Kb1 cxb5 16 Ne2 Ne4 (16...Nc6! should suffice for
a small plus) 17 Be3 Nc6 18 Nf4 Qf5 19 g4 Qh7, as in Zapata-Macieja,
Greensboro 2013, is very hard to evaluate.

c) 13 Ne4 Nxe4 14 Qxe4 Qd5 15 Qg4 (15 Qxd5 cxd5 16 g4 Nc6 17 Kb1
Rfc8 18 Be3 b5 19 Ne1 a5 = Milanović-Campora, Pančevo 1985)
15...Kh8 16 Kb1 Nd7 = Wei Yi-Ding Liren, Chinese Team Ch 2015.

d) After 13 Rhe1 c5 14 Bc3 Qc7 15 dxc5 Black must be careful:

d1) 15...Nbd7?! (Naroditsky-M.Bengtson, USA Chess League 2006)

is strongly met by 16 Nf5! since 16...Bxc5? fails to 17 Nxg7! Qf4+ (or
17...Kxg7 18 Qxd7) 18 Rd2! Qg4 19 Ne5 Nxe5 20 Rxe5, with a huge
advantage.

d2) 15...Rd8 16 Be5 Rxd3 17 Bxc7 Rxd1+ 18 Rxd1 Nbd7 19 Bd6 Kf8 20
Ne5 Nxc5 21 Bxc5 Bxc5 22 Nd7+ Nxd7 23 Rxd7 Bxf2 24 Ne4 Be3+ 25
Kd1 b6 26 Nd6 f5 =.

We now return to 13 Ne5 (D):

 Black to play
13...c
5

With the white knight on e5, 13...Nbd7 also seems logical. After 14 Qe2 c5
15 dxc5 Nxc5 Black is doing well: 16 Kb1?! (16 Bxh6! Qa5 17 Be3 Qxa2
18 Bxc5 Bxc5 and just to avoid immediate disaster White must find a crazy
tactical sequence with 19 c4! intending h6, when a perpetual check is
likely) 16...Qc7 17 Bc3 Rfd8 18 Ng4 Nd5 19 Be5 Qc6 20 Rhe1?! Rac8 and
Black was for choice in Rebord-Tazelaar, corr. 2011, which concluded 21
c3 Na4 22 Ka1 f6 23 Bd4 Ndxc3 24 Bxc3 Rxd1+ 25 Rxd1 Nxc3 26 bxc3 f5
27 Rd3
Bb4 0-1.

14 dxc5

White could also try 14 Qf3 cxd4 15 Qxb7; e.g., 15...Qd5 16 Qxd5 Nxd5
17 Ne2 Bd6 18 Nc4 Bc5, when 19 Bf4 Nc6 20 Bh2 f6 is hard to evaluate,
while 19 Be1 Nc6 20 Na5 Nxa5 21 Bxa5 d3 22 Rxd3 Bxf2 seems equal.

14...Qc7 (D)
 White to play

Black seems fine in this unbalanced position:

a) 15 Bc3 Nd5 (15...Na6!?) 16 Ne4 Na6 17 Bd4 Rac8 18 a3?! (18 Kb1!?)
18...Rfd8 19 c6 (Almasi-Dautov, Bundesliga 1996/7) 19...Nf4 =+.

b) 15 f4 Nbd7 16 Nxd7 Nxd7 17 Ne4 Nxc5 18 Nxc5 Qxc5 19 Bc3 (½-

½ Almasi-Dautov, Bundesliga 1997/8) 19...Rfd8 =.

c) After 15 Qe2, you probably want to avoid 15...Qxc5 16 Ng6! Re8 (the
point is that 16...fxg6? 17 Qxe6+ Kh8 18 Be3 leaves Black in trouble;
e.g., 18...Qc7? 19 hxg6 Ng8 20 Nf5 Rxf5 21 Qxf5 +–) 17 Ne4 Nxe4 18
Nxe7+ Qxe7 19 Qxe4 Nc6, as in Sahin-E.Atalik, Turkish Women’s Ch,
Antalya 2018, when 20 g4 might have been quite dangerous. Instead
15...Rd8! looks at least equal for Black; e.g., 16 Bc3?! Na6 17 Ne4 Nxe4
18 Qxe4 Nxc5 19 Qe3 Na4 and Black is better.

9F: 13 Kb1

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 h5 Bh7

9 Bd3 Bxd3 10 Qxd3 Nf6 11 Bd2 Be7 12 0-0-0 0-0)

13 Kb1

This flexible move is White’s most popular choice by some margin.

13...c5 (D)

White to play

Flexibility doesn’t really trouble us – if White plays slowly we have a default

plan.

14 Be3

The bishop moves again. When seeing this move, I sometimes feel the
urge to go back and recheck that the 11 Be3 lines are harmless. Other
moves:

a) 14 Bc3 Nbd7 15 dxc5 Nxc5 16 Qe2 Qb6 17 Bd4 Qa6 18 c4 Ncd7 19

Nf1 Qc6 20 Ne3 Rfd8 =+ Svidler-Leko, Madrid 1998.

b) 14 Ne4 Nc6! (14...Nbd7?! 15 g4! transposes to a sharp position from the

7...Nd7 move-order where White scores well) 15 dxc5 Qxd3 16 Nxf6+
Bxf6 17 cxd3 Rad8 18 Be3 Rd5 19 d4 Rfd8 is equal; e.g., 20 Kc2 a5 21
a3 a4 22 Rd3 Rf5 23 Rc3 Bxd4 24 Nxd4 Nxd4+ =.

14...Qc7!? (D)

The more common 14...Nbd7 tends to transpose, but I like to retain the
option of putting the knight on a6 or c6.

White to play

15 Ne4

Or:

a) 15 Ne2 cxd4 16 Bf4 Qc5 17 Nfxd4 Rd8 18 Qc3 Nbd7 19 Qxc5 Bxc5
20 Bg3 a6 =+.

b) 15 dxc5 Nbd7 16 Ne4 Rfd8 17 Qc4 a6 18 Nxf6+ Nxf6 19 a3 Nd5 20

Bd4 b5 21 Qe2 Bxc5 =+.

15...Nbd7
Now play merges with the 14...Nbd7 15 Ne4 Qc7 move-order.

16 dxc5

White is actually struggling after 16 Nxf6+ Nxf6:

a) 17 dxc5 Nd5 18 g4 Nxe3 19 Qxe3 Bxc5 20 Qc3?! Rac8 21 Qd2?! Rfd8

–/+ Ghatti-N.Kumar, Irving juniors 2017.

b) 17 g4?! Nxg4 18 Rhg1 f5?! 19 Qc4 Qc6 20 Ne5 Nxe5 21 dxe5 f4

(21...Kh7? 22 Rg6 +/– McShane-Jo.Gustavsson, Kuppenheim blitz 2003)
22 Bc1 b5 =+.

c) 17 c4 cxd4 (17...Rfd8!?) 18 Bxd4 Rac8 19 Rc1?! Rfd8 =+

V.Gurevich- Pitić, Mainz 1995.

16...Rfd8 (D)

White to play

In this interesting position chances seem equal.

17 Nxf6+

Other moves don’t offer any advantage either:

a) 17 Nfd2 (Jelica-Francisković, Croatian Women’s Ch, Sveti Martin

2009) 17...Nd5 18 g4 Ne5 19 Qe2 Nxe3 20 fxe3 Qc6 21 Rhg1 f6 =+.

b) 17 Qc4 a6 18 Nxf6+ Nxf6 19 Nd2 (19 a3 Nd5 20 Bd4 b5 21 Qe2 Bxc5

=+) 19...Nd5 20 Bd4 Rac8 21 Nb3 e5 22 Be3 b5 23 Qe4 Nxe3 24 fxe3 a5 25
Rxd8+ Rxd8 26 c6 a4 27 Nc1 Rc8 28 Rd1 Qxc6 29 Qxc6 Rxc6 =.

17...Nxf6 18 Qc4

Or:

a) 18 Qe2 Ng4! 19 Bd4 Bxc5 20 c3 Rac8 =+.

b) Against 18 Qb5, as in Abdusattorov-Kathmale, Internet rapid

2018, 18...Ng4 seems the easiest path to a good position.

18...Nd5 19 Bd4 b5!?

An interesting pawn sacrifice. Instead, in Papadopoulos-Francisković,

European Ch, Budva 2009 chances were equal after 19...Rac8 20 c3 Bxc5
21 Qxc5 Qxc5 22 Bxc5 Rxc5 23 Rd4 Rdc8.

Now we follow the game Luther-Rausis, Bundesliga 1998/9, which was

quite entertaining:

20 Qxb5

This greedy move is objectively best, but it will be tough to defend

the queenside against Black’s heavy artillery.

20...Rab8 21 Qe2 Bxc5 22 Ka1??

22 Be5 is probably best. Then, after 22...Bd6 23 Rd3, Black has several
moves that give reasonable compensation. Objectively White can defend,
but his practical task is very challenging.
22...Bxd4! (D)

White to play
23 Rxd4

After 23 Nxd4, the fancy 23...Nc3! decides.

23...Rxb2 24 Rxd5

Or 24 Kxb2 Rb8+ 25 Kc1 Rb1+ 26 Kxb1 Nc3+ 27 Kb2 Nxe2 –+.

24...Rdb8 25 Rd3 Rxc2 26 Qd1 Qc4 27 Rd8+ Rxd8 28 Qxd8+ Kh7

Here White resigned.

Lesson 10: Introduction to the Bareev Main Line

By: SJ

Against Alekhine you never knew what to expect. Against Capablanca you
knew what to expect, but you couldn’t prevent it.

Milan Vidmar

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

As we learned from the previous lesson, this is the only critical move.

8...Bh7 (D)

White to play

The bishop is important and should only be exchanged for its

white counterpart.

9 Bd3

The fact that this is very much the main line is another indication of the
respect that Black’s light-squared bishop commands. Analysis of 9 Bc4
and even rarer alternatives can be found in 10A.
9...Bxd3

Definitely not 9...Qxd4?? 10 Nxf7! +– – and you need a very open mind
even to consider the exchange instead happening on h7.

10 Qxd3

White’s development lead is striking and Black must be careful. However,

it should be noted that the difference in activity isn’t quite as big. Black’s
remaining bishop is easier to activate than White’s and the knight on g3,
although technically developed, isn’t actually contributing much to the
struggle.

10...Nd7! (D)

White to play

You may find this slightly surprising, as one of the ideas behind the
Bareev Variation was to delay this move and instead look for more active
options like ...c5 and maybe a later ...Nc6. Instead with the text-move
Black tries to gain the tempi that the white knight on e5 has spent to get
there, and he does so without a knight on f6. This move-order has at least
two advantages:
 After an exchange on e5, a white pawn on e5 won’t attack
anything. There is nothing on the d8-h4 diagonal obstructing an
attack on the white h-pawn by the queen and a bishop on e7.

11 Bf4

This is clearly the main alternative to 11 f4 (Lessons 11 and 12), which is

well illustrated by the fact that these are almost the only moves you’ll find
in the correspondence database.

Model Game 10

Bezemer –

Svetushkin Agios

Kirykos 2002

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11

Bf4 For some minor alternatives, see

10B. 11...Nxe5

11...Qa5+?! may come to the same thing, but gives White a useful extra
option as dxe5 won’t mean an exchange of queens: 12 c3 Nxe5 13 dxe5!?
(13 Bxe5 transposes to the main line) 13...Rd8 (13...Ne7!?) 14 Qe2 Ne7
(Bologan-Dizdarević, Sarajevo 2004) and now 15 Ne4! keeps Black under
pressure.

12 Bxe5

12 dxe5 is playable but after 12...Qxd3 13 cxd3 Rd8 14 Ke2 f5!? White
has some coordination problems.

12...Qa5+

Provoking the weakening c3 advance.

13 c3 (D)
After 13 Qd2?! Qxd2+ 14 Kxd2 Rd8 15 c3 c5, if anyone is better, it’s Black.

Black to play

While White has some extra space and faster development, Black’s main
assets are his more mobile bishop and his ability to castle into relative
safety on the queenside. For the moment his queen also prevents White
from castling queenside.

13...Nf6 14 Qe2

This allows Black to equalize comfortably. Queenside castling is for the

moment not an option but there are a couple of alternatives that demand
accurate defence, so each will get a separate examination under the
magnifier. See 10C for 14 Ne4 (and a few minor options) and 10D for 14 0-
0.

14...Be7

14...Nd7 is less ambitious, but looks solid: 15 Bf4 Be7 16 Ne4 Nf6 17
Nxf6+ Bxf6 18 h5? (18 Kf1 =) 18...Bxd4 –/+ I.Sarić-Elianov, Internet rapid
2019.

15 Ne4
15 Nf1!? is interesting. 15 Kf1 could well be the right way to take care of
the king; e.g., 15...0-0-0!? (15...0-0 =) 16 a4 Bd6 17 b4 Qc7 18 Bxf6 gxf6
and Black need not fear 19 Ne4?! (Sivokho-Solozhenkin, St Petersburg
Ch 2000) 19...f5 20 Nxd6+ Qxd6 =+.

15...Qd5 (D)

White to play

Black has caught up in development and White faces a few

tactical challenges thanks to his vulnerable g-pawn.

16 f3 Nd7 17 Bxg7 Rh7 18 Be5 Nxe5 19 dxe5 Qxe5

Quite an interesting position. Chances may be roughly balanced but in

many endgames Black’s strong bishop will give him an edge. Meanwhile
White has a harder time getting his king into safety.

20 h5?! Rg7?!

20...0-0-0 is more accurate.

21 g4?!

White should prevent Black from castling with 21 Rd1 (as he should
have done last move), though after 21...f5 22 Nf2 Qxe2+ 23 Kxe2 Bc5
24 Nd3 Bd6 Black may still have the upper hand.

21...0-0-0 22 Rd1 Rgg8 (D)

White to play

Black is clearly better, mainly because of the exposed white king. The rest
of the game is of little theoretical interest so please excuse the somewhat
cursory annotation.

23 Nf2 Qg3 24 Kf1 Rxd1+ 25 Qxd1 Rd8 26 Qe2 Qf4 27 Ne4 Rd7

The direct 27...f5 is more or less decisive.

28 Rg1 Bd8 29 Nf2

29 c4 is a little more stubborn.

29...Rd2

Probably 29...Bb6 would have won more quickly.

30 Nd3 Rxe2 31 Nxf4 Rxb2

White is a pawn down and his camp is completely disorganized.

32 Rg2 Rb1+ 33 Ke2 Bf6 34 c4 Rc1 35 Nd3 Rxc4

Here White might as well have resigned.

36 Ke3 Bd4+ 37 Ke4 Bb6+ 38 Ke5 Bd4+ 39 Kd6?

White walks into a mating-net in a lost position.

39...Bf6 40 Rd2 Rd4+ 41 Kc5 Kc7 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 (9 Bd3 Bxd3 10 Qxd3 Nd7 11 Bf4 Nxe5 12 Bxe5 Qa5+ 13 c3 Nf6 14 0-
0)

10A: (9 Bc4)

10B: (11 Qe2)

10C: (14 Ne4)

10D: 14 0-0

We are just about to enter theoretical waters but even the lesson’s main
move, 11 Bf4, is most of all an attempt to avoid critical lines. So none of
these magnifiers should have top priority in your study plan. Anyway, when
you decide to have a closer look, you would be well advised to start with
10C and 10D.
10A: Rare 9th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6)

8 Ne5 Bh7 (D)

White to play

There are a number of lines where White avoids the exchange of light-
squared bishops. Some have been tested by very strong players who
wanted to avoid a theoretical duel, but none of them are theoretically
critical.

9 Bc4

The bishop looks active here but if Black is careful it may actually be a
bit misplaced. Other unusual options:

a) 9 Qf3 Nf6 10 c3 Nbd7 11 Bf4 Be7 12 Nxd7 Qxd7 =.

b) 9 c3 Nd7 10 Nxd7 Qxd7 11 Bd3 Bxd3 12 Qxd3 0-0-0 13 Be3 Nf6 14 Ne4
Ng4 (14...Nd5 =) 15 Qe2 Nxe3 16 fxe3 Be7 = Abdusattorov-
Shankland, Doha 2015.
c) 9 Bf4 Nf6 10 Qd2 Nd5 11 a3?! Bxc2! 12 Nxf7 Kxf7 13 Bxb8 Rxb8 14
Qxc2 Qa5+ –/+ Wojciechowski-I.Appel, Polish Ch, Jurata 1937.

d) 9 Be3 Nd7 10 Nxd7 Qxd7 11 Bd3 Bxd3 12 Qxd3 Nf6 13 0-0-0 Be7 14
Ne4 Qd5 15 Nxf6+ Bxf6 =+ Pourramezanali-Ni Hua, Sharjah 2014.

e) 9 Qh5 Qc7 (D) and now:

White to play

e1) 10 Bd3 Bxd3 11 Nxd3 Nf6 12 Qe2 Nbd7 13 Bf4 Bd6 14 Be5 0-0 15 0-0-
0?! (15 f4 c5 =) 15...c5 16 Bxd6 Qxd6 17 dxc5 Nxc5 18 Kb1 Rfd8 19
Nxc5 Qxc5 =+ Dvoirys-Rodshtein, Israeli Team Ch 2019.

e2) After 10 Bf4 Nf6 11 Qf3, it is obviously possible to grab the c2-pawn
but White will win some time and have roughly enough compensation, so
11...Bd6 seems sensible; e.g., 12 Bd3 Bxd3 13 Nxd3 Bxf4 14 Nxf4 Nbd7 =.

e3) 10 Bc4 Nf6 11 Qe2 Bd6 12 c3 0-0 13 Nh5? Nxh5 14 Qxh5 c5 15 Be3 (15
Bxh6?! isn’t quite sufficient: 15...gxh6 16 Rh3 Bxe5 17 dxe5 Kh8 –/+)
15...cxd4 16 Bxd4 Nc6 17 Nxc6 Qxc6 –/+ Vavulin-Svane, Moscow
2017.
We now return to 9 Bc4 (D):

Black to play
9...Nd7 10
Qe2

Or 10 Bf4 Ngf6 11 c3 Be7 12 Nh5 Nxh5 (12...0-0 is viable too) 13 Qxh5

Nxe5 14 Bxe5 0-0 15 Qg4 (15 Rh3!?) 15...Bf6 16 Bxf6 Qxf6 17 0-0 =
Kokarev-I.Popov, Moscow 2014.

10...Nxe5

Not 10...Ngf6??, which loses to the standard knight sacrifice 11 Nxf7!.

11 dxe5 Ne7 (D)

 White to play

Black has an easy position to play, while White must be careful to

avoid ending up worse:

a) 12 Ne4? Qa5+ 13 Bd2 Qxe5 14 Bd3 Nd5 15 0-0-0 Nf4 (15...Ba3 16 c3

Nf4 17 Bxf4 Qxf4+ 18 Kb1 Be7 =+) 16 Bxf4 Qxf4+ 17 Kb1 0-0-0 =+
E.Berg-Berczes, Stockholm 2010/11.

b) 12 Bd2 Bxc2 13 Bxh6? (13 0-0! gives White enough play for the pawn,
but no more than that) 13...Rxh6 14 Qxc2 Qa5+ 15 Qc3 Qxc3+ 16 bxc3
Ng6
–/+ Schuermans-Nachbar, Belgian Team Ch 2007/8.

c) After 12 Bb3 a5 13 a4 Nd5 14 Bd2, Black’s two most interesting

moves are 14...Bb4 and 14...Qb6, planning ...Bb4 (as in Tseshkovsky-
Epishin, Russian Ch, Moscow 2004), in both cases with good chances
to play for a win.

d) 12 Be3 Nd5 (12...Qa5+ 13 Bd2 Qc7 14 0-0-0 0-0-0 is a safe path to

equality) 13 0-0-0 Qa5 14 Bd2 Bb4 (after 14...Qb6?!, as in Höglauer-
Pfreundt, Schwäbisch Gmünd 2017, 15 Bd3 Bxd3 16 Qxd3 0-0-0 17 Qf3
leaves Black under pressure) 15 Bxb4 (15 c3?? Nxc3! –+; 15 Ne4 Bxd2+
16
Rxd2 Bxe4 17 Qxe4 0-0-0 =) 15...Qxb4 16 Ne4 0-0 and now the sample
line 17 a3 (17 Nd6?! b5 18 Bxd5 cxd5 19 Rh3 {19 Qxb5?? Qf4+ 20 Rd2
Qxe5 –
+} 19...Rab8 =+) 17...Qe7 18 g3 b5 19 Bd3 Rab8 20 Nd6 Bxd3 21 Rxd3 c5
looks OK for Black.

10B: Rare 11th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8

Ne5 Bh7)

9 Bd3 Bxd3 10 Qxd3 Nd7 (D)

White to play
11 Qe2

This has scored well in the relatively few games where it has been
tested. Other moves don’t trouble Black:

a) The piece sacrifice 11 Ng6?? fxg6 12 Qxg6+ Ke7 is completely

incorrect: 13 0-0 (13 Nf5+ exf5 14 0-0 Ndf6 –+ Mazi-Zelčić, Ljubljana
1999) 13...Qe8
14 Qd3 (L.Xavier-E.Ramos, Cuiaba 2013) 14...Ngf6 and White has very
little for his knight.

b) 11 Nxd7 Qxd7 seems to help Black’s development:

b1) 12 Bf4 Nf6 13 0-0-0 Bd6 14 Be5 0-0-0 15 Qf3 Qe7 16 Kb1 Rd7 17 c4
Rhd8 18 Rc1 c5 19 Bxd6 Qxd6 20 d5 exd5 21 Nf5 Qe5 (21...Qf8!?) 22
Rhe1 Ne4 23 Ka1 g6 (23...Qf6 = I.Sarić-Berkes, European Ch, Legnica
2013) 24 Nxh6 f5 25 cxd5 Qxd5 26 Ng4 Qd4 =.

b2) 12 Ne4 Be7 13 Bf4 Nf6 14 Be5 Ng4 15 0-0-0 Qd5 16 Nc3 Nxe5 17 Qg3
Qc4 18 dxe5 0-0 = V.Kovalev-Navara, Minsk rapid 2015.

11...Ngf6 12 Bd2 c5 13 dxc5

Or:

a) 13 Nf5 is possible but seems rather pointless when the knight doesn’t
attack anything. After 13...Qb6 14 Nxd7 Nxd7 15 0-0-0 0-0-0, White
hasn’t achieved anything.

b) 13 0-0-0 cxd4 14 Nf5 Nxe5 15 Qxe5 Qd5 16 Qxd5 Nxd5 17 Nxd4 Bc5
18 c3 Bxd4 19 cxd4 Ke7 =+ A.Sokolov-Solozhenkin, Bethune 2000.

13...Bxc5 14 Nxd7 Qxd7 15 0-0-0 (D)

 Black to play
Now:

a) 15...Qc6 16 Bc3 Be7 17 Rhe1 0-0 18 Bxf6 (18 Nf5 exf5 19 Qxe7 Nd5 =+)
18...Bxf6 19 Nh5 e5 (19...Bxh4!? also looks playable but difficult) seems
to come close to equality:

a1) 20 Rd3?! Bxh4 21 Qxe5 Qg6 =+ Sredojević-Bosnjak, Grohote 2018.

a2) It’s not obvious why Stockfish prefers White after 20 Qe4 Qe6 21
Nxf6+ Qxf6 22 Qxb7 Qxh4 (22...Qxf2!?) 23 g3 Qf6. It seems Black will
have nice counter-chances on the half-open b- and c-files.

b) 15...Qa4 16 Bc3 Be7 17 Rd4 Qc6 18 Rc4 Qxg2 is very sharp. Predke-
Matlakov, Moscow 2017 continued 19 Re1 (19 Rd1 is an obvious
alternative but can hardly change the impression that the position is
relatively equal with the opposite castling as an unbalancing factor) 19...0-
0 20 Bxf6 Bxf6 21 Rg4 Qd5 22 Nh5 Kh7 23 Reg1 Rad8 24 Nxf6+ gxf6 25
Rg7+ Kh8 26 Qe3 Qh5, which looks a bit scary for Black, but Stockfish isn’t
worried.
10C: 14 Ne4 and Minor 14th Moves

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8

Ne5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7)

11 Bf4 Nxe5 12 Bxe5 Qa5+ 13 c3 Nf6 (D)

White to play
14 Ne4

This looks rather toothless but solves the problem of the offside knight
and has been played by some strong players. 14 0-0 is covered in 10D.
Some minor alternatives should also be mentioned:

a) 14 Bxf6 gxf6 15 0-0 Be7 16 Rfe1 0-0-0 17 b4 Qc7 18 Qe2?! Rhg8 19 Qh5
Rg7 20 Ne2 f5 21 g3 (Sadzikowski-Zelčić, Pula 2012) 21...f4 –/+.

b) 14 b4 Qd8 15 b5!? (15 a4 is steadier) 15...cxb5 16 Bxf6 gxf6 17

Qxb5+ Qd7 18 a4 Rg8 19 Nh5 Be7 20 g3 f5 21 Nf4 0-0-0 22 Qxd7+
Rxd7 23 Nd3 Bd6 = G.Jones-Elianov, Internet rapid 2019.
c) 14 Qe2 allows Black comfortable equality, as we saw in Model Game 10.

14...Nxe4

14...0-0-0 also seems playable if you prefer to keep the queens on:

a) 15 Nxf6 Qxe5+ 16 dxe5 Rxd3 17 Ke2 Rd8 18 Ne4 Rd5 = Bagi-

Berczes, Zalakaros 2014.

b) 15 Qf3 Nxe4 16 Qxe4 Qd5 17 Qxd5 Rxd5 18 0-0-0 b5 (18...h5 could be

safer) 19 Rhe1 Kd7 20 Kc2 Be7 21 g3 f6 22 Bf4 and now both 22...h5, as
in Soylu-Triapishko, Turkish Team Ch, Ankara 2018, and 22...g5 seem
close to equality.

c) 15 Bxf6 gxf6 16 Nxf6 Bg7 17 Qf3 Rxd4 18 0-0 Rxh4 19 g3 could be

White’s best try for a plus; e.g., 19...Qg5 20 Ne4 Qh5 21 Nd6+ Kb8 22
Qxh5 Rxh5 23 Nxf7 Rf8 24 Nd6 Rd5 25 Ne4 +=.

15 Qxe4 (D)

Black to play
Thanks to the simplification, White’s extra space is of little importance,
and Black’s remaining bishop, although a little passive, is unhindered by
pawns.

15...Qd5

This move, offering further exchanges, seems very logical and is the
only move to have been tested in practice.

16 Qxd5

In M.Pap-Wenninger, Vellmar 2018, White tried 16 f3, after which 16...h5

= and 16...Qxe4+ 17 fxe4 h5 = seem more logical than the 16...Qc4 of the
game.

16...exd5

16...cxd5 keeps a little more play in the game. White’s slightly more active
pieces should ensure him comfortable play but over time Black may
generate counterplay:

a) After 17 Kd2 h5 18 Rae1 Rc8 19 Re3?!, as in Karelina-Mirzoeva, Satka

2016, Black can get the better chances with 19...b5; e.g., 20 a3 f6 21 Bg3
Rc6 22 Rhe1 Kd7 =+.

b) 17 h5 Rc8 (17...Kd7 18 Kd2 +=) 18 a4 Rc6 19 Kd2 f6 20 Bf4 Kf7 21

Rhe1 Be7 22 Kd3 Rhc8 23 f3 += Kokarev-Rozum, Kazan 2014.

17 h5 f6 18 Bf4 Kf7 19 Ke2 Be7

It comes as no surprise that the two games reaching this position have
been drawn, but White seemed marginally more comfortable in both:

a) 20 Kf3 Rhe8 21 a4 Bf8 22 Rhe1 Rxe1 23 Rxe1 Re8 24 Rxe8 Kxe8 25 Kg4
+= Zajić-Csonka, Serbian Team Ch, Kragujevac 2018.

b) 20 Rhe1 Rhe8 21 Kf3 Bf8 22 Kg4 g6 23 hxg6+ Kxg6 24 Kf3 +=

B.Ochsner-Hammer, Nordic Ch, Sastamala 2016.
10D: 14 0-0

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 Bf4 Nxe5 12 Bxe5 Qa5+ 13 c3 Nf6)

14 0-0

Kingside castling is usually a very natural move but somewhat less so

when your h-pawn has advanced two squares.

14...Be7 (D)

I find 14...Ng4 tempting, and so have some very strong players.

However, completing development has an appealing logic, so we shall
prefer the text- move.

White to play

15 Rfe1

Or:
a) The only game I could find with 15 a4 was immediately
drawn. Fortunately 15...0-0 seems fine for Black.

b) 15 Ne4 0-0-0 16 Qf3 and rather than 16...Qd5 (as in Mochalin-

Sreeshwan, Moscow 2020), Black should go for 16...Nxe4 17 Qxe4 Qd5
18 Qxd5 cxd5 19 Bxg7 Rh7 20 Be5 Bxh4 =.

c) 15 f4 Ng4 (15...0-0-0!?, with roughly level chances, is worth a look)

and now:

c1) 16 Qf3?! f5 17 Rfe1?! 0-0 18 h5 (Abdumalik-M.Muzychuk, Hengshui

rapid 2019) 18...Bh4 19 Re2 Bxg3 20 Qxg3 Qd5 =+.

c2) 16 Qe2 is more challenging. Black must enter the line 16...h5 17 Nxh5
Rxh5 18 Qxg4 g6 19 g3 0-0-0, which in general looks playable for Black.
So does 20 c4 Qd2 too, according to my computer, but I must admit that it
seems a difficult position to handle over the board (for both sides).

15...0-0 16 Ne4

Or 16 Nf1 Nd7 17 Bg3 Rad8 18 Qf3 Nf6 19 Ne3 Rfe8 20 Ng4 Qd5 21 Nxf6+
Bxf6 = Kanmazalp-B.Socko, Graz 2018.

16...Nd7 17 Bg3

Black seems comfortable here:

a) 17...Rad8 18 Qe2 (or 18 Nd6 Qb6 19 Nc4 Qa6 20 Qe2 Nb6 21 Nxb6 Qxb6
=) 18...Nf6 = Vujaković-Zelčić, Croatian Team Ch, Šibenik 2016.

b) 17...Qh5 18 Qd1 Qxd1 19 Raxd1 Rfd8 = may be even easier.

Lesson 11: Bareev Main Line: Menchik’s Line

By: SJ

Those who know how to break all the rules and work around those
specific guidelines reach the very top.
Evgeny Bareev

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8

Ne5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 (D)

Black to play

In theory this move weakens e4 and hinders White’s dark-squared bishop

by placing another pawn on a dark square. Yet it’s the clear main line.

Now Black has a number of interesting continuations. Actually, that

number itself is one of the variation’s strengths:

It increases White’s workload if he wants to be well-prepared for all

of Black’s options.
It gives Black a chance to choose the style of play and to steer the
game into the kind of positions he prefers.
It’s a safety-net, as Black will have back-up lines if his pet line is
refuted or rendered unattractive.

11...Be7
Black chooses to go directly to e7, rather than first provoking c3 with a
bishop check on b4. This is currently the hottest move and the main
subject of this lesson. In addition we’ll investigate 11...Ne7, which is almost
virgin soil and discussed in 11A, and the natural 11...Ngf6, which seems
the most robust move and is the subject of Lesson 12. Two very serious
alternatives that we’ll not consider in any detail are 11...c5, which in some
ways seems the most principled challenge but also rather hard for Black to
handle, and 11...Bb4+, which was Carlsen’s choice in his first world
championship match against Anand. After 12 c3 Be7 the big question is
whether having the pawn on c3 rather than c2 helps or hurts White.

Model Game 11

Ramesh – Ter Sahakyan

Internet blitz 2020

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8

Ne5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7! 11 f4 Be7 (D)

White to play
This immediate attack on the h-pawn looks relatively natural and
somewhat reduces White’s options. Nevertheless it took 71 years after
Vera Menchik introduced it in 1931 until it was rediscovered.

12 Bd2

12 Nh5 was the move that at first made me hesitate to recommend

11...Be7. Although 12...Ngf6! (12...Bxh4+?! 13 g3 Bf6 14 Bd2 seems too
risky) 13 Nxg7+ Kf8 14 Nxe6+ (after 14 Qg6? fxg6 15 Nxe6+ Kg8 16 Nxd8
Rxd8 17 Nxg6 Rh7 Black’s piece is much stronger than White’s immobile
pawns) 14...fxe6 is theoretically rather harmless, it’s a line you don’t want
to face unprepared. You should also be aware that there are a few forced
repetitions lurking in the corners. See 11B for the analytical details of the
direct 15 Ng6+ and 11C for the more ambitious 15 Bd2!.

12...Nxe5

The move-order 12...Bxh4?! is imprecise thanks to 13 Nxd7!,

when 13...Kxd7 is strongly met by 14 f5!.

13 fxe5 Bxh4

The only consistent move.

14 0-0-0 Bxg3 15 Qxg3 Ne7! (D)

Returning the pawn; Velička has tried 15...g6?! and 15...Rh7 too,
but honestly, that’s too much suffering for a pawn!
 White to play

We have reached a critical position. White must act resolutely or risk

struggling a pawn down with an inferior bishop against a strong knight.
Actually, this may be the time to say some kind words about the
Sicilian Defence.

Q: Isn’t that a bit late now, having already spent hours and hours
studying the Caro-Kann?

My point is that the Sicilian is a very popular defence with a lot of

complicated lines. And those hours, days or even months that your
opponent has spent studying the Dragon, the Najdorf and the Sveshnikov
are ones that he hasn’t devoted to the Caro-Kann. As a matter of fact, you
have every chance to be the better prepared if you reach this position as
Black. That will be a great help as everything is now set for a sharp race
between attacks on the enemy kings.

16 Qxg7

This may seem obvious, but there is at least one dangerous alternative:
a) 16 Qf2 is untested and looks strange but prepares to meet ...Nf5 with g4,
so maybe you should have a look at it with your favourite engine one rainy
afternoon.

b) 16 Bb4?! is one of the ideas Black could have prevented with

11...Bb4+ (as 12 c3 blocks the e1-a5 diagonal). However, after 16...Nf5
(D) Black has been doing very well:

White to play
b1) 17 Qb3 gives Black a
choice:

b11) 17...Qd5 18 c4? (Somborski-Velikić, Zrenjanin 2018; 18 Qxd5 cxd5

=+) 18...Qxg2 19 Rhg1 Qe4 –+.

b12) 17...Qg5+!? 18 Kb1 and now 18...0-0-0?! 19 Qa4 a6 20 Bc5 offers

White some activity, while 18...Qxg2! leads to sharp play after 19 d5 cxd5
20 Qa4+ Kd8 21 Rhf1 Qe2!, but objectively Black is doing very well.

b2) 17 Qf4 g5 and now:

b21) 18 Qf3 Qd5 19 Qxd5 cxd5 20 g4 (V.Nielsen-K.Rasmussen, Horsens

2017) 20...Ng3 21 Rh3 Ne4 =+.

b22) 18 Qf2 g4 19 c4 Qb6 20 Bc5 Qa6 21 b3 g3 22 Qd2 0-0-0 23 Kb2 b6 24

Bd6 Nxd6 25 exd6 Rxd6 26 Qf4 Rd7 =+ Wei Yi-Giri, Wijk aan Zee 2017.

16...Rg8 (D)

White to play

17 Qxh6

Or:

a) 17 Qh7, keeping contact with the pawn on f7 and the b1-h7 diagonal,
makes sense but is untested. Black seems to be fine after, e.g., 17...Qd5
18 Rdf1 (18 Kb1 0-0-0 19 Qxf7 Nf5 20 Bxh6 Rxg2 =) 18...0-0-0 19 Rxf7
Rh8 20 Qd3 Nf5 21 Qb3 Qxb3 22 axb3 Rxd4 =.

b) 17 Qf6 Nf5 18 Bxh6 Qxf6 19 exf6 Rd8 (19...0-0-0 is OK too) 20 Bf4

Nxd4 (20...Rxd4 =) 21 Rd2 c5 22 c3 Nc6 23 Bd6 += Hou Yifan-Kariakin,
Wijk aan Zee 2018.
17...Qd5

17...Rxg2? loses to 18 Qf6!; e.g., 18...Kd7 19 Bb4 Qe8 20 Rh7 Rg6 21

Qxf7 Qxf7 22 Rxf7 Re8 23 Rh1 +–.

18 Kb1

This is the main move. After 18 Bg5 Qxa2 (D), both kings are
somewhat exposed and there are quite a few possible repetitions on
the cards:

White to play

a) 19 Bxe7?? Rxg2 0-1 was the abrupt end of Brendel-J.Grant, Acqui

Terme seniors 2017.

b) After 19 Qf6 Nd5 20 Qf3, 20...Rxg5 21 Rh8+ Ke7 22 Rxa8 Qa1+ 23

Kd2 Qxb2 24 Ke1 c5 25 Qe4 is a notable draw by repetition. Black has
other 20th moves, but hardly any realistic winning chances.

c) 19 c3 Kd7! (not 19...Nd5? 20 Rhf1 Qa1+ 21 Kc2 Qa4+ 22 Kd2 +–

B.Savchenko-Triapishko, Kolomna rapid 2018; however, 19...Rg6 20
Qh8+ Rg8 21 Qh4 Kd7! seems to be an alternative path to equality) is
probably
equal:

c1) According to Stockfish 20 Bxe7 Rxg2 21 Rd2 Qa1+ 22 Kc2 Rxd2+

23 Kxd2 Qxb2+ should end with a repetition. The perhaps least obvious
of the lines it gives is 24 Kd3 Rg8 25 Bg5 c5! 26 dxc5 Qg2 27 Bd2 Qd5+
28 Kc2 Qa2+ 29 Kd1 Qb1+ 30 Bc1 Kc8 =.

c2) After 20 Qf6 Rae8 21 Rh7 Qa1+ 22 Kc2 Qa4+ 23 Kd2 Rg6 24 Qf4 Qa2
25 Rc1, as in Martinez Alcantara-V.Carneiro, South American Junior Ch,
Manta 2017, 25...Qxb2+ 26 Rc2 Qa3 27 Rxf7 Kc8 seems to hold without
too much difficulty; e.g., 28 Bxe7 Rxe7 29 g4 Kd7 30 Rf8 Kc7 31 g5 Kb6 =.

Does the danger of a draw trouble you? Well, that’s the main reason we
have a Lesson 12 on the demanding 11...Ngf6.

18...0-0-0 (D)

White to play

19 Bg5

Or:
a) 19 Qf4 Rxg2 20 Bc3 Nf5 21 Rh2? Rxh2 22 Qxh2 c5 –+
Narendran- Mahalakshmi, Moscow 2018.

b) 19 Qh5 Rxg2 20 Bg5 Rd7 21 Qh8+ Kc7 22 Bxe7 Rxe7 23 Qh3 Rd7 =.

c) 19 Qf6 is clearly the main move. The position seems balanced with
quite a few reasonable repetitions after 19...Rxg2 (19...Nf5 keeps more
play in the position but doesn’t appear to offer Black many winning
chances):

c1) 20 Qxe7 Rxd2 21 Qxd8+ Kxd8 22 Rh8+ Kc7 23 Rxd2 (Womacka-

Velička, Prague rapid 2019) and now 23...c5 24 c3 Qe4+ 25 Kc1 Qe1+
26 Kc2 Qe4+ is the simplest route to equality.

c2) 20 Bc3 Nf5 21 Qxf7 (21 Rh7 Ne3 22 Qxf7 Rd7 23 Qf8+ Rd8 24 Qe7
Rd7 25 Qe8+ Rd8 26 Qe7 Rd7 27 Qe8+ ½-½ Predojević-I.Sarić,
Bundesliga 2018/19) 21...Rg7 22 Qh5 Ng3 23 Qh6 Nf5 24 Qh5 Ng3 25
Qh6 Nf5 26 Qh5
= Grandelius-Lupulescu, Baku Olympiad 2016.

19...Nf5 20 Qh5 Rd7 (D)

White to play
A white pawn must fall, and the d-pawn is clearly the most important.

21 Bf6?

Letting Black activate his rook is the wrong way to give up the pawn. After
21 Rhe1 Qxg2 22 Rg1 Qd5 23 c3 Qe4+ 24 Ka1 Kc7 Black has a good
game, but creating any real winning chances will be very hard.

21...Rxg2 22 Qh8+ Kc7 23 Qa8?

This decentralizing move creates no real threats and actually loses by force.

23...Qe4 24 Rc1

All is set for the conclusion.

24...Rxc2! 25 Rxc2 Nxd4 26 Rc1 Nxc2 27 Qxa7

There was nothing to do, as a recapture is met by a mate in one. The

final combination has been seen a thousand times but is always
charming.

27...Na3++ 28 Ka1 Qb1+ 29 Rxb1 Nc2# (0-1)

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 (11...Be7 12 Nh5 Ngf6 13 Nxg7+ Kf8
14 Nxe6+ fxe6 15 Bd2!)

11A: 11...Ne7!?

11B: 15 Ng6+

11C: 15 Bd2!

White’s first 11 moves hint that he may be well prepared and looking for
a theoretical duel, so if you are not quite up and running with the
Capablanca
Variation yet, 11...Ne7!? (examined in 11A) may be a good choice. It
should have some surprise value and there is no obvious refutation.

Anyway, as soon as your games start popping up in the databases you

should be fully prepared, and that includes having a good look at the 13
Nxg7+ sacrifice of 11B and 11C. Memorizing too many lines may not be
productive, but ensuring that you know the attacking ideas and defensive
resources certainly is.

11A: 11...Ne7!?

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4)

11...Ne7!? (D)

White to play

This interesting option is little explored. It may transpose to other lines,

but there are also independent lines where it’s important that Black:
 has prevented f5.
can capture on e5 without the recapture fxe5 coming with an attack
on a knight on f6.
after ...c5 has the move ...Nc6.

12 Bd2

Continuing as if the knight had gone to f6. 12 Be3, aiming to discourage

...c5, is a more critical reaction. Then:

a) 12...Nd5?! is a position that can also arise from the 11...Ngf6 12 Be3
Nd5?! move-order, but that very rarely happens (Black usually gives a
check on the a5-e1 diagonal) and after 13 0-0-0 Black should avoid
13...Nb4?! due to the strong pawn sacrifice 14 Qe2! Nxa2+ 15 Kb1; e.g.,
15...Nb4 16 Bd2 Nxe5 17 dxe5 Qb6 18 Bxb4 Bxb4 19 Rd3 +/– Wegener-
Zelčić, Austrian Team Ch 2012/13.

b) 12...Nxe5?! forces White to recapture with his f-pawn before bringing

the knight to d5, but 13 fxe5 Nd5 14 0-0-0 (14 Bd2!?), as in Enders-
P.Schmidt, Bundesliga 1998/9, leaves Black unpleasantly short of
counterplay; e.g., 14...Nxe3 15 Qxe3 Qd5 16 Kb1 Be7.

c) The much more challenging 12...c5! (D) turns out to be playable (as
pointed out by our editor), so while it leads to crazy complications, this is
the move to play if you want to avoid suffering.
 White to play

Note that some of the tactics are unlikely to be found over-the-board by

an unprepared player – as White or Black. Then:

c1) 13 Nxd7 Qxd7 14 0-0-0 0-0-0 15 Qb3 Nd5 16 dxc5 Qc6 gives
Black enough for the pawn, as, e.g., 17 Ne4 f5 18 Nd6+ Bxd6 19 cxd6
Rxd6 20 Bd4 Nxf4 21 Be5 Rxd1+ 22 Rxd1 Nd5 is close to equal.

c2) 13 dxc5 Nxe5 14 fxe5 is perhaps the best try objectively. After
14...Qxd3 15 cxd3 Nd5 16 Bf2 b6 Black has positional compensation, but
will have to defend with care.

c3) 13 Qb5 is the sternest test of Black’s resources, but he can prove
equality in some amazing tactical lines: 13...Nd5 14 0-0-0 a6! (the most
clinical, though the analysis is very ‘computery’; 14...Nxe3 15 dxc5 a6 16
Qa4 Nd5 17 Rhe1 b5 18 cxb6 Nxb6 19 Qe4 Ba3! 20 bxa3 0-0 may also be
OK, meeting 21 f5! with 21...Qc7!) 15 Qxb7 Nxe3 (D) and here:
 White to play

c31) 16 Rd3 cxd4 17 Rxd4 Qc8 (17...Rb8 is fully OK too) 18 Qxd7+

Qxd7 19 Nxd7 and now 19...0-0-0!? is the nicest and most clear-cut path
to equality.

c32) 16 Nxd7 Qc8! 17 Qe4 (now follows a lengthy forced sequence of

engine moves with a bizarre but necessary king walk; not 17 Qf3? cxd4)
17...Nxd1 18 Nb6 Nf2 19 Qf3 Qb8! 20 Qc6+ Ke7 21 Re1! Ra7! 22 Nc8+
Kf6! 23 Nh5+
Kg6 evacuates the king with an improbable computer “0.00” evaluation
(White can’t force mate, but Black can’t keep the extra material and live!)
after 24 Nxa7 Kxh5! 25 Qf3+! Ng4 26 Nc6! Qc8! (not the end of the forcing
play, but we must cut off the lines at some point) or 24 f5+ Kh7! 25 Nxa7
Qb4 26 Rf1 (26 c3 Qc4) 26...Bd6!. There are many pitfalls for both sides
along the way in this treacherous line. You are unlikely to be tested in this
crazy variation, but it is good to know it works!

We now return to 12 Bd2 (D):

 Black to play
12...c
5

Even 12...Nxe5 13 fxe5 c5!? 14 Qb5+ Nc6 15 Qxb7 Qc8 may be

playable. Jonkman-Okkes, Hoogeveen 2003 was drawn after 16 Qxc8+,
but 16...Rxc8 17 dxc5 Bxc5 18 Ne4 Nxe5 19 0-0-0 Be7 20 Bc3 leaves
White with some pressure.

13 Nxd7

White has nothing better:

a) 13 dxc5 Nxe5 14 Qxd8+ Rxd8 15 fxe5 Nc6 16 0-0-0 Bxc5 =+

Zwahr- Nabaty, Dresden 2012.

b) After 13 0-0-0 Nxe5 14 fxe5 cxd4 15 Bc3 Nc6 16 Bxd4 Qd5

Black certainly isn’t worse.

13...Qxd7 14 dxc5 (D)

 Black to play
14...Qxd3

14...Nd5!? 15 b4 a5 could also be worth testing.

15 cxd3 0-0-0

15...Nc6 16 Rc1 0-0-0 17 Ke2 comes to the same.

16 Ke2 Nc6 17 Rac1 Rd5 18 b4

White is a pawn up in a rather messy position. Živković-Zelčić, Bošnjaci

2010 continued 18...Be7 19 Bc3 Nd4+ 20 Kf2 Rhd8 21 Rh3 g6 22 Ne4
Kb8
= but White had a few interesting alternatives. Instead
Podgaets/Karpov/Odessky give 18...Nd4+!? 19 Kf2 Nb5, which looks
close to equal.

11B: Knight Sacrifice: 15 Ng6+

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4)
11...Be7 12 Nh5 Ngf6 13 Nxg7+ Kf8 14 Nxe6+ fxe6 15 Ng6+

This direct move has never been tested in practical play and is a challenge
for Black only if he has to avoid a draw at all costs.

Q: Isn’t that a serious practical problem? I quite frequently must try to win
with Black against lower-rated opposition.

Well, if White has reached this position, knows there are drawing lines and
is able to demonstrate one of them, he probably isn’t that weak. You must
remember the number of possible deviations by Black on the route here,
and you only need to back up to 11A to find a less known and less forcing
alternative.

15...Kg7 (D)

White to play

All White realistically can hope for here is a draw by repetition.

16 h5
Other moves are less critical:

a) 16 f5?! Re8 17 0-0 (17 fxe6? Bb4+ wins for Black) 17...Bd6 is at least
a little better for Black. Actually after a typical line like 18 Bf4 Bxf4 19
Nxf4 exf5 20 Qxf5 Nf8, Black has a quite clear plus – the main challenge
is that pawnless king positions like this are a lot harder to handle for
humans than for computers.

b) Against 16 Nxh8, Black’s best is probably 16...Qa5+! 17 Bd2 Qf5. After

18 0-0-0 Qxd3 19 cxd3 Nd5 20 g4 Rxh8 Black seems to have some
realistic winning chances. That being said, the obvious 16...Qxh8 is also
safe enough.

16...Rg8 17 Bd2

Or 17 Rh3?! Kf7! 18 Bd2 Nf8 19 Ne5+ Ke8 –/+.

17...Nf8

It probably wasn’t this position Bent Larsen had in mind when he

quipped that, “With a knight on f8 you will never get mated”. But it
applies here anyway: Black isn’t going to be mated.

18 Ne5 Qd5 19 0-0-0

Black’s main threat was ...Qe4+, exchanging queens and reaching a

very promising semi-endgame.

19...Qxa2 (D)
 White to play

This messy-looking position is probably a dead draw. After the forced 20

c3, Black’s most direct path to a draw is 20...Ba3. There is also a number
of other playable moves that more or less force White to take a draw,
mostly by checking alternatingly on g3 and d3.

11C: Knight Sacrifice: 15 Bd2!

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Be7 12 Nh5 Ngf6 13 Nxg7+ Kf8
14 Nxe6+ fxe6)

15 Bd2!

Having seen that White’s knight and queen cannot drum up any real
threats without reinforcements, this somewhat slow move seems the only
try for a plus.

15...Rh7! (D)
 White to play

Black is a piece for two pawns up and well on the way to organizing
his camp.

16 Rh3

This prevents the black king from running to h8, as there will be a rook
check on g3 and then a knight check on g6. Other moves:

a) 16 Qg6?! Rg7 17 Qxh6 Kg8 =+.

b) 16 g4?! Kg8 17 Rg1 Nxe5 18 fxe5 Ne8 19 0-0-0 (19 g5 hxg5 20 hxg5?!
Ng7 21 0-0-0 Nf5 –+ Gharamian-Iturrizaga Bonelli, Lisbon 2017)
19...Bxh4 20 Qg6+ Ng7 =+.

c) 16 0-0-0 Kg8 17 Qh3 (against the untested 17 Rdf1?!, I like 17...h5,

planning to exchange White’s e5-knight and placing my own on g4)
17...Nf8 18 f5 (D) leads to a position where there are many sharp options.
 Black to play

A race of opposite-wing attacks can easily occur if both sides wish (e.g.,
18...exf5 19 Qxf5 Qd5 20 Rde1 Qxa2 21 c4 a5 22 Rh3 Rg7 23 Ree3 Qa1+
24 Kc2 Qa4+ was sharp but ultimately equal in Ivić-Svane, Batumi
Olympiad 2018). But right now the game is rather balanced, with a wide
choice of relatively safe options for both players, and many draws by
repetition are possible. One sensible move here is 18...c5 (with ...Qd5
ideas in some lines; the immediate 18...Qd5 is riskier due to 19 c4 Qxd4 20
Qg3+ Rg7 21 Qxg7+ Kxg7 22 Bxh6+), when White may have nothing
objectively better than 19 Qg3+ Rg7 20 Qh3 (threatening Bxh6), when
Black can repeat with 20...Rh7 or roll the dice with 20...Qxd4!? 21 Rhe1
Ne4.

We now return to 16 Rh3 (D):

 Black to play
16...Rg7

This is the only move tested in practice. However, a pleasant aspect of this
position is that Black has a number of playable moves. In addition to the
text- move, 16...Nb6, 16...c5 and 16...Qe8 also seem to lead to equality –
at least Stockfish thinks so on normal search depths.

17 f5!

White cannot afford the slow 17 0-0-0? Nxe5 18 fxe5 Ng4, while after 17
Ng6+? Kf7 18 f5 exf5 19 h5 Nf8 Black is better: 20 Nh4 (20 Qxf5? Nxg6 21
hxg6+ Rxg6 –+) 20...Ne4 21 Nxf5 Nxd2 22 Nxg7 Kxg7 23 Qxd2 Ne6 =+.

17...Nxe5

Maybe 17...Rxg2 is a safer path to full equality:

a) 18 Bxh6+ Ke8 19 Kf1 Nxe5 20 dxe5 Qxd3+ 21 cxd3 Rxb2 22 exf6 Bxf6
=.
b) 18 Ng6+ Kg7 19 Rg3 Rxg3 20 Qxg3 Kh7 21 Qf4 Bf8 22 Nxf8+ Qxf8 23
fxe6 Nb6 and again chances are roughly equal; e.g., 24 0-0-0 Qg7 25
Rf1 Nbd5 26 Qf3 b5 27 b3 Qg6 28 h5 Qxh5 29 c4 Qxf3 30 Rxf3 bxc4 31
bxc4 Re8 32 cxd5 Nxd5 =.

18 dxe5 Qxd3 19 Rxd3 Rxg2 (D)

White to play
This seems
equal:

a) 20 exf6 Bxf6 doesn’t look critical: 21 0-0-0 (21 Bxh6+ Kg8 22 0-0-0 exf5
23 Rd7 Rg6 =) 21...exf5 22 Bxh6+ Kg8 was equal in G.Jones-Gumularz,
London 2017, even if White could have posed some challenges with 23
Rd7 Rg6 24 Be3 Re8 25 h5 (25 Bxa7 Rg7 26 Rxg7+ Kxg7 27 Bd4 f4 =)
25...Rg7 26 Rxg7+ Bxg7 27 Bxa7 Bh6+ 28 Kb1 f4 29 c3 Re5 =.

b) 20 Bxh6+ Kf7 is more complicated: 21 exf6 Rg1+ 22 Kd2 Rxa1 23 fxe7

Re8 24 f6 Rxa2 = or 21 Kf1 Rh2 22 exf6 Rh1+ 23 Ke2 Rxa1 24 fxe7 Re8
25 Bg5 Rxa2 26 Rb3 exf5 27 Rxb7 =.
Lesson 12: Bareev Main Line: 11...Ngf6

By: SJ

I am acutely conscious, from vast experience in opens, that guys around,

say, 2100 or more can definitely play chess and that one often has to work
very hard to beat them.

Nigel Short

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4

Having seen how the sharp lines of the previous lesson may easily lead to
repetitions and draws without much real fight, we shall in this lesson focus
on a continuation that scores worse than 11...c5 as well as the bishop
moves to e7 and b4 but keeps more play in the position.

Q: That sounds risky. Why does Black need to compromise?

If he wants to play for a win at all costs, he must avoid some of the
more forcing drawing lines of the previous lesson. The way to do that is
to enter positions where both sides have some freedom of choice and
may fight on after one or even two inaccurate moves.

11...Ngf6 (D)
 White to play

There are moves that just cannot be wrong, and this looks like one of them!
A knight is developed to a natural square while Black otherwise stays
flexible.

Model Game 12

Topalov – Dreev

FIDE Knockout rapid, New Delhi 2000

1 e4 c6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6

Alexei Dreev is one of the great experts on the Bareev

Variation. 8 Ne5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12

Bd2 (D) White obviously must prepare queenside castling.

 Black to play
12...Be
7

This modest developing move looks very sensible and is the most
common option. However, in order to evaluate it fairly, it should be
compared to the principled 12...c5. Then 13 Ng6? fails because of 13...c4!
14 Qxc4 fxg6 and in D.Gross-Ganaus, Austrian Team Ch 1999/00, Black
had a very clear advantage. Therefore 13 0-0-0 is critical and now
13...cxd4? 14 Ng6! gives White a plus; e.g., 14...Rg8 15 Nxf8 Nxf8 16 Bb4
+/– Hector-Rausis, Hamburg 2000. Instead 13...Be7 is playable but I
couldn’t find equality or promising counterplay after 14 d5! Nxe5 15 fxe5.
Black should play 15...c4! (and not 15...Nxd5? 16 Nh5! +/–) but after 16
Qxc4 Qxd5 17 Qa4+ Qd7 18 Qb3 Nd5, Black’s position is still difficult.

Black’s other alternative, 12...Bd6 is quite solid and may be easier for
Black to play, but after 13 0-0-0 Qc7 14 Ne4 (which hasn’t been popular
lately) 14...Nxe4 15 Qxe4, kingside castling seems risky and after 15...Nf6
16 Qe2 0-0-0, Black is rather short of counterplay.

13 0-0-0 (D)
Stockfish quite likes 13 Qf3 but after 13...0-0 White doesn’t seem to
have anything better than 14 0-0-0, transposing, and 13...Nb6!? is an
alternative.

Black to play
13...0-0

With opposite-side castling, play becomes very concrete – in theory it’s a

computer’s paradise. When you actually look into the lines of this position,
what the engine actually offers is frequently a list of lines evaluated as
‘0.00’, usually indicating that it has found repetitions which it considers best
play for both sides. Sometimes this evaluation changes when you follow
the lines further, but usually it’s correct and your main challenge is to
organize the material in a way that makes it possible to remember the often
complicated tactical lines.

Of the alternatives, 13...c5?! scores better, but after 14 d5 I don’t know

how Black should equalize. 13...Qc7 is probably playable but I didn’t find
14 Rhe1 0-0 15 f5! attractive for Black.

14 Qf3
This is clearly the main move; it’s the most popular and it scores best
(yes, there is a connection). Second in popularity is 14 Qe2, which is
discussed in 12B. For White’s minor 14th moves, including the slow but
flexible 14 Kb1, see 12A.

14...Qc7! (D)

White to play

Black prepares a quick ...c5.

15 c4

White hopes to meet ...c5 with d5. For some alternatives, including 15
Rhe1 and 15 Kb1, see 12C.

Note that with the text-move, the line merges with some where Black
had inserted an earlier ...Bb4+ and met c3 with ...Be7. That’s why you
will encounter games where the positions we examine from this point
on are reached with the move-numbers increased by 1.

15...a5
White’s dark-squared bishop isn’t ideally placed, and Black wants to
prevent Bc3, which can now be met by ...Bb4 or ...b5 depending on the
timing. If you want to see a demonstration of Black’s problems after
15...c5?! 16 d5!, you may look up the instructive game Movsesian-
Iordachescu, FIDE Knockout, New Delhi 2000 where White was probably
already winning after 16...Rae8 17 Rhe1 exd5 18 cxd5 Bd6 19 Bc3.

While the text-move is the most popular and has been used at top level,
the untried 15...Rfe8!? (D) is an interesting and logical alternative.

White to play

The point is to remain flexible: by giving the e7-bishop more options and
putting the rook on the e-file, Black provides tactical support to ...c5
ideas. Thus:

a) 16 Kb1 is met by 16...c5! 17 d5 Bd6! 18 Nxd7 (18 dxe6?! Nxe5 19 fxe5

Bxe5 20 exf7+ Qxf7 is great for Black) 18...Qxd7!, when the key point is
19 Bc3 e5 (supported by rook and bishop!), while 19 dxe6 Qxe6 20 Bc3
Bf8 (a harmonious arrangement) is also roughly level.

b) 16 Ne2?! is well answered with 16...c5, when ...Rfe8 is clearly a more

useful move than ...a5 (in comparison to the 16 Ne2 line mentioned in the
next note). White has little better than 17 g4 cxd4 18 g5 Nxe5 19 fxe5,
when 19...Qxc4+ 20 Kb1 Qd5 is safe, but 19...Nd7 20 Kb1 h5 gives White
more problems.

c) After 16 Rhe1 there are several options besides 16...a5 (which now has
the interesting point 17 a3 b5! 18 Nxc6?! {18 Qxc6 Qa7} 18...Bxa3! 19
bxa3 bxc4! with an attack). For instance, 16...Bf8!? prepares ...c5, which is
the answer to 17 Kb1, as after 17...c5! 18 d5 exd5 19 cxd5 Rad8 White’s
centre is creaking (though he maintains a dynamic equilibrium after 20 Bc3
Nb6 21 Ng4 Nfxd5 22 Nxh6+!). 17 Bc3 invites ...b5 ideas, possibly best
prepared with 17...a5 as 18 a3?! b5! works well. 18 Kb1 b5 (18...a4!?) 19
c5 (19 f5 Nb6) 19...Nd5 may be compared with line ‘d’.

d) 16 Bc3 reinforces the d5 advance if Black persists with the ...c5 plan,
but is inflexible and invites Black to hit back instead with 16...b5 (D).

White to play

The tactics work out OK since 17 Nxc6? (17 Qxc6?! Qxc6 18 Nxc6 Bd6
is similar, while 17 f5?! is answered with 17...b4!) 17...Bd6 exploits the
loose
f4-pawn and 18 Ne5 bxc4 puts the focus on the white king. White has
nothing better than 17 c5, a move clearly welcomed more by the black
knights than the white bishop, and ruling out the d5 thrust. Still, White has
space and kingside plans, and a tense pitched battle will result; e.g.,
17...Nd5 (shielding c6 while preparing ...f6 or ...f5) 18 Bd2 (18 Nh5 N7f6
19 Nxf6+ Bxf6 =) 18...f5! 19 Nh5 (19 Rde1 hopes to put pressure on e6,
but 19...Nxe5 forces 20 fxe5 and then 20...Qd8! disrupts White’s kingside
plans) 19...N7f6 (forced as White planned g4) 20 Rh3 (20 Nxf6+ Bxf6 21
g4 Bxe5 22 fxe5 Qf7 =) 20...Kh7 and now 21 Rg3 Rg8 22 Rg6 Raf8 23
Rg1 threatens 24 g4 Nxh5 25 Rxe6!, which 23...Qc8! parries, while after
21 Re1 Rf8 22 Rg3 Nxh5 23 Qxh5 Bf6 Black’s defences hold.

We now return to 15...a5 (D):

White to play
16 Kb1

This move is useful in general and specifically prepares Ne2 followed by

g4, where in at least one line it’s important that Black doesn’t have the
resource
...Qxc4+. The more direct 16 Ne2, preparing g4, is suggested as critical by
Podgaets/Karpov/Odessky. It prepares to advance the g-pawn and takes
some of the point out of ...Bb4 but still appears untested. As it’s not entirely
clear what’s Black’s best, we look at 16...c5 in 12D and 16...h5 in 12E.

The otherwise natural 16 Rhe1 allows 16...Bb4, when Black may be

satisfied. Schmittdiel-Lauber, Bundesliga 2011/12 continued 17 f5 Rfe8 18
a3 Bxd2+ 19 Rxd2 Rad8 20 Rde2 Nf8 =.

16...Rad8?!

Again, this is the ‘main line’, played in a couple of top-level games, but it is
legitimate to ask if this is really the right rook. 16...Rfd8!? (D) makes a lot of
sense (and is one of my two recommended repertoire options, along with
the earlier 15...Rfe8!?), as we have already seen that providing the f8-
square for the bishop has its uses, while having played ...a5, the queen’s
rook can expect to have a role on the a-file.

White to play
Then:

a) 17 Bc3 Bb4 = is a standard theme in this whole line.

b) 17 Bc1 a4 (the immediate 17...c5 18 d5 Ra6 is possible too) 18 a3 c5 19
d5 Ra6 20 Rhe1 Rb6 shows the rook becoming useful; e.g., 21 Nd3 Nf8 22
Bd2 Rb3 23 Bc3 b5 with chaos and counterplay.

c) 17 a3 is met by 17...c5! (compare the line 16...Rad8?! 17 a3! below) 18

d5 exd5 19 Nf5 Bf8 (without this move Black would be lost!) with unclear
play after 20 g4 g6.

d) 17 Ne2 is met by the standard 17...h5 intending ...c5.

e) 17 h5 Bb4 (17...Bf8?! 18 Bc3 offers White more) 18 Bc1 c5!? (Black is

willing to sacrifice a bishop) 19 d5 (19 a3 cxd4 20 axb4 axb4 gives Black
the a-file play we mentioned after 21 Rxd4 Qb6) 19...exd5 20 a3 (20 Nf5
Nxe5 21 fxe5 Qxe5 22 Rd3 {22 a3?! dxc4} 22...Qe4 with unclear play)
20...Nxe5 (20...dxc4!?) 21 fxe5 Qxe5 22 cxd5 Rxd5 is messy but roughly
balanced.

f) 17 Rhg1 (intending Ne2 and g4) 17...Nh7!? (hitting h4 puts a spoke

in White’s plans) 18 Be3 (18 Ne2 Nxe5 19 dxe5 and now the non-
greedy 19...Rd7 is equal) 18...Nxe5 19 fxe5 Bxh4 20 Ne4 with unclear
play.

g) 17 Rhf1 Bb4 (this direct approach seems best) 18 Bc1 a4 (D).

 White to play

19 Ne2 (or 19 a3 Bf8 intending ...c5) 19...h5 (another standard theme) 20

c5 (20 a3 Bf8 21 g4 Nxg4 22 Nxg4 hxg4 23 Qxg4 f5 followed by ...Nf6
keeps Black’s defences sturdy) 20...a3 21 b3 (or 21 bxa3 Ba5 22 Bb2
Rab8 intending ...b6 with counterplay; e.g., 23 Ng3 b6! 24 Nxc6 bxc5 25
Nxd8 Nd5!) 21...Ng4!? (21...g6 is safer) 22 Nxg4 hxg4 23 Qxg4 Nf6 24
Qf3 Nd5 with positional compensation and counterplay with ...b6.

17 Bc1?!

Black’s idea was 17 Bc3?! Bb4! with equal chances. The more active try
17 Ne2?! was analysed in some detail by Podgaets/Karpov/Odessky, but
overall Black seems fine in the sharp lines after both 17...h5 and 17...c5.

However, the untried 17 a3! (D) looks very much like the critical test.

Black to play

White prevents ...Bb4 while retaining Bc3 ideas and leaves it rather
unclear where Black’s counterplay is coming from. Unless Black can
demonstrate a satisfactory reply, you may wish to take a close look at
Black’s alternatives
on moves 15 and 16. The basic problem for Black is that 17...c5? is no
good because of 18 d5! exd5 19 Nf5, and there doesn’t appear to be a
way to prepare ...c5 that doesn’t allow a strong response. Detailed
analysis suggests Black is in real trouble here:

a) 17...Qb6!? hopes to exploit White’s previous move to put pressure on

d4, b2 and a3, but after 18 f5! nothing works for Black: 18...c5 (18...Qxd4?
loses outright to 19 Nxd7! Rxd7 20 Bg5!) 19 Rhe1 Nh7 20 Ng4 with a huge
attack; e.g., 20...Bf6 (20...cxd4 21 Nh5) 21 Bxh6 Bxd4 22 Bc1 +/–. Maybe
this is murky enough for Black to fight on in practice, but objectively White
is in great shape to smash through on the kingside.

b) 17...Ne8 18 Ne4 Nd6 (desperately trying just to get a solid game;

18...c5? fails to 19 d5! Nxe5 20 fxe5 exd5 21 Nc3 and Nd5) 19 Nxd6
Bxd6 20 g4 Nxe5 21 dxe5 Bc5 22 g5 and Black has nothing better than
simply giving a pawn away with 22...h5 23 Qxh5 +/–, when an attempt to
set up a fortress is unlikely to succeed.

c) 17...Rfe8 brings the ray of hope that 18 Bc3 (stamping out 18...c5? due
to 19 d5) 18...Rc8 intends counterplay with ...b5. However, Stockfish
provides the stronger 18 Rhf1! (D) and demolishes all attempts to defend
the black position; e.g.:
 Black to play

c1) 18...Qb6 19 f5 Qxd4 20 Nxf7! Ne5 21 Nxe5 Qxe5 22 Rfe1 Qc5 23

fxe6 Qxc4 24 Nf5 Rd3 25 Qe2 Qb5 (25...Bxa3? 26 Bc1 +–) 26 g4! Nd5
(26...Bxa3? 27 Bc3! +–) 27 Nxe7+ followed by 28 g5 +/–.

c2) 18...Bf8 19 Bc3 Rc8 (intending ...b5; 19...c5 20 d5 exd5 21 cxd5 Nb6
22 Bxa5 c4 23 d6!) 20 f5! b5 21 Nxd7 Qxd7 22 d5 e5 (22...exd5? 23 Bxf6
gxf6 24 Nh5 Qd6 25 c5!! +–) 23 dxc6 Qxc6 24 Qxc6 Rxc6 25 cxb5 Rc5 26
a4 Rc4 27 Rfe1 Rxa4 28 Rxe5 Rb8 +/–. It is possible (as also in ‘c1’) that
with brilliant defence Black can reach an endgame that might be drawable
with great difficulty.

17...Rfe8 18 Ne2 (D)

18 a3 shouldn’t worry Black much, as the logical 18...c5 19 d5 Bd6 20

Nxd7 Qxd7 21 dxe6 (21 Nh5 Nxh5 22 Qxh5 Bf8 23 Qf3 exd5 24 Rxd5 Qe7
=) 21...Qxe6 22 Rhe1 Qxc4 23 Rxe8+ Nxe8 24 Qxb7 Rb8 looks OK for
Black.
 Black to play
18...c5?
!

This looks very thematic but is still inaccurate. Considering the earlier
inaccuracies it has minimal theoretical relevance, but 18...h5! is probably
OK for Black. One important line runs 19 Ng3 (after 19 c5 Bf8 20 Nc3 the
freeing 20...b6!? should be considered, with the surprising point that 21
Qxc6 Qxc6 22 Nxc6 Ra8! gives Black compensation) 19...g6 20 Rhe1 Bf8!
21 Bd2 Bg7, when the kingside has stabilized and it’s time to see whether
Black can get something going on the queenside. After 22 c5 a4, both 23
Nc4 b6! and 23 Rc1 Nxe5 24 fxe5 Ng4 look OK for Black.

19 g4!

I’m not sure this is winning, but I’m convinced that Black now has a difficult
defensive task.

19...cxd4 20 g5 Nxe5 21 fxe5 Nh7 22 gxh6 Qxe5 23 hxg7

I don’t believe this position is the most critical for the variation, but it’s quite
interesting. The pawn-structure is extremely unbalanced and while both
sides
have some pluses and minuses when it comes to their pieces, there seem
to be no immediate mating threats.

23...f5?!

Having spent some time analysing the position with Stockfish (which gives
unusually volatile evaluations), it seems that Black may be able to hold
after 23...Qf6 provided that he defends with computer-like precision.

24 Ng3?! Qc7 25 Qe2 Bc5 26 Nh5?! e5 27 Ng3

This may not have been the best way to nurse White’s plus.

27...Qf7 28 Qf1 f4 29 Ne4 Qf5

Stockfish evaluates 29...b6 as close to equal.

30 Qd3 Be7?! 31 Rdg1 Rd6 32 c5 Re6 33 h5?

Play isn’t very precise at this stage of the game (as we are a long way from
the opening, most of the errors are merely labelled with question marks) so
I assume there was some time-pressure. Here Stockfish suggests 33 Ka1
with a clear plus. One point seems to be that in some lines White can
follow up with Qb5, attacking a rook on e8 and then improve his knight
position.

33...Bg5? 34 Ka1? Bh6? 35 Qf3 Ng5 36 Nxg5 Bxg5 37 Qd5 Bh6 (D)
 White to play
38
Rg6?

No matter how unfair to the human players under time-pressure, I would

like to point out the surprising computer move 38 Bxf4!, when Black loses
in all variations:

a) 38...Bxf4 39 h6 Qh7 40 Qxb7 and it’s understandable that Black

is helpless despite his extra piece.

b) It’s harder to see 38...Qxf4 39 Rf1 Qg5 40 Rhg1 Qe7 41 Rg6 Bf4, when
White’s only winning move is the relatively calm 42 h6; e.g., 42...Qf7 43
h7+! Kxh7 44 g8Q+! Rxg8 45 Rxe6 and Black is completely lost even if it
may not be obvious to a human eye. A typical line is 45...Rg6 46 Qh1+ Kg7
47 Rxe5 Rh6 48 Qf3 Rf6 49 Rd5 and even if you don’t see how the
computer picks up the black queen in all lines, it’s easy to believe that
White is winning.

38...Qf7 39 Qe4 Rc8 40 Rhg1 Rxc5 41 Bd2 Rec6 42 a3 Rc8??

A tragic blunder in an equal position.

43 Rxh6 1-0

(Major parts of the analysis of this game’s opening stage –

particularly moves 15-17 – were provided by our editor, Graham
Burgess.)

Theory Magnifiers

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12 Bd2 Be7 13 0-0-0 0-0 (14
Qf3 Qc7 15 c4 a5 16 Kb1 Rfd8!?)

12A: (14 Kb1)

12B: 14 Qe2!?

12C: (15 Rhe1)

12D: 16 Ne2 c5

12E: 16 Ne2 h5

We have reached the centre of the theoretical debate. If you have

followed the critical lines up to this point, you should be prepared for a
theoretical discussion. My advice would be to start with the main line
and work backwards.

Note that if you choose to go with the 15...Rfe8!? option proposed in our
model game (rather than 15...a5 and meeting 16 Kb1 with 16...Rfd8!?),
then you need not study 12D or 12E in order to complete your repertoire.

12A: 14th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8

Ne5 Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6)

12 Bd2 Be7 13 0-0-0 0-0 (D)

 White to play
14 Kb1

A flexible but somewhat slow move. At least one of the alternatives

also requires a precise reply:

a) 14 Ne4 Nxe4 15 Qxe4 Nf6 ½-½ Luther-Kachiani

Gersinska, Recklinghausen 1998.

b) After 14 Ne2, planning g4, the thematic 14...c5! seems sufficient for
equality: 15 g4 Nxe5 16 fxe5 Nxg4 17 Qf3 h5 18 Nf4 g6 19 Rhg1 Qxd4
20 Nxh5 Nxe5 (20...gxh5? 21 Bf4 +–) 21 Qe2 Nc4 22 Rxg6+ fxg6 23
Qxe6+ Rf7 24 Qxg6+ Rg7 25 Qe6+ Rf7 26 Qg6+ with a repetition.

c) 14 Rhe1 c5 also seems fine for Black as 15 Nxd7?! Qxd7 16 f5 cxd4 17

fxe6 fxe6 is harmless: 18 Qg6 (18 Qb3 Nd5 19 Re4 Bc5 20 Rg4?! Kh8 21
Nh5 Qf7 22 Nf4 Nxf4 23 Rxf4 Qe7 =+ Feygin-Dautov, Bundesliga 2000/1)
and now 18...Bd6! is good for Black, since after 19 Ne4 Nxe4 20 Rxe4 e5
21 Rg4 (given by Dautov as equal) 21...Rf2 22 Bxh6 Bf8 Black’s
counterattack takes over.
14...c5

This is our programmed reaction to quiet moves.

15 Be3 (D)

15 dxc5, as played in Topalov-Dreev, FIDE Knockout rapid, New Delhi

2000, is probably inaccurate due to 15...Nxe5! 16 fxe5 Qxd3 17 cxd3 Ng4
18 Ne4 Nxe5, when Black has a small plus.

Black to play
Now:

a) 15...Nxe5 16 fxe5 Nd5 seems fully playable, but there are some
tactics lurking below the surface:

a1) 17 Bf2 Nb4 (17...cxd4 18 Bxd4 Nb4 19 Qe2 Qd5 {19...Qc7!?} 20 c4 Qa5
21 a3 += R.Almeida-Tazelaar, corr. 2013) 18 Qb3 c4 19 Qxc4 Rc8 20 Qb3
Rxc2 21 Be1 Rxg2 22 Bxb4 Bxb4 is complicated but possibly roughly
equal.

a2) 17 Bxh6 has the point that 17...gxh6? 18 Nh5! Kh8 19 c4! gives White a
decisive central pawn-roller; e.g., 19...Nb4 20 Qe3 Kh7 21 a3 Nc6 22 d5.
Black’s only move is 17...c4! 18 Qxc4 (18 Qd2? c3!) 18...gxh6 19 Qd3
b5!, the point being that White’s c-pawn is restrained. Then 20 Nh5 Kh8
21 Qd2 Kh7 22 Qd3+ is a draw by repetition.

b) 15...Qc7!? keeps the game more tense. A sample line runs 16 Nxd7 (16
Qe2 Rad8 17 Bf2 Bd6 18 c4 a5 19 Rd3 cxd4 20 Bxd4 Nc5 =+; 16 Ne4
Nd5 17 Ng5 N7f6 18 dxc5 Nxe3 19 Qxe3 Bxc5 20 Qe2 Rad8 21 Rxd8
Qxd8 =) 16...Nxd7 (16...Qxd7 is playable too) 17 Nh5 (17 d5!?, as played
in Jansa- Möhn, Pardubice 2015, is tricky, but 17...exd5 18 Nf5 Rfe8 19
g4 Nf6 provides chances for both sides) 17...Rad8 18 d5 exd5 19 Qc3 Nf6
20 Nxf6+ Bxf6 21 Qxc5 Qd7 =.

12B: 14 Qe2!?

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12 Bd2 Be7 13 0-0-0 0-0)

14 Qe2!?

One subtle difference between this move and the main line 14 Qf3 is
that here the move Rh3 has more point.

14...Qc7 (D)

This preparatory move is probably best even if some very strong players
have tried the immediate 14...c5.
 White to play
15 Kb1!?

It should come as no surprise that this position may arise from various
lines with an earlier Kb1. Black seems to be OK after other moves:

a) 15 Rhe1 c5 16 dxc5?! Bxc5 (16...Nxc5!?) 17 Nxd7 Qxd7 18 Kb1 Qc6

= Wan Yunguo-Manush, Bangkok 2017.

b) 15 f5 Bd6 16 fxe6 Nxe5 17 dxe5 Bxe5 18 exf7+ Rxf7 19 Ne4 (19 Nf5?!,
as in Sarkar-P.Taylor, London 2000, can be met by 19...Re8 –/+) 19...Rd8
=.

c) As hinted, 15 Rh3 is a dangerous try. White’s idea is to meet ...c5 with

d5. If Black then captures with his knight, White will have a lot of piece
pressure after Nh5. Black probably should prevent this plan with
15...Rad8!, taking play to the centre files:

c1) For 16 Kb1, see note ‘c’ after Black’s 15th move below.

c2) 16 f5 Bd6 17 Nxd7 (17 Nxf7 Rxf7 18 fxe6 Re7 19 Nf5 Nf8 20 Nxe7+
Qxe7 21 Re1 Re8 =+) 17...Qxd7 18 fxe6 Qxe6 19 Qxe6 fxe6 20 Re1
Nd5!
and Black is OK since 21 Rxe6? loses to 21...Nf4 22 Bxf4 Bxf4+ 23
Kd1 Rxd4+ 24 Ke1 Bd2+ 25 Ke2 Bc1.

15...Rad8!? (D)

White to play

The idea is to keep one rook to cover f7 and put the other one on the d-file
to ensure that ...c5 isn’t met by d5. Now:

a) 16 c4 c5 17 Be3 a6 18 Qf3?! cxd4 19 Bxd4 Nxe5 20 fxe5 Nd7 21 Qe2

Rc8 22 b3 b5 23 cxb5 axb5 24 Bb2 b4 –/+ T.Bueno-Tinjaca Ramirez, corr.
2000.

b) 16 f5 Bd6 17 Nxd7 Qxd7 18 Bg5 Bxg3 19 Bxf6 gxf6 20 Qg4+ Kh7 21

fxe6 fxe6 22 Qxg3 =.

c) 16 Rh3 c5! was Black’s reason to put his rook on the d-file: 17 f5 (not
17 d5? Nxd5 18 Nh5, when 18...N5f6! rebuffs the attack thanks to the pins
after 19 Rg3? Nxh5 20 Qxh5 Nxe5 21 fxe5 Bf6! –+) 17...cxd4 18 Nxf7 Rc8
19 Nxh6+ gxh6 20 Bxh6 e5 21 Bxf8 Kxf8 =+.

d) After 16 Nf1 I think Black should avoid 16...h5, as in O.Bronstein-

Macovei, European Under-16 Ch, Prague 2016, as White’s attack with
g4 will be quick after 17 Nh2!. Instead it seems Black should try to
generate counterplay with 16...Nd5:

d1) The straightforward 17 g4 Nxe5 18 fxe5 c5 isn’t clear; e.g., 19 g5 h5

20 Ng3 cxd4 21 Nxh5 Ne3 22 Bxe3 Qxe5 and chances seem equal.

d2) 17 c4 N5b6 18 g4 doesn’t really improve; e.g., 18...Nxe5 19 dxe5 Na4

20 Rh3 Qd7 and Black has sufficient counterplay.

12C: 15th-Move Alternatives

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12 Bd2 Be7 13 0-0-0 0-0)

14 Qf3 Qc7 15 Rhe1

Placing the rooks on the central files is fairly natural but not consistent with
a direct attack against Black’s castled king. Unlike our main line (15 c4), it
fails to prepare a good reply to the ...c5 break. The same goes for another
normal-looking move, 15 Kb1 (D), which is flexible but slow:
 Black to play

a) 15...Rad8 looks natural but after 16 c4 Black must be careful since

16...a5?! transposes to Model Game 12, where 17 a3! is hard to answer.
16...Rfe8 may be worth a look, meeting 17 Bc3 with 17...Bf8, still
preparing
...c5.

b) There’s no good reason to avoid 15...c5!, apart from a desire to avoid

simplification. 16 Nxd7 Qxd7 17 dxc5 Qc6 18 Qxc6 bxc6 19 Bc3 Nd5 20
Be5 (Mielczarek-Baranowski, Polish corr. Ch 2011) 20...Ne3 21 Rd3
Bxc5 22 Nh5 f6 23 Bd6 Bxd6 24 Rxd6 Rad8 25 Rxe6 Nxg2 =.

15...c5 (D)

After 15...Rac8 16 c4, White is ready to meet ...c5 with d5.

White to play
16 Nxd7 Qxd7 17 dxc5
Rac8!

The move-order 17...Qa4?! 18 Kb1 Rac8 gives White the extra option
19 Qxb7!, while 17...Qd5 18 Qxd5 Nxd5 19 Ne4 Bxh4 20 g3 Be7 21 c4
Nf6
(J.Krebs-Priebe, German corr. Ch 2006) 22 Nxf6+ Bxf6 23 Be3 will be
a tricky ending for Black, as White’s majority is the more dangerous.

18 Bc3

18 Kb1 allows various ideas, including 18...Rxc5 19 Bc3 Qa4, transposing.

18...Qa4

Compared to playing this a move earlier, the queen now hits an extra target.

19 Kb1 Rxc5

This looks OK for Black since 20 Qxb7 is met by 20...Rxc3! 21 bxc3

Qxf4 with dynamic equality thanks to White’s weak pawns and exposed
king.

12D: 16 Ne2 c5

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12 Bd2 Be7 13 0-0-0 0-0 14
Qf3 Qc7)

15 c4 a5 16 Ne2 c5 (D)
 White to play

Black immediately starts his queenside counterattack, rather than trying

to slow down White’s kingside play.

17 Bc3

This looks sensible. More direct moves seem to backfire:

a) 17 d5?! Nxe5 18 fxe5 Qxe5 19 Nc3 Qd4 20 Bxh6 Qxc4 21 d6 Bd8 =+.

b) 17 g4?! cxd4 18 g5 Nxe5! 19 fxe5 Nd7 20 gxh6 (20 Kb1 h5 21 Bf4

Rad8 22 Ng3 Nxe5 23 Bxe5? Qxe5 24 Nxh5 d3 –/+) 20...g6!? (20...Nxe5
21 Qg3
g6 22 Nxd4 Qxc4+ 23 Bc3 is given as hard to evaluate by
Podgaets/Karpov/Odessky; nothing is made clearer by the natural 23...Nc6
24 Nxc6 Qxc6) and Black seems better: 21 h5 (21 Nxd4 Qxc4+ 22 Kb1?!
Qxd4 23 h7+ Kxh7 24 h5 g5 25 Bxg5 Nxe5 26 Qe2 Qg4 27 Qc2+ f5 28
Bxe7 Rf7
–/+) 21...Nxe5 22 Qe4 Nxc4 23 hxg6 (23 Kb1? g5! –/+ blocks
White’s avenues of attack) 23...Nd6+ 24 Qc2 Qxc2+ 25 Kxc2 fxg6
=+.

17...cxd4
17...Bd6 also seems to work. 18 dxc5 (18 Nxd7 Nxd7 19 d5 exd5 20 Qxd5
Bxf4+ 21 Nxf4 Qxf4+ 22 Qd2 Qxd2+ 23 Rxd2 Nb6 24 b3 a4 =) 18...Bxe5
19
fxe5 Nxe5 20 Qg3 Nfg4 and now:

a) 21 Rd4 Nf6 22 Rd2 Nfg4 23 c6 (23 Rd4 repeats) 23...bxc6 24 Rd4 Nf6
25 Rf1 Ne8 26 Re4 Nd7 27 Qxc7 Nxc7 =.

b) 21 Bd4 Rad8 22 Rhe1 h5 is probably equal; e.g., 23 Nf4 g6 24 Nxh5

gxh5 25 Bxe5 Rxd1+ 26 Kxd1 Qxc5 =.

18 Nxd4 Nxe5 19 fxe5

The position is double-edged but roughly equal:

a) 19...Nh7 20 b3 (20 Qe2 Rfc8 21 b3 Bb4 22 Nb5 {or 22 Bxb4 axb4 23

Nb5 Qc5 =} 22...Qc5 23 a4 Bxc3 24 Nxc3 Nf8 25 h5 Rd8 =) 20...Qxe5 21
Nb5
Qc5 22 Rd7 Bf6 23 Rc7 Qb6 24 Rxb7 Qc5 and the position remains
sharp, with dangers for both sides.

b) 19...Qxe5!? is playable, though after 20 Nc6 Qc7 21 Nxe7+ Qxe7 22

g4! Nh7 White has enough compensation to avoid a disadvantage.

12E: 16 Ne2 h5

(1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 h4 h6 7 Nf3 e6 8 Ne5

Bh7 9 Bd3 Bxd3 10 Qxd3 Nd7 11 f4 Ngf6 12 Bd2 Be7 13 0-0-0 0-0 14 Qf3
Qc7 15 c4 a5 16 Ne2)

16...h5 (D)
 White to play

Contrary to traditional advice, Black moves forward on the wing where he

is defending. However, the position is very concrete and general
guidelines must take the back seat.

17 g4!?

This seems critical. Podgaets/Karpov/Odessky only give 17 Ng3 g6,

which looks fine for Black:

a) 18 Kb1 Rfd8 19 Bc3 Bb4 =.

b) 18 Qd3 Nxe5 19 fxe5 Ng4 20 Ne4 c5 21 d5 exd5 22 cxd5 Qxe5 23 d6

c4 24 Qe2 Bxd6 25 Bc3 Qf4+ =.

c) 18 Rhe1 Rad8 and now:

c1) After 19 f5? Nxe5 20 dxe5 Ng4 Black seems better: 21 f6 Bc5 22
Nxh5 Nxe5 –/+, 21 fxg6 fxg6 22 Qe4 Kg7 –/+ or 21 Nxh5?! Nxe5! 22 Qc3
Nd3+ 23 Kb1 e5 –+.
c2) 19 Kb1 Bb4 20 Bc3 Bxc3 21 Qxc3 Ne8 22 c5?! Ndf6 23 Qf3 Ng7 =+.

17...hxg4

Or:

a) 17...Nxg4 18 Nxg4 hxg4 19 Qxg4 transposes.

b) 17...Nxe5? 18 fxe5 Nxg4 19 Nf4 gives White strong play, since

19...g6? 20 Rhg1! is more or less lost for Black.

c) 17...c5 18 gxh5 cxd4 19 h6 Nxe5 20 fxe5 Qxe5 21 Rdg1 is very

dangerous for Black, but far from clear.

18 Nxg4 Nxg4 19 Qxg4 f5 20 Qg3 Nf6 21 h5 Ng4 (D)

White to play

Black has stabilized the kingside and the game should not end
quickly. Probably chances are equal but Black’s practical task
seems somewhat harder:
a) 22 Bc3 Bf6 23 Kb1 and now Black has various options including
counterplay with 23...b5 (e.g., 24 d5 exd5! 25 h6 Bxc3 26 Qxc3 b4), and
23...Kh7!?, simply neutralizing any ideas of White pushing his pawn to
h6 and h7.

b) 22 h6 g6 also seems to hold for Black if he knows what he is doing:

b1) 23 h7+ Kh8 24 Bc3 Bf6 25 Kb1 b5 26 d5 exd5 27 cxb5 Rae8 28 Bxf6+
Rxf6 29 Nd4 cxb5 30 Rde1 Re4 31 Nf3 Rf8 32 Nh4 Rf6 33 Nf3 =.

b2) 23 Ng1 Bf6 24 Nf3 b5 25 Nh2 may be the toughest test of Black’s
fortress but after 25...Kh7 26 Nxg4 fxg4 27 Qxg4 bxc4 Black is probably
OK; e.g., 28 Rde1 Rae8 29 Rhg1 Qf7 30 Re4 Rg8 31 Bc3 Qd7 32 Rge1 c5
=.
3: Advance Variation

Lesson 13: Arkell/Khenkin Introduction

By: TRH

Chess is a fighting game which is purely intellectual and includes chance.

Richard Réti

1 e4 c6 2 d4 d5 3 e5 (D)

White doesn’t develop or threaten anything and gives Black access to the
f5- square. Yet this strange move is currently White’s most popular choice.

Black to play
3...c
5
For an inexperienced player, Black’s last move might seem mysterious.
Why is it OK to move the c-pawn for a second time, as early as move
three? It’s not really a loss of tempo compared to standard lines, because
Black usually ends up playing ...c5 anyway. Black’s most common choice
is 3...Bf5, but by delaying this move Black hopes to develop the bishop all
the way to g4. To avoid this, White has to play accurately.

One reason for the Advance Variation’s popularity is the fact that Black has
only two replies that have stood the test of time. This, the Arkell/Khenkin
Variation, is one of them. It’s arguably a gambit, but definitely of the
healthier variety and in my opinion the easier line to prepare for Black.
While 3...Bf5 seems to be in excellent theoretical shape, a practical
difficulty is the number and variety of White’s replies, ranging from the
raving 4 g4 via 4 h4 to the quiet 4 Nd2, but with Van der Wiel’s 4 Nc3 e6 5
g4 and Short’s 4 Nf3 as the most respected systems.

Q: Is there really no third option? Maybe a tricky tactical line that will work
as a surprise weapon?

It’s not particularly tactical but if you desperately need a surprise weapon,
you could try the experimental 3...Qc7!?. See Lesson 18 for some
preliminary analysis.

Model Game 13

Perez Candelario – Delchev

Spanish Team Ch, Melilla

1 e4 2011
c6

Up to the Olympiad in 2014, I was training with grandmaster Aleksander

Delchev and achieved my best result ever. Delchev is a great fighter, and
has won a lot of open tournaments. To do so, he has to win as Black, and
the Caro-Kann can be the perfect weapon.

2 d4 d5 3 e5 c5 4 Nf3
4 dxc5 is by far White’s most popular move and will be the subject of
Lessons 14 to 17. Among the quite numerous alternatives, the text-move is
in my opinion the most critical. You will find some analysis of 4 c3 in 13B
and of White’s other 4th moves in 13A.

4...Bg4 (D)

4...Nc6 is a good alternative, and even more popular. However, this pin
normally leads to much sharper play, where White also has real chances to
go wrong.

White to play
5 dxc5

For the interesting 5 c4 and some less critical moves, see 13C.

5...Nc6 6 Bb5

With the idea of playing b4. For 6 c3, see 13D.

6...Qa5+
This forces White to place his knight on c3.

7 Nc3 e6 8 Be3

White can also play 8 Bd2 Qc7 9 b4 (D), with which he scores rather well.

Black to play

However, with some accuracy over the next few moves, Black should be
OK. After 9...a5 10 a3 Bxf3 11 gxf3 (11 Qxf3? is refuted by the simple
11...axb4, when White’s rook on a1 is hanging) 11...Qxe5+ 12 Qe2 Black
has a choice:

a) With White’s compromised structure it is natural to want to take the

game a step closer to an endgame with 12...Qxe2+?! 13 Kxe2 Ne7 14
Na4, as in Pijpers-Nabaty, European Ch, Batumi 2018, but this queenless
middlegame turns out to be treacherous for Black.

b) So 12...Qf5 seems a better option; e.g., 13 Bd3 (13 Bxc6+ bxc6 14

Nb5 must be answered with 14...Kd7!, when Black is OK) 13...Qf6 14
Rd1 Be7 15 Nb5 Kf8 16 h4 h5 17 c4 (17 Kf1 Qe5 reoffers the queen
exchange now that Black’s game is more stable) 17...axb4 18 cxd5 exd5
19 axb4 (Castro Salguero-E.Ramos, corr. 2017) and now 19...b6 (or
even 19...Ra2 20 Qe3
Rxd2 21 Rxd2 Nxb4) looks

playable. We return to 8 Be3 (D):

Black to play
8...Ne7 9
a3

After 9 h3?! Bxf3 10 Qxf3 a6 11 Ba4? (11 Bxc6+ Nxc6 =+ J.Gyimesi-Gierth,

corr. 2008) 11...0-0-0! Black is threatening ...d4 and is much better.

9...0-0-0

Black has alternatives, but I like this direct move.

10 b4 Bxf3 11 gxf3

11 Qxf3 Nxb4 12 0-0 Nxc2 13 Rac1 was played in Mekhitarian-

M.Santos, Campinas 2010. The position is very sharp, but Black is
probably to be preferred after 13...Nxe3 14 fxe3 Nf5.

11...Qc7 (D)
 White to play

White has an extra pawn and the bishop-pair, but also has problems with
his king safety and weak pawn-structure.

12 Bxc6

The immediate 12 Bf4?! g5 13 Bg3 Nxe5 is better for Black.

12...Nxc6 13 Bf4 a6

Not 13...g5? 14 Nb5! +/–. In the previous variation, the b5-square was
not available to the white knight.

14 Bg3 g5 15 h4 (D)

Alternatively:

a) 15 0-0 h5 is very dangerous for White, Edouard-Fridman, European

Ch, Plovdiv 2012.

b) 15 Qe2 Nd4 16 Qd3 Nf5 is at least equal for Black, Pijpers-Admiraal,

Leiden tt 2017.

c) 15 Qd2 h5 and Black has no problems, Tiglon-Ostrovsky, Charlotte 2018.

Black to play

15...Bg7 16 hxg5 Bxe5 17 Ne2 f6!

By opening up the position, Black will give White’s king a hard time.

18 gxf6 Qf7

18...h5 has been played in a few games, but I like Delchev’s interpretation.

19 Rb1

19 Bxe5 Nxe5 20 f4 Nf3+ 21 Kf1 Qxf6 22 Rb1 e5 23 Ng1 Nd4 24 fxe5

Qxe5 is equal according to the engine, but Black is much more active.
Black prevailed in Neusch-Desquiens, corr. 2011.

19...Bxg3 20 fxg3 Qxf6

White’s position is very dangerous as there is no way to exchange pieces
or to find a safe place for the king.

21 Rb3 e5 22 Qd2 h5 23 Rh4 e4 24 Kf2?

24 Rf4 keeps White in the game.

24...Ne5 25 Rf4 Qg5 26 b5 (D)

This is White’s only chance – a counter-attack on Black’s king.

Black to play

From now on you will see a lot of mistakes by both players, but don’t
judge them too harshly.

26...h4 27 g4 h3?

It is getting amazingly complicated, but 27...Nc4 brings Black close to

victory: 28 Qd4 h3 29 bxa6? e3+ 30 Rxe3 h2 31 a7 Qh4+ 32 Ng3 and
now Black wins by promoting to a knight: 32...h1N+!.
28 bxa6

Suddenly White is winning!

28...h2 29 a7 Qh4+ 30 Ke3?

Once again, the tables turn! 30 Ng3! h1N+ 31 Ke3 Nc4+ 32 Kd4 and
incredibly, now that White has escaped the checks, it’s Black who is
mated.

30...Nc4+ 31 Kd4 (D)

Black to play
31...Kc7?

The position is so complicated that it’s almost impossible to navigate. The

engine says Black is winning after 31...Kd7 32 Rxb7+ Kc6 but I dare you
to continue successfully after 33 Rff7.

32 Rxb7+?

After this there are no more adventures – Black is winning! White could have
forced mate with a difficult line starting 32 Rf7+! Kc6 33 Kc3!!.

32...Kxb7 33 Qb4+ Ka8 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 e5 c5 (4 Nf3 Bg4 5 dxc5 Nc6 6 c3)

13A: (4 Ne2)

13B: 4 c3

13C: 5 c4

13D: 6 c3

13A is significant mainly due to the transposition from the 2 Ne2 line (see
30A). The quiet 4 c3 of 13B isn’t theoretically challenging, but not as
innocent as it might appear, so do take a look at it. Finally, 13C and 13D
are sharp and principled lines, so an examination of the main lines should
come quite high on your list of priorities.

13A: Rare 4th Moves

(1 e4 c6 2 d4 d5 3 e5 c5)

4 Ne2

This has been played by quite a few strong grandmasters (frequently from
the move-order 2 Ne2 d5 3 e5 c5 4 d4).

White has also tried 4 c4, and after 4...cxd4 (D) there is a split:
 White to play

a) 5 Nf3 Bg4 transposes to the line 4 Nf3 Bg4 5 c4 cxd4 of 13C.

b) 5 Qxd4 is unproblematic for Black. Simplest and probably strongest is

5...e6, when I start to wonder what the queen is doing on d4. After 6 cxd5
Nc6 7 Bb5 Qxd5 8 Qxd5 exd5 9 Nf3, instead of 9...Ne7, as in
Nevednichy- Dobre, Romanian Ch, Baile Olanesti 2013, I suggest 9...a6
10 Ba4 (after 10 Bxc6+ bxc6, Black’s bishop-pair compensates for his
slightly weakened structure) 10...Bg4 11 0-0 Bc5 12 Nbd2 Ne7 = Milde-
V.Khan, corr. 2013.

c) 5 Ne2 is a rare line, but played by Gawain Jones twice. The idea is to
recapture with the knight on d4, and getting there via e2 means that the pin
with ...Bg4 is no longer an issue. Surprisingly, 5...dxc4!? has never been
tried. Taking the pawn demonstrates the downside of having the knight on
e2. Then:

c1) 6 Qxd4?! is strongly met by 6...Bd7, when Black is already better.

For instance, 7 Qxc4 Nc6 8 Bf4 Rc8 and the white queen is misplaced,
leaving Black at least slightly better.

c2) After 6 Nxd4 Qa5+ 7 Nc3 Qxe5+ Black wins a pawn, but is behind in
development. I believe chances are balanced. A possible continuation is
8 Be2 Nf6 9 0-0 e6 10 Re1 Be7. Engines surely are party poopers
sometimes; here they say ‘0.00’, but the position is by no means boring.

We now return to 4 Ne2 (D):

Black to play
4...Nc6

Black has some other good options but I like this move.

5 c3

White isn’t active enough to open the centre.

5...Bf5 6 dxc5

After 6 Ng3 Bg6 the always creative grandmaster played 7 e6 and went on
to win in Shirov-Plazuelo Pascual, Roquetas de Mar 2019 after Black
accepted the sacrifice, but after 7...cxd4 8 cxd4 I suggest 8...Qd6, when it
looks like Black is doing fine.
6...Nxe5 (D)

In general, exchanging the c-pawn for the e-pawn can be considered a

small success for Black.

White to play
7 Nd4

7 Ng3 contains a little trap, so you should be careful: 7...Bc8 (it turns
out 7...Bg6?? drops a piece to 8 f4 followed by 9 f5, trapping the
bishop) and White has tried:

a) 8 Na3 e6 9 b4 Nf6 10 Bb5+ Bd7 11 Qe2 Nc6 12 0-0 Be7 13 Nc2 h5

and Black grabbed the initiative and went on to win in Himanshu-
Ashutosh, Udaipur 2019.

b) 8 Bb5+ Nc6 9 0-0 e6 and now:

b1) 10 Be3 Nf6 11 Bxc6+ bxc6 and in Alcazar Jimenez-Andrade, World

Under-16 Ch, Oropesa del Mar 2000 White weakened the e4-square with
12 f4?!, which Black should have immediately exploited with the thematic
12...h5!.

b2) 10 b4 g6 11 Be3 Bg7 12 Bd4 was played in B.Smith-Zelčić, Zagreb

2016. Here Black should have played ambitiously with 12...f6; e.g., 13
Nd2 Ne7 14 f4 0-0 =+.

7...Bd7

Now:

a) 8 f4 Nc6 is fine for Black, and after 9 Qb3? e5 Black already had a
decisive advantage in Chuprov-Airapetian, Taganrog 2011; e.g., 10
fxe5 Qh4+ and White can’t play 11 g3 because of 11...Qe4+.

b) After 8 Qh5 Qb8 9 Bf4 Nd3+ 10 Bxd3 Qxf4 11 0-0 Nf6 12 Qe2 White
went on to win in Sedina-Botta, Lugano 2011 but after 12...g6! I believe
Black is the one with the advantage.

13B: 4 c3

(1 e4 c6 2 d4 d5 3 e5 c5)

4 c3

This is played relatively often, usually in a quite harmless manner.

4...Nc6 (D)
 White to play
5 Be2

White is trying to trick us into unfamiliar positions. White’s point is to delay

the development of his king’s knight until Black has committed his light-
squared bishop. Nevertheless, it is rather a rare choice for White.

5 Ne2 transposes to 13A, and other moves are harmless:

a) The delayed 5 dxc5 is no longer critical and after 5...Nxe5 6 Nf3 Nc6
Black will build a strong centre and was comfortable in Teodoro-
Spraggett, Canadian Ch, Toronto 1996.

b) 5 Be3 is best met by 5...Nh6, planning ...Nf5. The principled 6 Bxh6

gxh6 is probably better for Black as White has lost precious time with his
dark- squared bishop, and Black’s bishop-pair compensates for the
damaged pawn- structure.

c) 5 Nf3 cxd4 6 cxd4 Bg4 7 Be2 e6 has occurred in quite a few games,
such as Pirverdiyev-Jobava, Poti 2015. Black will continue kingside
development, leading to equality.
5...Bf5

A sensible alternative is 5...Qc7, leading to a position discussed in Lesson

18 under the move-order 3...Qc7 4 c3 c5 5 Be2 Nc6 (note ‘b2’ to White’s
4th move in 18A).

6 Nf3

Now there is no reason to delay this extremely natural move.

6...e6 7 0-0 (D)

Black to play

We have reached a relatively rare version of the Short Variation,

normally reached via 3...Bf5 4 Nf3 e6 5 Be2 c5 6 0-0 Nc6 7 c3.

Q: Does this mean we have to learn two different openings?

Not really; the critical move in the traditional move-order given above is
6 Be3 as White puts more pressure on the c5-pawn and keeps the
option of playing c4. The diagram position is not a critical test for Black.
7...cxd4 8 cxd4

White’s centre gives him more space, and Black’s problem is that e7 is a
natural square for both the knight and the bishop. The right way to play
this type of position is to untangle slowly. This is possible because of the
closed nature of the position.

8...Nge7 (D)

White to play

9 a3

Or:

a) At first sight 9 Nh4? might seem critical, but it’s no good due to 9...Qb6!,
when White has problems protecting d4. After 10 Nf3 Be4 Black was
clearly on top in Vdovychenko-Dzhumabaev, Russian Team Ch, Sochi
2012.

b) 9 Nc3 Nc8 10 Be3 Nb6 11 Rc1 Be7 12 Na4 Nxa4 13 Qxa4 0-0 =
Anand- Speelman, Linares 1992.
c) 9 Be3 was played in another high-level game. Black’s set-up was very
similar to our main game: 9...Nc8 10 Nbd2 Be7 11 Ne1 0-0 12 f4 Bg6
(pre- empting g4 as now 13 g4 f6 would leave Black at least equal) 13
Ndf3 Nb6 and Black threatens 14...Nc4, provoking a weakness on
White’s queenside. After 14 b3 a5 15 a4 Nb4 16 Nd3 Rc8 the position is
equal. In the next few moves, Black executes an instructive knight
manoeuvre: 17 Nfe1 Nd7 (the knight was misplaced on b6, and is now
heading for c6) 18 Nxb4 Bxb4 19 Bb5 Nb8 20 Nd3 and in Kariakin-Leko,
Nice rapid 2008, Black could have gained the advantage with 20...Bc3 21
Rc1 Qb6.

9...Nc8

This might seem strange, and it’s one of many possibilities for Black, but
the idea is to make room for the dark-squared bishop. The knight will find a
good place on b6 or even a7.

10 Nbd2

After 10 b4 Be7 11 Nbd2 Nb6 12 Bb2 0-0 13 Rc1 Rc8 14 Nb3?! Nc4! 15
Bxc4 dxc4 16 Rxc4 (Hamdouchi-Adams, Cap d’Agde rapid 1994) Black
would have good compensation even without the nice tactic 16...Nxb4!
17 Rxc8 Qxc8 18 axb4 Bc2, after which Black will end up regaining his
pawn, while retaining a big positional advantage.

10...Be7 (D)
 White to play

As can be seen from the moves so far and the comments, play is more
about plans than memorizing long lines. So from here we give the
complete game Shirov-Karpov, Las Palmas 1994 which will show how
strategically rich this type of position is.

11 b3

This move is useful against a knight on b6, but as we shall see, Black
doesn’t have to be so cooperative.

11...a5

This way Black makes sure White’s pawn stays on b3.

12 Bb2 N8a7 13 Re1 0-0 14 Nf1

Perhaps White should have tried to play the knight to b1 and c3 instead.
In the game, Black’s initiative on the queenside grows fast.

14...Rc8 15 Ng3 Bg6 16 Qd2?! Qb6 (D)

After this move, Black is slightly better. Also good was 16...a4 17 bxa4
Na5 as Black threatens ...Rc2 in addition to ...Nb3 or even ...Nc4.

White to play
17 Bd1 Rc7 18 h4 h6 19
Re2

It’s hard for White to create an attack. The best I can find is 19 Qf4 Rfc8
20 Re3, trying to swing heavy artillery over to the kingside.

19...Rfc8 20 Qf4 Nb8

Another deep move. The knight was blocking the rooks, but from b8 it
can easily go to d7 and f8 to give extra protection to the king.

21 Re3 Nb5 22 Ne2 Nd7 23 Nh2

Now Black really wants to put more pressure on the a3-pawn. If it’s forced
to move, Black will gain access to the b4-square. Karpov found an
impressive way to attack it.

23...Rc6 24 Ng4 Qd8

Attacking h4.

25 Rh3 Qf8

Suddenly it’s the a-pawn that’s under fire.

26 a4 Na3

Black is planning ...Nc2, from where it threatens the rook and can
manoeuvre to d3 via b4 or even e1. White didn’t want to see any of this
and decided to take it.

27 Bxa3 Bxa3 28 Rg3 h5 29 Ne3 Bb2 30 Ra2 Bc1 31 Qg5 Qb4 (D)

White to play
32 Nf4?

Black is better, but nothing is decided after 32 Kh2.

32...Bxe3 33 fxe3 Rc1 34 Nxg6

This desperate attack leads nowhere.

34...Rxd1+ 35 Kh2 fxg6 36 Qxg6 Qe7 37 Rf2 Qxh4+ 38 Rh3 Qxf2

Here White resigned.

13C: 5 c4 and Rare 5th Moves

(1 e4 c6 2 d4 d5 3 e5 c5)

4 Nf3 Bg4 5 c4

Together with 5 dxc5, this is the most challenging option. 5 Bb5+ Nc6 6
dxc5 transposes to Model Game 13.

5...cxd4 (D)

White to play

6 cxd5

Or:
a) 6 Qa4+ Bd7 7 Qb3 dxc4 8 Bxc4 e6 9 Nxd4 Nc6 10 Nxc6 Bxc6 is
equal. For instance, 11 0-0 Ne7 12 Nc3 (Nisipeanu-Zelčić, Bosnian
Team Ch, Neum 2004) 12...Nf5 and Black should have no problems.

b) 6 Qxd4 Bxf3 7 gxf3 and now the best option is to sacrifice a pawn for
the initiative: 7...Nc6 8 Qxd5 Qc7 is rarely played, but in my opinion it’s
very logical. White lags behind in development and has a hard time
finding a safe place for his king, which means Black should keep the
queens on. Then 9 f4 (9 Bf4? e6 10 Qe4 0-0-0 11 Nc3 Bb4 –/+ Saravana-
Ostrovsky, Internet blitz 2020) 9...e6 10 Qe4 f5 11 exf6 (11 Qe3 0-0-0 12
Bg2 g5! gives Black the initiative) was played in Linchevsky-Kovchan,
Russian Team Ch, Sochi 2007. Here I suggest 11...0-0-0, after which
Black has excellent compensation. The greedy 12 Qxe6+? Kb8 13 Be3
Nxf6 leaves White in a hopeless situation, despite being two pawns up.

6...Qxd5 7 Nc3 Qa5 (D)

My first intention was to recommend 7...Bxf3 but following 8 Nxd5 Bxd1 the
game normally gets very sharp after 9 Nc7+ or 9 e6. However, the novelty
9 Kxd1 gives Black surprising difficulties; e.g., 9...Kd7 10 Bb5+ Nc6 11 e6+!
fxe6 12 Nb4 Nf6 13 Nd3! and despite being two pawns up, I could not find
a path to equality and had to abandon the whole line.
 White to play
8 Bb5+

The straightforward 8 Qxd4 Bxf3 9 gxf3 is at best equal for White.

9...Nc6 10 Bb5 Qxb5! 11 Nxb5 Nxd4 and now:

a) 12 Nxd4 a6! (taking squares away from the white knight) 13 Be3 (13 e6
g6 and when Black’s bishop appears on g7, Black has the advantage)
13...e6 14 Ke2 Ne7 15 Rac1 Nd5 = Bak-Tyton, British League (4NCL)
2012/13.

b) 12 Nc7+ Kd7 13 Nxa8 Nc2+ 14 Ke2 Nxa1 leads to a funny situation.

Both knights are in the corner, but don’t forget that Black is a piece up. The
engines give 15 Rd1+ Ke6 16 Nc7+ Kf5 as slightly better for Black.

8...Nc6 9 Bxc6+

9 Qxd4 Bxf3 10 gxf3 transposes to the previous note.

9...bxc6 10 Qxd4 Bxf3 11 gxf3 e6

Black has the better structure, which means White has to create something.
12 Be3

12 0-0 Ne7 13 Rd1 Rd8 14 Qxd8+ Qxd8 15 Rxd8+ Kxd8 16 Be3 was
played in Iordachescu-Zelčić, Dresden 2003 and here I suggest 16...Kc7
with the point that after 17 Bxa7 Ng6 18 Bd4 Nh4 Black regains the pawn
with good play.

12...Ne7 13 0-0-0

Or 13 0-0 Nf5 14 Qc4 Rc8 15 f4 Be7 = Ni Hua-Kazhgaleev, Sydney 2015.

13...Nd5 (D)

A rare move.

White to play
14 Qc4

This must be White’s only critical try.

14...Rc8 15 Nxd5 exd5 16 Qg4

White has to continue forcefully, or else Black will be very
comfortable. After 16 Qb3 Bc5 17 Rhg1 Bxe3+ 18 Qxe3 g6 Black has
no problems, especially as 19 e6 is met by 19...0-0!.

16...Rb8 17 e6

Or 17 Kb1 Qb4 =.

17...f6 18 Kb1

18 e7 Bxe7 19 Qxg7 Rxb2 20 Qxh8+ Kd7 21 Kxb2 and it’s a draw

by perpetual check.

18...Bd6 19 Rd2

19 f4 Rxb2+!? =.

19...Rb7

Black has neutralized White’s initiative.

13D: 6 c3

(1 e4 c6 2 d4 d5 3 e5 c5 4 Nf3 Nc6)

5 dxc5 Bg4 6 c3

With this move, White keeps the option of playing b4 and Be3. At the
same time, White might prepare an early Qb3.

6...e6 (D)

Please not 6...Nxe5??, as 7 Nxe5 Bxd1 8 Bb5+ picks up a couple of pieces.

 White to play

The c5-pawn is under attack and White has two ways of defending it.

7 Be3

7 b4 is also possible. With the bishop on g4, I like spending a move

to prevent Bb5 with 7...a6!?. After 8 Nbd2 Nxe5 Black is doing fine:

a) 9 Qa4+ and now:

a1) 9...Qd7 10 Qxd7+ Nxd7 11 Bb2 Ngf6 12 Nd4 Be7 13 N2b3

(Predojević- Zelčić, Zadar 2010) 13...0-0 =.

a2) 9...Nd7 10 Ne5 Nf6 11 c4?! (11 Nxg4 Nxg4 12 Be2 ½-½ was a quick
draw in A.Tate-Esserman, Caleta 2017) 11...a5 12 Nb3 axb4?! 13 Qb5
Be7
–/+ Topalov-Nakamura, London 2016.

b) After 9 h3 Bxf3 10 Nxf3 Nxf3+ 11 Qxf3 g6 White has the bishops while
Black has the better centre and pressure on the long diagonal. 12 Be2
Bg7 13 0-0 Nf6 14 Bg5 h6 15 Bxf6 Qxf6 16 Rac1 Qxf3 ½-½ Skaperdas-
Kapnisis, Paleochora 2013.
c) 9 Be2 Nc6 equalizes: 10 0-0 (10 Nd4 Bxe2 11 Qxe2 Be7 12 N2b3 Bf6
13 0-0 Nge7 14 Rd1 0-0 = Naiditsch-Naer, Porto Vecchio 2015) 10...Nf6
11 h3 Bh5 12 Bb2 Be7 13 Nd4 Bxe2 14 Qxe2 (Frolianov-Alekseev,
Kaliningrad 2015) and here the simple 14...0-0 leads to an equal position.

7...a6 (D)

White to play

This is a relatively rare move. Also in this line we take time to control the
b5- square, and White is unable to take advantage of his slight lead in
development. Black’s idea is simply to take on e5 without allowing a check
on b5. Black often takes on f3, but I see no reason to give up the bishop-
pair that easily.

8 Nbd2

Or:

a) 8 Bd4?! allows 8...Bxf3 9 Qxf3 b6 10 cxb6?! Nxd4 11 cxd4 Qxb6,

when Black regains the pawn with a slightly better position, Hossain-Abu
Sufian, Dhaka 2017.
b) At first, the engine is enthusiastic about unpinning with 8 Qb3. It is best
met by 8...Bxf3 9 Qxb7 (9 gxf3 is not critical as 9...Qc7 gives Black lots of
compensation due to White’s weaknesses) and now after the new move
9...Nge7 Black is fine in all variations. After 10 gxf3 Ra7 two retreats need
to be examined:

b1) 11 Qb3 Nf5 12 Qa4 Rc7 and Black is planning to take on e3 and
capture on c5, while the queen is ready to reach h4.

b2) 11 Qb6 should be met by 11...Qa8!. It took me a while to find this

continuation, but it solves all Black’s problems. Black is preparing
...Nf5,
...Be7, ...0-0 and ...Rb8. White has difficulty finding a safe space for his
king, which adds up to sufficient compensation for Black; e.g., 12 Nd2 Nf5
13 Nb3 Nxe3 14 fxe3 Be7 and Black has good play for the two sacrificed
pawns.

8...Nxe5 (D)

White to play

9 Be2

Or:
a) After 9 h3 Bxf3 10 Nxf3 Nxf3+ 11 Qxf3, 11...Qc7 is a clever move to
prevent Qg3 and provoke b4. Miras Garcia-D.Woodard, corr. 2015
continued 12 b4 Bd6 13 Rc1 Be5 14 Bd3 Ne7 15 0-0 0-0 16 Rfe1 Nc6 17
a4 g6 and Black retreated his bishop to g7, preparing to advance his
central pawns. I believe chances are roughly equal.

b) 9 Qa4+ Qd7 10 Qf4 (10 Qxd7+ Nxd7 11 b4 Ne7 shouldn’t pose Black
any problems) 10...Nxf3+ 11 Nxf3 Bxf3 12 Qxf3 Rc8 13 Bd4 Ne7 14 Bd3
Nc6 15 0-0 Nxd4 16 cxd4 is equal; e.g., 16...Be7 17 Rad1 Bf6 18 Bc2 b6
19 b4 a5 20 Rb1 axb4 21 Rxb4 bxc5 22 Ba4 cxb4 and in this highly tense
position, the game was agreed drawn in R.Neubauer-Scheiba, corr. 2014.

9...Nc6 (D)

White to play
Now:

a) After 10 0-0 Be7 11 h3, there is no reason to give up the bishop-pair.

Jasinski-Biedermann, corr. 2015 continued 11...Bh5 12 Nd4 (12 b4 Nf6 is
roughly equal) 12...Bxe2 13 Qxe2 Nf6 14 b4 0-0 15 Bf4 Re8 with a
roughly equal position.
b) 10 Qa4 Nf6 11 h3 (11 c4, as in Mokshanov-Chigaev, Russian Under-
18 Ch, Sochi 2015, is best met by 11...Bh5 = so as to recapture with the
knight on d5) 11...Bxf3 12 Bxf3 was played in A.Markgraf-A.Braun,
Bundesliga 2016/17, and here 12...Be7 13 0-0 0-0 leads to an equal
position. White has the bishop-pair, while Black has the better centre and
healthier structure.

c) 10 h3 Bxf3 11 Nxf3 was played in S.Zhigalko-Paichadze, European Ch,

Minsk 2017. As White normally goes for b4, an interesting idea is to place
the bishop on the long diagonal; e.g., 11...g6 12 0-0 Bg7 followed by
...Nge7 and ...0-0, and Black is not worse.

Lesson 14: Arkell/Khenkin 4 dxc5 e6 Introduction and 5 Nf3

By: TRH

The greatest challenge to any thinker is stating the problem in a way that
will allow a solution.

Bertrand Russell

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5

Although some of the moves of the previous lesson clearly had their points
and demanded precise play from Black, accepting the pawn must be
critical. It is therefore no surprise that there are more games with this
move than with all the alternatives combined.

4...e6 (D)

This is currently the hottest move, and probably best. It’s clearly a
practical plus for us that White must also prepare for the previous main
continuation, 4...Nc6.
 White to play

In this lesson we shall investigate the natural 5 Nf3. The main line, 5 a3, is
examined in Lessons 16 and 17, while 5 Be3 and rarer moves are the
subject of Lesson 15.

Model Game 14

Lennartz – R. Mayer

corr. 2013

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6

As we shall see in Model Game 16, Volokitin-Van Foreest, there are

situations where White doesn’t follow up a3 with b4. In this game, we
will investigate White’s possibility of leaving out the move a3 altogether.

5 Nf3

Judging just by the number of games, we would conclude that this natural but
somewhat bland developing move is clearly the main alternative to 5 a3
and 5 Be3. However, a more principled move, demonstrating the drawback
of delaying ...Nc6, is 5 Qg4, which is analysed together with 5 Bd3 in 15A.

5...Bxc5 6 Bd3 (D)

Except for 6 a3 (transposing to Model Game 16), this is the only

reasonable set-up. It was recently played by Kariakin.

Black to play
6...Ne7

This is the plan we shall also recommend against lines with a3 – not
risking being out-manoeuvred by tricky move-orders. Unfortunately, in
Kariakin- Abdrlauf, Internet blitz 2020 Black played 6...Nc6, so we must
wait for Kariakin to reveal his repertoire.

7 0-0 Ng6

Targeting the e-pawn. Please note that 7...0-0?? loses to 8 Bxh7+!.

8 Nbd2

From this moment on, the game can no longer transpose into a3 lines
from Lessons 16 and 17 where we delay, or even skip, ...Nc6. White’s last
chance to enter the typical a3 lines is 8 a3, which transposes to the
position we discuss in Model Game 16 via the move-order 5 a3 Bxc5 6
Nf3 Ne7 7 Bd3 Ng6 8 0-0.

Another idea is 8 Qe2 – see 14A for this and other 8th-move options
for White.

8...Nc6 (D)

White to play

We are in time to attack the e-pawn in the most natural way.

9 Nb3

Practically the only move. White has to be precise, as the automatic 9

Re1? gives Black the advantage: after 9...Qb6! 10 Qe2 Nf4 11 Qf1 Nxd3
12 cxd3 Nb4 Black had the bishop-pair and the initiative in F.Niehaus-
H.Möller,
Berlin Ch 2011.

9...Bb6

9...Be7 is a decent alternative, though a little passive, and the statistics

are clearly in White’s favour.

Instead, the bishop is excellently placed on b6. White is still faced with
the threat of Black taking on e5, and a decision must now be made.

10 Re1

For 10 Qe2, see 14B.

10...0-0 11 Bg5

This has the aim of disrupting Black’s set-up, and this way preventing
...Bc7. For White’s other options, see 14C.

11...Qc7 (D)

White to play
This is not the only move, but going for the e-pawn sets White the
most problems.

12 Bxg6

Or:

a) 12 Qe2? Ngxe5 13 Nxe5 Nxe5 14 Qxe5 Qxe5 15 Rxe5 f6 gives Black

a slight advantage, at least!

b) 12 a4 is a tricky move, indirectly protecting the e-pawn because of

the possibility of flicking in the a5 advance. Black must avoid
12...Ncxe5? 13 Nxe5 Nxe5 14 a5, when White wins a piece, so these
options remain:

b1) 12...Bd7 13 Qd2 Qb8 14 Qe2 Bc7 15 Bxg6 fxg6 16 c4 Rf5 is

completely equal according to the engines, but I would have been worried
when looking at my queenside, especially after 17 Nc5. Black drew
comfortably in Petronijević-De Homont, corr. 2016. I find Black’s play
instructive, but I wouldn’t have decided upon such a plan in a practical
game.

b2) 12...f6 is a very ambitious and double-edged decision. After 13 exf6

gxf6 two options are:

b21) 14 Bd2 threatens a5. After 14...Qf7 15 a5 Bc7 a complex fight

lies ahead, but I see great potential in Black’s position.

b22) 14 Bh6 Rd8 gives us a slightly weakened kingside, but the more I
look at this, the more I like it for Black. His centre is about to start rolling,
and the King’s Indian player in me wants to go for ...Kh8 and ...Rg8; White
has to be careful.

12...fxg6

Black would like to play ...h6 and ...g5. After capturing on g6, White no
longer has a light-squared bishop, which means we are not really
talking about weakening the kingside after such moves.

13 Be3 Bd7 14 Rc1

White is planning c4.

14...Bxe3 15 Rxe3 (D)

Black to play
15...Rf4!

Prophylaxis against the above-mentioned

plan. 16 c3 Raf8 17 Rc2 h6 18 g3 Re4 19

Rce2 After 19 Rxe4? dxe4 White drops the

e-pawn. 19...Rxe3 20 Rxe3 g5 21 h3 Qb6 22

Qa1

I don’t know how to comment on such a move. It looks like the engines
had completely taken over the thinking...

22...a5 23 Nbd4 ½-½

The position is probably equal, but there is a lot of play left. Black can play
for a queenside minority attack or try to improve his bishop to g6, h5 or
even e4.

Theory Magnifiers

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 Nf3 Bxc5 6 Bd3 Ne7 7 0-0 Ng6 (8 Nbd2

Nc6 9 Nb3 Bb6 10 Re1 0-0 11 Be3)

14A: (8 Qe2)

14B: 10 Qe2

14C: (11 Be3)

The lines in this lesson may not be very theoretical but they are all quite
unbalanced and White has some kingside attacking potential. Setting
aside some time for these magnifiers, looking for moves that would be
scary to face unprepared, could be time well spent.

14A: 8th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 Nf3 Bxc5)

6 Bd3 Ne7 7 0-0 Ng6 (D)

 White to play

8 Qe2

Or:

a) 8 a3 transposes to Model Game 16.

b) 8 Nc3 should be met by 8...Bd7, preventing Na4. The knight is

misplaced on c3, and Black is doing fine; e.g., 9 Bg5 Qc7 10 Re1 (10 Nb5
Bxb5 11 Bxb5+ Nc6 =+) 10...a6, intending ...Nc6 with pressure on White’s
e-pawn.

c) 8 Bg5 Qc7 9 Bxg6 fxg6 10 c4 dxc4 11 Nbd2 b5 12 a4 Bb7 13 axb5 h6

14 Be3 (14 Bh4 0-0 15 Bg3 Bd5 16 Qc2 g5 = Teske-Molinaroli, Pardubice
2012) 14...0-0 (14...Bxe3?! 15 fxe3 0-0 16 Qc2 +/–) 15 Qe2 Bxe3 16
Qxe3 Nd7 =.

d) 8 Re1 Nc6 (D) and now:

 White to play

d1) 9 c3 0-0 10 Be3 Be7 11 Bd2 Bd7 12 a4 f6 13 exf6 Bxf6 14 Na3?! e5 =+

T.Thorhallsson-H.Olafsson, Icelandic Team Ch 2013/14.

d2) 9 Qe2 0-0 10 Bg5?! Nd4 11 Bxd8 Nxe2+ 12 Rxe2 Rxd8 13 g3 Bd7
(13...f6!?) 14 Nbd2 a6 15 Nb3 Ba7 = Fluvia Poyatos-Arizmendi
Martinez, Escaldes 2018.

8...Nc6 (D)
 White to play

9 c4

Or:

a) 9 a3 Qc7 (9...0-0!?) 10 Re1 Nd4 11 Nxd4 Bxd4 12 Bxg6 hxg6 13 c3 and

instead of 13...Bc5?! 14 Nd2 += Lazović-Zelčić, Zagreb 1997, 13...Bb6
keeps the option of ...Qc4; e.g., 14 Nd2 Bd7 15 Nf3 Qc4 16 Qxc4 dxc4
gives Black the bishop-pair and no reason to complain.

b) After 9 c3 0-0 10 b4 (D) (10 Bg5 Qc7 11 b4 Bb6 12 Re1? Ncxe5 13

Nxe5 Qxe5 14 Qxe5 Nxe5 15 Rxe5 f6 –/+ Muratović-Dizdarević, Zenica
2012) both bishop retreats are possible:
 Black to play

b1) I was surprised to see I have reached this position ... with the white
pieces! Following 10...Be7 11 b5 Na5, 12 g3 f6 leaves Black at least
equal, while after 12 Be3 Qc7 13 Bxg6 fxg6 14 Nbd2 Bd7 15 Bd4, as in
T.R.Hansen-D.Jørgensen, Copenhagen 2002, Black is clearly better.
However, he misplayed the middlegame and I managed to win.

b2) 10...Bb6 11 a4 f6 12 Bxg6 hxg6 13 exf6 gxf6 14 Be3?! e5 15 Nbd2 Kg7

=+ M.Savić-Nabaty, Belgrade 2012.

9...dxc4 10 Bxg6

After 10 Bxc4?! 0-0 White’s e-pawn is rather weak, and Black’s ideas
of playing ...Qc7 or ...Nd4 leave him slightly better.

10...hxg6 (D)
 White to play

Now the play becomes very sharp:

a) 11 Rd1 Qc7 12 Nc3 a6 13 Be3 Bxe3 14 Qxe3 Ne7 gives

White compensation, and I believe the chances are roughly
balanced.

b) 11 Qxc4 Qd5 12 Qf4 Nd4 13 Nc3 Nxf3+ 14 gxf3 Qd8 15 Be3 was
played in Kurnosov-Zlochevsky, Serpukhov 2002, and here I suggest
15...Bxe3 16 fxe3 Bd7 17 Ne4 Qe7 18 Nd6+ Kf8 with a very tricky position.
It looks risky to allow a knight to d6, but Black has counterplay against
White’s weakened kingside and the bishop will find a nice spot on d5.
Chances are roughly equal.

14B: 10 Qe2

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 Nf3 Bxc5 6 Bd3 Ne7 7 0-0 Ng6)

8 Nbd2 Nc6 9 Nb3 Bb6 10 Qe2

This is less common than 10 Re1 but has been used by some very strong
players.

10...0-0 (D)

10...f6 is also possible, when White’s best is 11 exf6 Qxf6 12 Bxg6+

Qxg6 13 Be3 Bc7 14 Bc5. With his king still in the centre, this looks
slightly dangerous for Black.

White to play
11
Bg5

Alternatively White can try 11 c4 dxc4, and now:

a) 12 Bxc4?! Qc7 13 Re1 Ncxe5 14 Nxe5 Nxe5 15 Qxe5 Qxc4 –/+

Tapie Amione-Adly, Mexico City 2016.

b) After 12 Bxg6 fxg6 13 Qxc4 Rxf3 (13...Bd7 =) 14 gxf3 Nxe5 15 Qe4,

15...Qf6 more or less forces a draw: 16 f4 Qh4 17 Qxe5 Qg4+ 18 Kh1
Qf3+ with a repetition. With 15...Nf7!?, Black can keep the tension and
play for a win in a level position.
11...Qc7 12 Rae1

Against 12 Rfe1?! Black has the same trick as in the main line, although
here it’s even better for Black, since the f-pawn will end up under fire:
12...Ncxe5 13 Nxe5 Qxe5 14 Qxe5 Nxe5 15 Rxe5 f6 16 Bxf6 Rxf6 17 Re2
(Kharchenko-Lomasov, Moscow 2016) 17...e5 18 Rxe5 Rxf2 19 Kh1 Bd7
–/+.

12...Ncxe5 13 Nxe5 Qxe5 14 Qxe5 Nxe5 15 Rxe5 f6 16 Bxf6 gxf6

I believe Black has every reason to be optimistic thanks to his bishop-pair.

14C: 11th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 Nf3 Bxc5 6 Bd3 Ne7 7 0-0 Ng6 8

Nbd2 Nc6 9 Nb3 Bb6)

10 Re1 0-0 (D)

White to play
11 Be3
This has been played by a few strong players, including GMs Van
Kampen and Predojević. Otherwise:

a) 11 a4?! Bc7 12 Bxg6 fxg6 13 c3 Rf5 14 Nbd4 Nxd4 15 cxd4 b5 16

Bd2 bxa4 17 Rxa4 Bd7 = Voveris-Dard, corr. 2018.

b) After 11 Bxg6 fxg6 12 Be3 Black has the clever tactic 12...g5!?.
Following 13 h3 (13 Nxg5?? Bxe3 14 Rxe3 Qxg5 –+; 13 Bxg5 Bxf2+! 14
Kxf2 Qxg5
=) 13...h6 I am happy about Black’s chances if the light-squared bishop
can reach g6 or h5; e.g., 14 c3 Bd7 15 Bxb6 Qxb6 =.

c) Another option for White is 11 c3, taking control of the d4-square

and preparing Nbd4. Black should force matters with 11...Bc7:

c1) The strange 12 Nbd4? Ncxe5 13 Nxe5 Nxe5 –/+ left Black a
healthy pawn up in Kett-T.Brown, Welsh Ch, Cardiff 2010.

c2) In case of 12 Bg5 Qe8 13 Bxg6 fxg6 the black queen is well placed
on e8, as ...h6 followed by ...g5 is on the agenda.

c3) 12 Bxg6 fxg6 13 a4 was played in Yu Yangyi-Cheparinov, Hengshui

blitz 2019. Here I suggest 13...h6, which is a typical move in this structure.
In the absence of White’s light-squared bishop, Black shouldn’t fear
expanding on the kingside. 14 h3 g5 15 Nbd4 Bd7 is complex but roughly
equal.

11...Bc7! (D)

11...Bd7, as in Predojević-Miladinović, Slovenian Team Ch, Rogaška

Slatina 2009, is imprecise due to 12 Bxb6 Qxb6 13 Qd2 +=.
 White to play

White’s idea is to exchange off Black’s powerful bishop, but we don’t have
to be so cooperative. This has been played in five correspondence games,
in all of which Black drew comfortably. This proves that the move is sound,
but as we all know, correspondence games are heavily influenced by
engines, and we have to look at it from a human perspective.

12 Bc5

Alternatively:

a) After 12 Bxg6 we capture with the f-pawn to open a file for the rook:
12...fxg6 13 Bc5 Rf5 14 Nbd4 Nxd4 15 Qxd4 and Black has 15...g5!, a
standard move in this type of structure. Black takes control of the f4-square
and introduces the possibility of playing ...g4. Next up is ...Bd7 and ...h6.
The bishop can consider swinging to the kingside via e8, and the half-open
f-file will be useful. After 16 Qg4 Bd7 the engine says it is completely equal,
but I believe Black is already slightly better, B.Mulligan-De Homont, corr.
2016.

b) 12 Bd4 intends to protect the e5-pawn. Then:

b1) Black succeeded in holding a draw with 12...Nxd4 in Alderisio-Andeer,
corr. 2014, but I don’t like the way this game went. After 13 Nbxd4 Nf4 14
Bf1 the engine agrees that the game’s 14...Ng6 is best, but to me it looks a
bit too desperate for half a point. A draw is not necessarily a bad result, but
let’s fight for more.

b2) 12...Nf4!? 13 Bf1?! (13 Bb5 is probably better, but Black should have
no problems after 13...Bd7) 13...f6! 14 exf6?! Nxd4 15 Nfxd4 Qxf6 looks
like a French Tarrasch that has gone horribly wrong for White.

12...Ngxe5!

Moving the rook is possible, but let’s play with fire! White’s idea is that
12...Re8 13 Bxg6 fxg6 leaves Black’s rook misplaced on e8.

13 Bxf8 Qxf8 14 Nxe5 Nxe5 (D)

White to play

Black has good compensation and should complete his development and
slowly expand while trusting the strength of his bishop-pair and pawn-
centre. Black is not worse, as demonstrated by correspondence chess as
well as in a
blitz game between two top grandmasters:

a) 15 c3 Bd7 16 Nd4 Rd8 17 Bc2 g6 18 Qe2 Qd6 with at least

equality, M.Suarez-De Homont, corr. 2016.

b) In L.Dominguez-Cheparinov, Hengshui blitz 2019, White returned the

material by playing 15 Bxh7+ Kxh7 16 Rxe5 Bxe5 17 Qh5+ Kg8 18 Qxe5.
Black’s slightly weakened kingside is compensated by his stronger centre,
so the position is equal.

Lesson 15: Arkell/Khenkin 5 Be3 and Rare 5th Moves

By: TRH

If you play the Caro-Kann when you are young, what will you play when
you are old?

Bent Larsen

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 Be3

Strictly speaking, 3...c5 is a pawn sacrifice, and it’s important to see

what happens if White decides to hold on to it. In this lesson you will
see that Black gets sufficient compensation in all lines.

Model Game 15

H. Koll – Grayland

corr. 2009

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 (D)
 White to play

5 Be3

Or:

a) Unsurprisingly, 5 b4?! a5 scores massively for Black, as White can’t

keep the pawn. After a few more moves, 6 c3 axb4 7 cxb4 b6 8 Be3 bxc5
9 bxc5 Nd7, we see that it’s just bad for White.

b) 5 Bd3 has the tricky idea of meeting 5...Bxc5?! with 6 Qg4 Ne7 7 b4.
For information on a better response for Black, see the note to White’s 5th
move in 15A.

c) 5 Qg4 has been played three times by Vachier-Lagrave, so no matter

how strange it looks, it has to be treated seriously. The idea is to stop
Black from taking on c5, as the g7-pawn would be hanging. See 15A for
some analysis.

5...Nd7 (D)
 White to play
6 Bb5

This is the consistent move as it protects both advanced pawns. For 6 Nf3
and some minor alternatives, see 15B.

6...Ne7 7 c3

White also has another way of hanging on to the pawn by 7 Nf3 but once
again Black gets enough compensation. After 7...Nf5 8 b4 Nxe3 9 fxe3 a5
10 c3 Be7 11 0-0 0-0 12 a3 Qc7 13 Bxd7 Bxd7 14 Qd4 Rfb8 Black
continues with ...b6, and once the position opens up, the bishop-pair will
count, D.Poljak-Holroyd, corr. 2015.

7...a6 8 Bxd7+

8 Ba4 is another approach by White, but once again Black gets

sufficient compensation: 8...Nc6 9 b4 Ndxe5 10 Na3 Be7 11 f4 Ng6
and Black was doing fine in Utesch-Z.Szczepanski, corr. 2016.

8...Bxd7 9 Nf3 Nf5 10 Bd4

With this move, White secures his extra pawn. However, the
game continuation is typical for this variation.

10...Nxd4 11 cxd4 b6 (D)

White to play

12 cxb6 Qxb6 13 Qb3 Rb8

The position is equal.

14 Qxb6

After 14 0-0 Qc7 15 Qe3 Rxb2 16 Rc1 Qb8 17 Ne1 Be7 18 Nd3 Rb6
Black was pressing throughout the whole game in T.Geier-De Homont,
corr. 2015 and won in the end.

14...Rxb6

White has to weaken the c3-square, which leads to long-term suffering.

15 b3 Rc6 16 0-0?! f6!

Black adds tension to the centre and prepares to keep his king in the
centre. After all, we are approaching the endgame.

17 Rd1 Kf7 18 Ne1 Be7 19 Nd2 Rhc8 20 Ndf3

Black has a slight but stable advantage thanks to his two bishops and
control of the open c-file.

20...Rc3 21 Rd3 Bb5 22 Rxc3 Rxc3 23 g3 a5 24 Kg2 Ba6 25 Rd1 a4

26 Rb1? (D)

Black to play
26...a3!

Black fixes the a2-pawn as a target.

27 Rd1 Rc6 28 h3 Bb4 29 Rb1 Be2 30 exf6 gxf6 31 g4 Kg6 32 Ng1 Bb5
33 Ngf3 Rc8 34 Kg3 Bf1 35 Nh4+ Kf7 36 Rd1 Bd6+ 37 f4 Rc3+ 38 Nhf3
Bb4 39 Ra1 Kg6 40 h4 h5 41 g5 Bd6 42 Kf2 Bh3 43 Ng1 Bg4 44 Ne2
Rh3 0-1

Material is equal, but White has been completely outplayed and his pawns are
about to fall.

Theory Magnifiers

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 (5 Be3 Nd7 6 Nf3)

15A: (5 Qg4)

15B: (6 Nf3)

White’s continuations in these magnifiers are relatively rare and nothing

to lose sleep over.

In 15A I suggest a fun and promising way of playing for Black, and I am
sure you will enjoy looking at these lines.

15B is quieter, but you are probably more likely to face this variation.
Once again I suggest a combative approach that should end up in a
balanced position.

15A: 5 Qg4 and Minor 5th Moves

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6)

5 Qg4

This looks odd, as the queen doesn’t really attack anything, but Black
cannot yet capture the c5-pawn.

5 Bd3 is a cunning try by White as 5...Bxc5?! 6 Qg4 Ne7 7 b4 Bb6 8 Nf3

gave him the upper hand in Vitiugov-C.O’Donnell, Batumi Olympiad 2018.
However, by delaying the capture on c5 with 5...Nd7 Black gets a good
game after 6 Nf3 Nxc5 7 0-0 (D).
 Black to play
Then:

a) The Azeri grandmaster Guliev defended the position after 7...Ne7 four
times in the 1990s, getting good positions each time, and it was repeated
in Andriasian-Epishin, Villorba 2019, where Black also achieved an
advantage.

b) However, I think 7...Nxd3 is more precise. Black played very accurately

in the following game: 8 Qxd3 Bd7 9 c4 dxc4 10 Qxc4 Rc8 11 Qe2 Qa5 12
Nc3 Bb4 13 Bd2 Ne7 14 Rfc1 0-0 15 a3 Bxc3 16 Bxc3 Qb6 17 Ng5 Bb5 18
Qh5
½-½ S.Olsen-Martin Molinero, corr. 2016. Black could have played on
with 18...Bd3, though the game remains equal.

5...h5 (D)
 White to play

This is of course somewhat weakening but we shall see lines where the
advanced h-pawn is very useful, and it’s not at all clear where White’s
queen belongs.

6 Qg3

This is the preference of Vachier-Lagrave, but 6 Qf4 is more common,

allowing the recapture of the pawn by 6...Bxc5. After 7 Bd3 Nc6 8 Nf3
Nb4 9 0-0 Nxd3 10 cxd3 Ne7 Black has scored well.

6...Nc6 7 Nf3 h4

The idea is to force the white queen to make a decision right away.

8 Qf4

8 Qg4?! Nh6 9 Bxh6 Rxh6 is good for Black:

a) Against 10 Nc3, as in Vlassov-Sestiakov, Moscow Ch 2001, 10...f5!,

exploiting White’s exposed queen, is better for Black. Compare with line ‘b’.
b) 10 Bd3 should also be met by 10...f5!; e.g., 11 Qf4 (11 exf6 Qxf6 is
excellent for Black, who will quickly seize the centre with ...e5, C.Carton-
Renet, French Team Ch 2001) 11...Bxc5 12 Nc3 Bd7 and Black dominates.
If 13 0-0-0?!, Black can grab the pawn with 13...Bxf2, while 13 0-0 Be7 with
the idea ...g5 leaves Black better.

8...Bxc5 9 Bd3 (D)

Black to play

Black has many decent options in this position, but I can’t resist giving
the following move.

9...f5

This looks crazy, but White is the one who must be precise. If White plays
quietly, Black can most likely continue to pile pressure onto the e5-pawn
with ...Nh6-f7 and possibly ...Qc7. And yes, if possible, Black can
eventually consider rolling over White by playing ...g5. Now:

a) 10 exf6?! gxf6 11 Nxh4 e5 12 Bg6+? Kd7 and the engine evaluates this
as a decisive advantage for Black.
b) 10 g4 Nge7 11 gxf5 Nxf5 12 Qg4 (12 Bxf5? 0-0! –/+) 12...0-0 13 0-0?!
Qc7 =+ Anokhin-De Homont, corr. 2015.

15B: 6 Nf3 and Minor 6th Moves (1 e4

c6 2 d4 d5 3 e5 c5 4 dxc5 e6)

5 Be3 Nd7 (D)

White to play

6 Nf3

Or:

a) 6 f4 has been played in a handful of games, including Durarbayli-Indjić,

Saint Louis 2018. Here, before capturing on c5, Black should include
6...Nh6; e.g., 7 Nf3 Nf5 8 Bf2 Bxc5 9 Bxc5 Nxc5 and Black is at least
equal.

b) 6 c4 Bxc5 7 Bxc5 Nxc5 8 cxd5 exd5 and now:

b1) 9 b4 Ne4 10 Bb5+ Kf8 11 Qd4 Ne7 12 Ne2 Nf5 13 Qd3 a6 14 Ba4 Qb6
with equal chances in Tanabe-Sethuraman, corr. 2018, which concluded 15
0- 0 Qxb4 16 Bb3 Nc5 17 Qxd5 Be6 18 Qf3 Nh4 19 Qe3 Nf5 20 Qf3 Nh4
21
Qe3 ½-½.

b2) 9 Nc3 Ne7 10 b4 seems to be a speciality of Chyzy, who has played it

in three games. 10...Ne6 (chosen by two of his opponents) is fine, but I
like 10...Ne4 11 Nxe4 dxe4 12 Qxd8+ Kxd8 =+, as in Chyzy-Tokarczyk,
Police 2019, since White’s e-pawn will quickly come under attack.

b3) 9 Qd4 Qa5+ 10 b4 Ne6 11 Qd2 Qb6 12 Nf3 Ne7 = Pessoa-Dhanish,

corr. 2014.

6...Bxc5 7 Bxc5 Nxc5 (D)

White to play
8 Qd4

From here the queen hits the knight and is ready to move to the g4-
square. 8 Nbd2 does not pose Black any real problems: 8...Ne7 9 Nb3
Bd7 10 Nxc5 Qa5+ 11 Qd2 Qxc5 is very similar to a French Defence.
Here Black has almost completed development and is ready to exchange
his bad bishop with
...Bb5. 12 0-0-0 has been played in five games, which were all drawn.
After 12...Rc8 13 h4 (Nefedov-Krzyzanowski, European corr. Team Ch
2017) I suggest the logical 13...0-0 14 h5 h6. Black will continue with
...Bb5 and/or
...f6 with equality.

8...Qc7 (D)

This is a combative move, although rarely played. 8...Qb6 is more

common and probably sufficient for equality, but 9 a4 is somewhat
challenging.

White to play

9 Nc3

Or:

a) 9 c4 is best met by 9...Ne7, when 10 cxd5 Nxd5 11 Nc3 0-0 12 Nxd5

exd5 is fine for Black. 13 Qxd5? Rd8 14 Qc4 Be6 15 Qc3 Qb6 followed by
...Rac8 or ...Ne4 gives Black a decisive initiative.

b) 9 Nbd2 Ne7 10 Nb3?! was tried in Milliet-J.Houska, Internet blitz

2020. After 10...Nd7, with ...Nc6 coming up, Black is already better due
to the
weak e-pawn.

c) 9 b4 Ne4 10 Nbd2 Nxd2 11 Qxd2 Ne7 12 Bd3 Ng6 13 0-0 Nxe5 14

Nxe5 Qxe5 15 Bb5+ Bd7 16 Bxd7+ Kxd7 17 c4 Rhc8 18 f4 Qd6 19 cxd5
Ke8 = Egelstaff-B.Mulligan, corr. 2017.

d) 9 Qg4 f5 (I prefer this over 9...g6, which was played twice in a 2020
Milliet-J.Houska Internet blitz match) 10 Bb5+ Kf7 11 Ng5+ Kf8 12 Qd4 (12
Qg3? Qa5+ 13 Nc3 d4 –+) 12...h6 13 Nf3 Ne7 14 0-0 was played in
Zielasko-Enescu, corr. 2016. Here I suggest 14...Ne4 15 Nbd2 (15 c4 a6
16 cxd5 axb5 17 d6 Qc5 18 dxe7+ Kxe7 =) 15...Kf7 16 Nb3 g5 with a
highly complicated position where Black’s chances should be no worse.

9...a6 10 Bd3

Or 10 Qg4 f5 11 exf6 Nxf6 12 Qg3, when both 12...Qxg3 and 12...Qe7

lead to equality.

10...Ne7 11 0-0 Bd7

Chances seem balanced; e.g.:

a) 12 Ne2 h6 prepares castling without allowing a bishop sacrifice on

h7. Chances are equal.

b) 12 a4 Rc8 13 Rfd1 Nxd3 14 Qxd3 Ng6 15 Qe3 0-0 16 Rd2 (De

Vilder- T.De Jong, Wijk aan Zee 1999) 16...f6 gives Black a comfortable
position.

Lesson 16: Arkell/Khenkin Introduction to 5 a3

By: TRH

Intelligence is what you use when you don’t know what to do.

Jean Piaget

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3
This move is usually preparation for b4 but may be played at various points
and combined with natural developing moves, most notably Nf3 as in
Lesson
14. Some care is required to avoid being tricked by White’s various
move- orders.

5...Bxc5 (D)

In her book, Houska prefers 5...Nc6, presumably to induce 6 Nf3, which

blocks the Qg4 option for White. I don’t think that’s strictly necessary, and
as we shall see, leaving the knight on b8 for some time has its points.

White to play

One minor advantage of the immediate capture is that all Shaw has to
say about it is that “6 b4 Bb6 7 Nf3 is likely to transpose whenever
...Nc6 is played.” As we shall see, there’s a little more to it than that.

Model Game 16

Volokitin – J. van Foreest

Bundesliga 2017/18
1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 Nf3

6 b4 is the main move and the subject of Lesson 17. For now we shall
only mention that the immediate advance has the drawback of giving
Black more options about where to move his bishop. Our repertoire move
is 6...Be7, but both 6...Bb6 and 6...Bf8 are valid.

With the move chosen, White hopes to play a later b4 under better conditions.

6...Ne7

White’s capture on c5 weakened his e-pawn, so we should attack this

pawn as quickly as possible. First on the agenda is to get the knight to
g6.

7 Bd3 (D)

White’s only real alternative to this move is 7 b4, which is covered in 16B.

Black to play
7...Ng6
7...0-0?? is an embarrassing mistake as White has the standard Greek
Gift sacrifice 8 Bxh7+!, which is winning for White with correct play.

8 0-0

8 Nbd2 Nc6 practically forces White to capture on g6, which is favourable

to Black.

8...Qc7! (D)

I believe this is the strongest move. 8...0-0 has been played in many high-
level games, but White rarely replies with the critical 9 Ng5 Be7 (9...h6?
10 Nxf7! and 9...Qc7 10 Qh5 h6 11 Nf3! leave White with dangerous
attacking possibilities on the kingside) 10 Qh5 Bxg5 11 Bxg5 and I
believe White has the better chances.

White to play
9 b4

Once again, White can consider delaying or omitting b4:

a) 9 Re1 0-0 (9...Nc6?! is strongly met by 10 b4; in general, it’s a good
idea to wait with ...Nc6 until White can no longer gain tempi with b4-b5)
10 Nc3 a6 11 Be3 Be7 12 Na4 Nd7 13 Bd4 b5 14 Nc3 should be no
problem for Black. I suggest 14...Nb8 to challenge the bishop on d4 with
...Nc6.

b) Similar is 9 Nc3 a6 10 Re1 but with b4-b5 no longer possible, Black can
play 10...Nc6, which gives White issues with his e5-pawn. 11 Bxg6 should
be met by 11...hxg6. Capturing this way obviously makes sense, to use the
half- open h-file. Black can continue with ...Rh5 to exert pressure on the e-
pawn. The king can be brought to safety by ...Kf8 and possibly later go to
g8. Black’s bishops and solid position give him at least equality.

We now return to 9 b4 (D):

Black to play
9...Be
7

One of the reasons why we hurried with the development of the knight to
g6 was to leave the e7-square for the bishop.

With the queen occupying c7, 9...Bb6?! is inferior. White gets the upper hand
by forceful play: 10 a4 a6 11 a5 Ba7 12 b5 +/–.

10 Bb2 0-0 11 g3

One of the drawbacks of developing the bishop to b2 is that the f4-square

is unprotected.

11 Nbd2 a5 12 b5 Nd7 13 Re1 Nc5 (I believe Black is fine in structures

like 13...b6 14 g3 Bb7, as in T.Craig-Andeer, corr. 2013) 14 Bf1 b6 15 g3
Bb7 gives Black a slight advantage, even if White went on to win in
Carnicelli- Palozza, Italian Junior Ch, Salerno 2018.

11...a5 (D)

White to play

An idea to remember! This is the main reason why we delayed ...Nc6. We

are aiming for a stable square on c5.

12 b5 Nd7 13 Qe2 Nc5 14 h4

This is a typical reaction to a knight on g6, but a well-known principle is that

you should have a stable centre when going for a flank attack, which
explains Black’s next move.

14...f6

Not the only move, but now things become sharp.

15 Bxg6 hxg6 16 exf6?!

16 Nbd2 is a safer choice, leaving the game approximately equal.

16...gxf6 (D)

White to play

Black is better thanks to his strong pawn-centre and bishop-pair.

17 c4?! Nb3!

At first sight this looks like a mistake, but Black has everything figured out.

18 Qd3 Bc5! 19 Qxb3 Qxg3+ 20 Kh1 Qh3+ 21 Kg1 b6

The light-squared bishop is about to join the action, practically deciding
the game in Black’s favour.

22 Qd1 Bb7 23 Nh2 Bd6 24 f4 Bc5+ 25 Kh1 dxc4+ 26 Rf3 Rad8 27 Qe2
Rd3 28 Nd2 Rxd2 0-1

A crushing attack!

Theory Magnifiers

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 (6 Nf3)

16A: 6 Qg4

16B: (6 Nf3)

The following two magnifiers are very interesting, although they are
two completely different concepts.

16A is critical, as White immediately targets the weak g-pawn. I know that
strong players like playing this way with the white pieces, so I advise you
to pay extra attention to this line.

16B is also challenging but not because of forced attacking lines. The
reason is that it seems like White has tricked us away from our desired
set-up, but there is a downside to White’s play as well.

16A: 6 Qg4

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5)

6 Qg4

This is a popular move. As young and ambitious talents tend to go for

this line, it has to be treated seriously.

6...Ne7 (D)
This is the way to play, although both 6...g6 and 6...Kf8 are playable
options. New to this line, I chose 6...Bf8 against Neikšans, but I won’t
repeat that.

White to play
7 Nf3

This move has been played twice by former World Junior Champion
Aryan Tari. I know he does his homework, and probably he has found
some potential in White’s position. Other ideas:

a) Taking the pawn with 7 Qxg7?! gives White nothing. A few games have
featured 7...Ng6, which is unproblematic for Black, but this is a nice
moment to go our own way. The engine also gives 7...Rg8 a stamp of
approval, and in a game it would have been my choice. Black will get full
compensation for the two pawns; for instance, 8 Qxh7 Qb6 9 Nh3 Nbc6 10
f4 Bd7 with a position that resembles a French Winawer, and it’s useful to
look for differences. In the French, White reaches a similar position with a
worse structure, while Black has given up his dark-squared bishop. I think
the latter point is the more important. After 11 Nc3 the engine evaluates
most reasonable options for Black as equal, whereas White must often find
very
precise moves to avoid being worse. This indicates that the position is
easier to play for Black. Now 11...0-0-0 12 Qd3 Kb8 is razor sharp. The
engine says it’s completely equal, despite White’s material advantage. I’m
a fan of the initiative, and would certainly take Black.

b) 7 b4 is strongly met by 7...h5! 8 Qxg7 (8 Qh4?! Qc7 9 Bb2 Ng6 10 Qg3

h4 11 Qb3 Bb6 –/+ and Black won in Gulbis-T.Thomas, corr. 2017) 8...Rg8,
which seems fine for Black:

b1) 9 Qh7?! Bd4 10 Ra2 Bxe5 11 Qxh5?! (A.Liang-Santarius, Sitges

2016) and the engines say 11...Bf6 gives Black an almost decisive
advantage.

b2) After 9 Qh6 Bd4 10 Ra2 Bxe5 11 Bd3 Nd7 12 Nf3 a5, Black was at
least equal in Fabig-U.Baumgartner, corr. 2016.

c) In case of 7 Bd3, I propose 7...0-0. If 8 Nf3 we transpose back to the

main line, while against 8 b4 I have prepared a shocking move. The
following line is a proof that working with engines can open your eyes to
new ideas. 8...f5!? (8...Bb6 is a sound move, but why not take the driving
seat?) 9 exf6 (D).

Black to play
Now at first I liked the amazing 9...Ng6?!, but in the final analysis it leaves
Black fighting for equality after 10 f7+!. Fortunately, 9...Bxf2+! 10 Kxf2
Rxf6+ is a good option. Black has only one pawn for the knight, but his
strong centre and White’s permanently exposed king provide excellent
compensation. 11 Ke1 (Black regains the piece after 11 Nf3?! e5 12 Qh5
e4, when White is the one fighting for equality) 11...e5 and now:

c1) 12 Qh5 e4 13 Bb5 (after 13 Bg5? Rf5 14 Bb5? Bd7! 15 Bxd7 Qxd7
there is no good way to prevent ...h6, so Black is much better) 13...Nbc6
14 Ne2 Qb6 with good compensation.

c2) After 12 Qh4 it took me a while to land on the slow but effective move
12...h6. Black defends against White’s threat, and takes control of the g5-
square. 13 Ne2 (13 Bb2 Nf5 14 Bxf5 Bxf5 15 Nf3 Nc6 gives Black sufficient
compensation, and the main question remains where the white king will
hide) 13...Nf5 14 Bxf5 Bxf5 15 Rf1 Nd7 and I believe Black is doing fine. In
the sample line 16 Nbc3 Bxc2 17 Rxf6 Nxf6 18 Bxh6 gxh6 19 Qg3+ Kf7 20
Rc1 Bb3 21 Qxe5 Qe7 White has returned the material, and we are
approaching an equal endgame.

We now return to 7 Nf3 (D):

 Black to play
7...0-0

It is noteworthy that having played this line as Black (choosing 7...Qb6 in

Topalov-Navara, Shamkir 2019), Navara switched to the white side as
recently as August 2020 in Navara-Toman, Czech Ch, Plzen 2020 (where
he faced 7...Ng6). However, in my opinion, castling immediately is better,
and hopefully will come as a surprise, as it has only been played in one
grandmaster game.

8 Bd3

If 8 b4 Bb6 9 Bd3 (9 Bb2 can be answered the same way) I want to go for
the uncompromising 9...Ng6 10 Bb2 Nd7 11 0-0 f5. This is a key move in
this line. Now:

a) 12 Qg3 Bc7 and thanks to the threat of playing ...Nf6-e4, Black was
doing fine in Jarmula-Licznerski, Polish Junior Team Ch, Karpacz 2013.
However, it must be said that it seems like both players missed this idea
during the game.

b) After 12 exf6 Nxf6 13 Qg5 Ne4, the engine slightly prefers White, but I
believe Black’s activity is enough for equality. Black ended up winning in
V.Colin-Rausis, Sautron 2018, but with the recent cheating scandal in
mind, it’s difficult to trust the result.

8...f5 (D)
 White to play
9 Qg3

Or 9 exf6 Rxf6 10 Qh5 h6 11 0-0 Bd6, and now:

a) After 12 c4 Nbc6 13 Nc3 Bd7 14 Be3 Rc8 Black should be doing

fine, with ...Be8 coming.

b) 12 Bd2 Nbc6 13 Bc3 gives Black several options, but 13...e5 equalizes:
14 Nxe5 d4 15 Nxc6 Nxc6 16 Bb4 Bxb4 17 axb4 Nxb4 18 Be4 Qf8 with the
idea of playing ...Bf5.

c) 12 b4 Nd7 13 Bb2 and now 13...e5 leads to some nice tactics. The key
point is that after 14 Nxe5? Nxe5 15 Bxe5 g6, to avoid losing a piece,
White must play 16 Qe2 Re6 17 f4 Nc6 –/+.

9...Ng6

This is better than 9...Nbc6, as played Tari-Sadzikowski, Reykjavik

2019, since Black stays flexible on how to develop his b8-knight.
Additionally, White has to look out for ...f4.
10 0-0 Nd7

I can’t see Black being worse in this position. White has to be careful not
to lose his e-pawn. Additionally, Black has ideas of moving the bishop
from c5 to prepare ...Nc5-e4. The position is equal.

16B: Delayed b4 lines

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5)

6 Nf3

6 Bd3 Ne7 7 b4 Bb6 is similar to the main line in this lesson, but delaying
Nf3 gives White the option of a more aggressive set-up. 8 Qg4 is
interesting. Then 8...Ng6 9 Nf3 reaches a position known from other move-
orders. 9...0-0 transposes to the note to White’s 8th move in 16A, but I
favour 9...f5, the main point being that after 10 Qh5 (Wang Yue-Li
Yunshan, Makati women 2018) 10...a5 the move 11 Ng5 is harmless due
to 11...Nc6, when 12 Nxh7?
Nxe5 gives Black a winning advantage.

6...Ne7

Against 6 b4, I recommend 6...Be7 in Lesson 17. Here White can force
the bishop to b6, but there are a couple of downsides.

7 b4 Bb6 (D)
 White to play

It may seem that White has achieved something by forcing us to place

our bishop on b6 rather than on e7, like we do in Lesson 17. However,
White’s problem is that his knight has gone to f3, which means the
queen can no longer reach g4. This saves us a lot of trouble.

8 Bd3

Or 8 Bb2 a5, and now:

a) After 9 Bd3 axb4 10 axb4 Rxa1 11 Bxa1, I like 11...Na6 as Black

immediately targets the b4-pawn. Additionally, with the pawn on b4,
White has weaknesses on the queenside. Then:

a1) After 12 c3 Ng6 13 0-0 0-0 14 Nbd2 Bd7 15 Re1 (Nanjo-Dragnev,

Batumi Olympiad 2018) I like 15...Nc7, aiming for the b5-square. For
instance, 16 c4 dxc4 17 Nxc4 Bb5 and the position is equal with a lot of
play left.

a2) 12 Bc3 d4 (Black has alternatives, but this equalizes with ease) 13
Bxd4 Nxb4 14 Bxb6 Qxb6 15 0-0 Nxd3 16 Qxd3 0-0 17 Rd1 and the
game was
suddenly drawn in Packroff-Tornow, corr. 2017. Black is at least not worse.

b) 9 Nbd2 axb4 10 axb4 Rxa1 and now:

b1) 11 Bxa1 was played in Durarbayli-Ganguly, Ulaanbaatar 2018.

Here I suggest 11...0-0 12 Bd3 Ng6 13 g3 Na6 and with moves like
...Qe7 and
...Bd7 on the agenda, the position is OK for Black.

b2) 11 Qxa1 0-0 12 Be2 (12 Bd3 Na6 13 c3?!, Laube-De Homont, corr.
2016, and after 13...Bd7 with the idea ...Nc7, Black is not worse) 12...Na6
13 Qa3 Bd7 14 0-0 Nc7 = D.Olofsson-Holroyd, corr. 2015.

We now return to 8 Bd3 (D):

Black to play

8...a5 9 b5

This is not dangerous. Otherwise:

a) 9 Bb2 transposes to note ‘a’ to White’s 8th move above.

b) 9 Nc3 is a tricky move, hoping to jump quickly to b5 and d6. However,
we have a counter-trick ready: 9...Nbc6! 10 Rb1 (on 10 b5?, Black’s idea is
to pick up a pawn by 10...Nxe5! 11 Nxe5 Bd4 –/+) 10...axb4 11 axb4 Ng6
practically forces 12 Bxg6, when both ways to recapture look slightly better
for Black.

9...Nd7 10 0-0

After 10 Bb2 Ng6 11 0-0 (Hector-Tikkanen, Swedish Ch, Ronneby

2018) Black should play 11...a4 =.

10...a4 11 Bd2

This was played in Slovak-Landa, Czech Team Ch 2018/19. Here Black

should target the e5-pawn: after 11...Qc7 12 Qe2 Ng6 13 Bxg6 hxg6,
the position is probably objectively equal, but I wouldn’t mind playing
Black, who can typically continue with moves like ...Rh5, ...Nc5 and
...Bd7.

Lesson 17: Arkell/Khenkin Main Battleground

By: TRH

Beware of false knowledge; it is more dangerous than ignorance.

George Bernard Shaw

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 b4 (D)

 Black to play

This is the main move, immediately forcing Black to decide where to

place his dark-squared bishop.

6...Be7

I spent a lot of time on 6...Bb6. After all, this is the most common move and
fits well with our repertoire. However, even if I didn’t stumble on any real
theoretical problems, there were definitely some practical challenges. The
line that worried me the most was 7 Qg4 Ne7 8 Nf3 0-0 9 Bd3 Ng6 10 0-0.
White has dangerous attacking prospects on the kingside, which is why
Black has to get rid of the white e-pawn to connect the two flanks.
Following 10...f5 11 exf6 Qxf6 a long series of ‘only’ moves leads to a
beautiful draw in the end: 12 Qh5 Nd7 13 Ng5 h6 14 Bxg6 Qxa1 15 Bf7+
Kh8 16 Nxe6 Nf6 17 Nxf8 Nxh5 and now 18 Ng6+ is a fascinating draw.
However, Black has no chances to deviate, and I suspect that White has a
few other tries. I would be somewhat worried in this line, which is why I
recommend 6...Be7.

7 Nf3

This natural developing move is the most common. For White’s alternatives,
including the critical 7 Qg4, see 17A.

7...a5!?

One of the disadvantages of playing b4 is that Black has this move,

which gives him access to the important c5-square.

Model Game 17

Degraeve – T.R.

Hansen Internet rapid

2018

1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 b4 Be7 7 Nf3 a5!? (D)

White to play

Rambaldi doesn’t like this move but I think it’s fine as long as Black
knows what he is doing.

8 b5 Nd7
This is an excellent square for the knight.

9 Bd3

9 c4 may be critical. For analysis of this and some minor 9th moves, see 17B.

9...f6

This is Black’s main idea in this line. We shall see that White will
have problems keeping control of the e5-square.

10 Bb2 Nh6 (D)

White to play

I was the first to bring this position to daylight, and by now I have had
it twice. Both my opponents played the ‘obvious’ move, but it’s a
mistake.

11 0-0?!

Or:
a) 11 Qe2 is a better try, but shouldn’t worry Black. After 11...0-0 12 0-0
Nf7 13 exf6 Bxf6 14 c4 Nc5 15 Nc3, as in Warakomska-Giacomini,
Hoogeveen 2018, Black should play 15...d4 16 Ne4 Nxd3 17 Qxd3 e5,
when he has nothing to complain about.

b) 11 Nbd2! (D) is more accurate than castling, because the rook on

a1 is now protected by the queen as well.

Black to play

The position after 11...Nf7 12 exf6 has rarely occurred, but is critical. After
12...Bxf6 13 Bxf6 Qxf6 14 c4! (14 0-0? is a mistake as it transposes back
to the main game, and so allows Black to play 14...e5! –/+) 14...Nc5 15
Bc2 0-0 16 0-0, the move 16...a4! (as played in Agopov-Feuerstack, 2nd
Bundesliga 2019/20) at first sight looks a little strange, but the idea is to
dominate the knight on d2. Now the black knight will be able to stay on c5
long-term as Nb3 is no longer possible for White. After 17 cxd5 exd5 we
have a non- standard isolated queen’s pawn, but I believe Black’s chances
are pretty good.

11...Nf7
This is Black’s point. White is forced to take on f6.

12 exf6

12 Re1? fxe5 13 Nxe5 Ndxe5 14 Bxe5 Nxe5 15 Rxe5 Bf6 and Black wins
the exchange.

12...Bxf6 13 Bxf6

I was pleased to see that Ushenina has reached this position twice with the
black pieces. After 13 Nc3 Nc5 14 Re1 0-0 15 Bf1 a4 16 Rb1 Qd6 the
games split: 17 Na2 Bxb2 18 Rxb2 e5 –/+ Lei Tingjie-Ushenina, Hengshui
blitz 2019 and 17 g3 Bd7 18 Bg2 Rac8 =+ Abdumalik-Ushenina, Hengshui
blitz 2019.

13...Qxf6 (D)

White to play

White’s big problem in this position is that he doesn’t have time to play 14
c4 as the rook on a1 is hanging.
14 Nbd2?! e5!

Much stronger than the automatic 14...0-0?, which I played the first time I
got this position. This gave White the time he needed to play 15 c4! in
Fressinet- T.R.Hansen, Porticcio 2017. Black’s position is not that bad, but
this e-pawn advance is much stronger. Black is dominating completely
thanks to his powerful centre. First of all, White has to deal with the threat
of ...e4.

15 Be2 e4 16 Ne1

Now one more vital move is required for Black to gain a clear advantage.
Once again, White has to play c4 to challenge Black’s centre, and this is
what I had to stop.

16...0-0?

16...Nc5!, with the idea of meeting 17 c4 with 17...d4, is a dream come

true for Black, but in a rapid game I failed to find this.

17 c4 Nde5 18 cxd5 Nd6

I have sufficient compensation, but no more than that. In the end White
won the game. With a short time-limit, both players made some mistakes
and the rest of the game is of no theoretical interest.

19 Nc2 Bd7 20 a4 Qg5 21 Ra3 Nd3? 22 Nxe4 Nxe4 23 Rxd3 Rac8 24 Ne3
Bh3 25 Bg4 Nc3 26 Bxh3? Nxd1 27 Rfxd1?! Rcd8 28 g3 Qf6 29 R3d2
Qc3? 30 d6 Qb4 31 Rd4 Qb2 32 R1d2 Qa3 33 d7 g6 34 Kg2 Kg7 35
Bg4 h5 36 Be6? Qe7? 37 Bh3 Qf6 38 Nd5 Qf3+?!

The final mistake. After 38...Qe5 it’s still possible to fight on.

39 Kg1 g5 40 Nc7 g4 41 Ne6+ Kh8 42 Bg2 Qa3 43 Nxf8 Qxf8 44 Bxb7

Qc5 45 Bc6 Kg7 46 Kg2 Qf5 47 Rf4 Qc5 48 Re4 Qc3 49 Re8 Qf6 50 Rxd8
Qxd8 51 Re2 1-0

Theory Magnifiers
1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 b4 Be7 (7 Nf3 a5!? 8 b5
Nd7 9 c4)

17A: (7 Qg4)

17B: (9 c4)

We have reached the Arkell/Khenkin main lines, indicating that your

opponent may be well prepared. If you don’t want him to have a head
start, these two magnifiers are mandatory study material. You should at
least play through the main moves and have a look at any sideline that
looks intimidating. Both 17A and 17B are critical, although I believe Black
has every reason to be happy.

17A: 7th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 b4 Be7)

7 Qg4

Once again this move is critical. Rarer options:

a) 7 Bb2 a5 8 b5 Nd7 9 Nf3 transposes to note ‘a’ to White’s 9th move

in 17B.

b) After 7 Bd3 a5 8 b5 Nd7 I can’t see how White can avoid transposing
back into the main line. For 9 Nf3 see Model Game 17.

c) 7 c4 doesn’t seem theoretically challenging but has been recommended

by Dreev. A model response is 7...a5 8 b5 Nd7 9 Bb2 (9 Nf3 is the main
line of 17B) 9...Nh6 10 Nf3, transposing a position we see in note ‘a’ to
White’s 9th move in 17B.

We now return to 7 Qg4 (D):

 Black to play
7...g6

This is probably best, but I have spent a considerable time analysing

the riskier 7...g5!?, which you may wish to explore.

8 Bd3

Or:

a) There are a few games with 8 Nf3 a5 9 b5 Nd7 but most have
transposed to the main line of this section after 10 Bd3.

b) 8 h4? was chosen in Kardoeus-Kolb, Willingen 2019, but it’s just a bad
move. Black gets the upper hand by playing 8...a5 9 b5 Nd7 10 Nf3 (10
Bb2 f6 11 Nf3 Nh6 12 Qf4 Nf7 is much better for Black as he is about to
build a huge centre) 10...Qc7 with a double attack on c2 and e5.

8...a5 9 b5 Nd7 (D)

 White to play

This position has occurred in a number of correspondence games. It’s

worth noting that the ‘defenders’ of the black side often play this line
repeatedly, indicating that it may be trusted.

10 Nf3 h5 11 Qf4

11 Qg3 h4 12 Qf4 has been played in a few games. A safe and sensible
option is then the solid 12...b6 (as used in a couple of correspondence
games), when Black should have very few problems, and may even be
better already. But if you want something more ambitious and
unbalancing, I can suggest the great-looking novelty 12...f5!? (D).
 White to play

If White plays slowly, Black will manoeuvre the g8-knight to f7, putting
pressure on the e5-pawn and preparing ...g5 while the d7-knight has
an excellent post on c5. Then:

a) 13 exf6?! is strongly met by 13...Bxf6 14 Bxg6+?! Kf8. It takes some

guts to play this way with the black pieces, but it helps to know that the
engines give Black a significant plus. Black is about to build a big centre
with ...e5.

b) 13 Nd4 g5 leads to a unique position. It’s all about the weak e5-pawn.
After 14 Qe3 Rh6 Black threatens ...f4, and White has to act. One
possible line is 15 f4?! gxf4 16 Qxf4?! Qc7 17 Nf3 Rg6 –/+.

c) 13 h3 Nh6 14 Qe3!? (more accurate than 14 0-0 Nf7, when Black’s

position has great potential) and now Black can change tack with 14...Nc5
(heading for e4, while White must be careful not to allow ...g5 at the wrong
moment; instead after 14...Nf7?! 15 Bb2 Black’s counterplay with ...g5 is
not possible for tactical reasons) 15 0-0 (15 a4 is strongly met by 15...g5!,
when 16 Nxg5?! Ne4 17 Nxe4 fxe4 18 Be2 Nf5 gives Black great
compensation; another try is 15 Nbd2, when after 15...g5 16 0-0 {16
Nxg5?? d4 –+} 16...g4
Black is at least no worse) 15...Ne4 (D) and now:

White to play

c1) 16 Nbd2 g5 17 Nd4 Qc7 is an incredibly sharp position. Obviously

both players have many options, but the following are possible lines: 18
Rb1 (or 18 c4 Qxe5 19 Bb2 g4 20 Rab1 Bg5 21 Qe2 Qf4 22 cxd5 Qxd2
23 Bxe4 Qxe2 24 Nxe2 Rg8 25 Bd3 gxh3 26 gxh3 Bf6+ 27 Kh2 Bxb2 28
Rxb2 e5 =) 18...Qxe5 19 Bb2 Qf4 20 Qxf4 gxf4 21 Nxf5 Nxd2 22 Nxe7
Kxe7 23 Bxh8 Nf7 24 Bc3 Nxb1 25 Rxb1 e5 and Black’s strong centre
provides good compensation.

c2) 16 c4 g5 17 Bxe4 fxe4! 18 Nh2 Nf5 19 Qe2 Qc7 20 Ng4 Nd4 is also
far from clear; e.g., 21 Qb2 Qxc4 22 Be3 Ne2+ (rather than 22...Nf5?! 23
Nd2 Qd3 24 Nf6+ Bxf6 25 exf6, when White makes inroads) 23 Kh2 Bd7
24 a4 Nf4 25 Nd2 Qb4 26 Qa2 Nd3 is a computer line where Black’s
queen emerges unscathed, though Black will still have to defend
accurately.

We now return to 11 Qf4 (D):

 Black to play
11...Nh6

Alternatively 11...Nc5 with the same idea; e.g., 12 a4 Nh6 13 0-0 b6 14

Bb2 h4 15 Nd4 Bb7 16 Qf3 Nf5 17 Nd2 Qc7 18 N2b3 h3 and in
E.Moskalenko- Dittmar, corr. 2016 the position was strategically complex,
with equal chances.

12 0-0

12 a4 is answered in the same way: 12...b6 13 0-0 transposes to the main line.

12...b6

Or 12...Nc5 13 Nc3 b6 14 Nd4 (14 Ne2 Bb7 15 a4 Nf5 16 Bb2 Rc8 17 Ned4
Ne4 18 Rfd1 Bf8 19 Nxf5 gxf5 20 Bd4 Bh6 = Karpenko-U.Baumgartner,
corr. 2015) 14...Bb7 15 Nc6 (White tries to strike immediately since Black
will be very comfortable if he is allowed to complete development)
15...Bxc6 16 bxc6 Qc7 and now:

a) After 17 Rb1 Nxd3 18 cxd3 Qxc6 19 Nb5 Nf5 20 Be3 0-0 White had
compensation for a pawn, but nothing more than that, in
Poleshchuk- R.Mayer, corr. 2012.

b) 17 a4 Nf5 18 Ba3 Kf8 19 Rfd1 Qxc6 20 Bxf5 gxf5 ½-½ S.Colin-

Rüfenacht, corr. Olympiad 2016.

13 a4 Bb7 14 Ba3

14 Nbd2 g5 15 Qe3 Bc5 16 Qxg5 Qxg5 17 Nxg5 Bd4 18 Rb1 Bxe5 19

Nb3 Rc8 20 f4 Bc3 =+ Bekemann-U.Baumgartner, corr. 2016.

14...Nc5 (D)

White to play

The nice thing about this line is that it gives room for creativity, with
many possible set-ups and attacking chances. Now:

a) 15 Nbd2 Rc8 16 Rfd1 g5 17 Qe3 Qc7 18 Bb2 Rg8 was complex but
equal in F.Poetz-De Homont, corr. 2015.

b) 15 Rd1 g5 and in A.Ilyasov-De Homont, corr. 2017 we saw a very inspired

attack by Black: 16 Qe3 g4 17 Nd4 Bg5 18 Qe1 g3 19 hxg3 h4 20 gxh4
Bxh4 21 g3 Ng4 and after quite a few adventures the game ended in a
perpetual check.

17B: 9th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 c5 4 dxc5 e6 5 a3 Bxc5 6 b4 Be7)

7 Nf3 a5!? 8 b5 Nd7 (D)

White to play
9 c4

White opens the position. Otherwise:

a) 9 Bb2 Nh6 10 c4 (10 Bd3 f6 transposes to Model Game 17, while 10

Nbd2 f6 11 Bd3 is note ‘b’ to White’s 11th move in that game) 10...0-0 is
very interesting:

a1) 11 cxd5 exd5 12 Qxd5? (this pawn is poisoned; 12 Be2 Nf5 13 Nc3
Nb6 is interesting and probably equal) 12...Nxe5 13 Qxe5 Bf6 –+.
a2) 11 Be2 dxc4 12 Nbd2 c3 13 Bxc3 b6 14 0-0 Bb7 (Black is at least
equal) 15 Bb2 Rc8 16 Re1?! Nc5 and although Tomczak-Bartel, Polish Ch,
Warsaw 2018 was won by White in the end, at this point Black has a
pleasant advantage.

b) After 9 Be2, 9...f6 seems consistent:

b1) Although it has never been played here, 10 c4 is always a move to

be taken seriously. Black can play 10...fxe5 11 cxd5 exd5 12 Qxd5 Ngf6
and now the only critical move seems to be 13 Qa2 as it stops Black
from castling. However, 13...Nc5 comes with multiple ideas: ...Nd3+,
...Be6 and even ...Qd5. Black has no problems and the position is
roughly equal.

b2) C.Yoo-Ostrovsky, Charlotte 2019 continued 10 exf6 Bxf6 11 Ra2,

when White’s position is probably not as suspicious as it looks. Here
Black should play 11...Ne7 followed by ...0-0, ...b6 and ...Bb7 with a fine
game.

9...Qc7 (D)

White to play
10 Bf4
10 cxd5? Nxe5 11 Bb2 Nxf3+ 12 Qxf3 Nf6 leaves Black better.

10...dxc4 11 Be2 b6 12 Nbd2 c3 13 Nc4 Bb7

Black has two ways of handling this position: either ...Bc5 to free e7 for
the knight, or ...Nh6.

14 Rc1

14 Qd3 Bc5 15 Nd6+ Bxd6 16 exd6 Qc8 17 Ne5 Ngf6 18 Rc1 0-0 19 0-0
Nd5 = D.Malin-Ovcharenko, corr. 2016.

14...Bxf3 15 gxf3 Bc5 16 Rxc3 Ne7 17 Nd6+ Kf8 18 Rc2

Rambaldi stops here, concluding that White has an attack.

18...Rd8

Black is threatening 19...Ng6, leaving White’s knight in trouble.

19 Qc1 h6 (D)
 White to play

With the idea of playing ...g5. This way we provoke a further weakness
in White’s camp, which is especially relevant if White decides to castle.

20 h4

I’m not sure if this is strictly necessary, but it puts pressure on Black.

20...Nxe5 21 Bxe5

After 21 Ne4 Rd5 White has compensation for the pawn in the shape
of Black’s uncastled king, but this can easily go downhill for White as
well.

21...Rxd6 22 Qf4 Nf5

Black seems to be doing well:

a) 23 Bxd6+?! Nxd6 24 a4 Nxb5 =+.

b) 23 Bd3 Qd7 24 Bxd6+ Qxd6 25 Qxd6+ Nxd6 and I wouldn’t mind playing
this endgame with the black pieces.

Lesson 18: Experimental Line: Johnsen Variation (3...Qc7)

By: SJ

Accursed creator! Why did you form a monster so hideous that even
you turned from me in disgust?

Frankenstein, Mary Wollstonecraft Shelley

1 e4 c6 2 d4 d5 3 e5

Finding an alternative line against the Advance Variation proved quite

challenging. Having rejected moves like 3...Na6, 3...g6 and 3...h5 as
leading
to cramped positions that are probably very good for White and even
harder for Black to play well in practice, we ended up with a very rare
move – it has only been tested in three games according to our databases:

3...Qc7 (D)

White to play

From c7 the queen observes the important c5- and e5-squares and thus
prepares ...c5 while retaining the option of ...Bf5 or ...Bg4 depending on
White’s play. If you are considering 3...Qc7 as a way to minimize theory
study, I am sure you are wondering how much detail you need in order to
play it as a surprise weapon. Well, that depends on the level of your and
your opponent’s play. If you regularly meet opponents who are able to
come up with critical lines even when surprised, your first task is to decide
whether the line is actually playable as a regular part of your repertoire.

Model Game 18

Hansch – Baldauf

Karlsruhe 2019
1 e4 c6 2 d4 d5 3 e5 Qc7

This rare move is not even mentioned in Dreev and Shaw’s repertoire
books, and it is only possible to write a full lesson on it thanks to
transpositions to other lines where ...Qc7 comes somewhat later. This
makes the lesson somewhat like a Frankenstein’s monster, stitched
together from pieces of other lines.

4 Nc3

It is far from clear that this is White’s best, but it’s a fairly natural reply and
quite likely to lead to positions that have been tested in practice before.
The same can be said about 4 Ne2, which we discuss in 18A (together
with a couple of more minor options) and 4 Bd3, preventing ...Bf5, in 18B.
It’s possible that 4 c4 is a strong move, but it has not been tested much. It
can be found in 18C.

Here we shall just point out that although 4 Nf3 Bg4 is arguably what Black
is aiming for, he must be careful not to play ...c5 too early as in some lines
White may blast everything open with c4. Some sensible lines are 5 Be2
e6 6 0-0 Nd7 7 Nbd2 Ne7 8 c4 dxc4 9 Nxc4 Nf5 10 Ne3 Nxe3 11 fxe3 Be7
+= and 5 Nbd2 e6 6 h3 Bh5 7 Be2 Ne7 8 0-0, when Black must be wary of
opening the game in lines like 8...c5?! 9 dxc5 Nd7 10 c4!, and prepare
...c5 with the safer 8...Nd7, when he has reasonable chances.

4...Bf5

This position more frequently arises after the moves 3...Bf5 4 Nc3 Qc7.
From that move-order it has been played by Peralta, Dreev (in blitz) and a
very young J.van Foreest. Black’s main idea is to delay ...e6, so that if
White plays an early g4, the bishop may fall back to d7.

5 h4 (D)

For White’s alternative 5th moves, including 5 g4, see 18D.

 Black to play

This game actually (and quite typically) started out from a different move-
order. Only after 3...Bf5 4 h4!? (Shaw’s recommendation) 4...Qc7!? 5 Nc3
(leaving Shaw’s repertoire) did the game reach the diagram position.
Another possible move-order is 4 Nc3 Qc7 (as mentioned above) 5 h4.

There are more than 30 games with this position – some of them
between very strong grandmasters.

Q: So we didn’t avoid main-line theory after all?

Well, assuming that we had recommended 3...Bf5, then after 4 Nc3,

4...Qc7 would definitely have been a way to avoid the main lines as it’s a
rare reply (with fewer than 20 games). But 4 Nc3 is far from White’s only
option in that line.

5...h5

As a matter of fact, the somewhat peculiar 5...a6 is Black’s most popular

option here. The idea obviously is to play ...c5 without having to worry
about Nb5.
6 Nge2

It looks natural to develop a new piece with some future kingside threats.
For alternatives, including the equally natural 6 Bd3, see 18E.

6...e6 7 Ng3 Bg6 (D)

White to play

I believe Black has at least full equality here and quite possibly the
easier position to play.

8 Be2

This may be inaccurate, but even 8 Be3 Nd7 9 Bd3 Bxd3 10 Qxd3 c5
may not be fully satisfactory for White.

8...c5 9 f4?!

Or 9 Be3 Nc6 =+.

9...cxd4 10 Qxd4 Nc6 11 Qa4 0-0-0

Black already has an obvious advantage thanks to White’s
weakened kingside and misplaced g3-knight.

12 Bd3 Bxd3 13 cxd3 Nh6 14 Be3 d4 15 Nb5 Qb6 16 Bc1 Rd5 17

Na3 Bb4+ 18 Ke2 f6 19 Nc4 Qc7 20 Ne4 Kb8

White has been gradually outplayed by his somewhat stronger IM

opponent and is now clearly losing.

21 exf6?!

Nominally White will get enough pawns for his piece but the open lines
and White’s exposed king strongly favour Black’s active forces.

21...b5 22 Qb3 bxc4 23 dxc4 d3+ 24 Kd1 Rd4 25 Nc5 Qb6 26 fxg7 Rg8
27 Nxe6 Ng4 28 Rf1 Re4 29 c5 (D)

Black to play
29...Re1+!

I don’t find the dual 29...Nf2+! 30 Rxf2 Re1# distracting at all.

30 Rxe1 Nf2# (0-1)

Theory Magnifiers

1 e4 c6 2 d4 d5 3 e5 Qc7 (4 Nc3 Bf5 5 h4 h5 6 Bd3)

18A: (4 Ne2)

18B: 4 Bd3

18C: 4 c4

18D: (5 g4)

18E: (6 Bd3)

This line may be considered an anti-theory weapon, and there is no

guarantee that it will hold up against GM scrutiny. Therefore your first step
should probably be to look for a genuine refutation. And if there is one, it is
most probably hidden in 18B or 18C. When (or if) you are satisfied that the
line is safe to play, you should quite possibly test it in online blitz games
before digging into the details of the other magnifiers.

18A: 4 Ne2

(1 e4 c6 2 d4 d5 3 e5 Qc7)

4 Ne2

This doesn’t look particularly natural but there are actually quite a few
games with this move coming from the move-order 2 Ne2 d5 3 e5 Qc7 4
d4.

Other options:

a) 4 Nd2 c5 5 dxc5 Qxe5+ 6 Be2 Qd4 7 Ngf3 Qxc5 8 0-0 Nf6 9 c4 e6

10 Nb3?! Qd6 11 cxd5 Nxd5 =+ Dotsenko-Rupel, St Petersburg 2012.
b) 4 c3 c5 and in addition to 5 Bd3, transposing to 18B, White has
the following options:

b1) 5 Nf3 Bg4 6 dxc5 Bxf3 7 Bb5+ Nc6 8 Qxf3 Qxe5+ 9 Be3 Nf6 10 0-0 e6
11 Nd2 +=.

b2) 5 Be2 Nc6 is of extra interest to us, as it’s a reasonable transposition

from Lesson 13 (from the move-order 3...c5 4 c3 Nc6 5 Be2 Qc7). Black is
OK after 6 Nf3 Bg4 7 Nbd2 e6 8 0-0 cxd4 9 cxd4 Nge7 10 h3 Bh5 11 a3 a5
12 Re1 Nc8 13 Nf1 Be7 14 Ng3 Bg6 15 h4 h6 =.

b3) 5 Be3 Nc6 6 f4 (6 Bd3!?) 6...cxd4 7 cxd4 Bf5 8 Nc3 e6 9 g4 Bg6

10 Nge2 +=.

4...c5 (D)

White to play

5 c4

This looks critical. Otherwise:

a) 5 c3 Nc6 can hardly be a problem for

Black: a1) 6 a3 Bf5 7 Ng3?! cxd4 8 cxd4

Nxd4 =+.

a2) 6 Bf4 looks clumsy and isn’t much of a problem if you are OK with
6...e6 and a French kind of position. Black has lost a tempo with his c-
pawn, but White’s pieces are poorly placed. Instead 6...Qb6?! 7 dxc5!
works out badly for Black, while 6...Bg4?! 7 dxc5 is also tricky.

b) After 5 Bf4, 5...e6 6 c4 was a little tricky for Black in Vilchez Barroso-
Rodriguez Lopez, Andalucian Ch, Granada 1991. Instead, quite typically
for this line, Black should go for the critical 5...Nc6 6 e6 Qa5+ 7 Bd2 Qc7 8
exf7+ Kxf7, when his chances certainly are not worse and his loss of
castling rights is of little consequence.

5...dxc4 6 Bf4 g5! (D)

White to play

Like it or not, this crazy-looking move is more or less forced if Black

wants his full share of play. Now:
a) 7 Be3 Nc6 8 Qa4 Bd7 =+.

b) 7 Bxg5 Nc6 8 Nbc3 Be6!? 9 d5 0-0-0 =+.

c) After 7 Bg3 Nc6 play gets quite irrational as the computer lines are not
very forcing but hard for humans to understand. For instance, 8 Nec3 (8
Nbc3 Be6 9 Qa4 0-0-0 10 d5 Bxd5 11 Nxd5 Rxd5 12 Nc3 Rd8 13 Bxc4 e6
=+) 8...cxd4 9 Bxc4 Nh6 10 Nb5 (10 e6? Qa5 11 exf7+ Nxf7 12 Qh5 e6 –+)
10...Qa5+ 11 b4 Nxb4 is roughly equal. After, e.g., 12 0-0 Bg7 13 e6 0-0 14
exf7+ Nxf7 15 Nd2 we are ready for a complex middlegame. According to
Stockfish, chances are still level, and Black has half a dozen playable
moves.

18B: 4 Bd3

(1 e4 c6 2 d4 d5 3 e5 Qc7)

4 Bd3

White logically prevents ...Bf5 but the move is a little slow and
encourages Black’s alternative idea.

4...c5 (D)
 White to play
5 c3

This position could also arise from 4 c3 c5 5 Bd3.

5 Nf3 Bg4 should probably transpose to line ‘a’ of the next note after 6
c3 Nc6, as 6 Nc3 e6 7 Nb5 Qb6 seems fine for Black.

5...Nc6 6 Ne2

Or:

a) 6 Nf3 seems inconsistent and 6...Bg4 7 0-0 e6 looks very comfortable

for Black.

b) 6 Be3 Qb6 (6...Nh6!?) 7 Qd2 cxd4 and White must choose:

b1) 8 Bxd4 Nxd4 9 cxd4 e6 10 Nc3 Bd7 11 Nge2 Ne7 12 0-0 and
12...g6!? could be Black’s best attempt to make something out of his
strong dark- squared bishop.
b2) 8 cxd4 Nb4 9 Be2 (9 Ne2 Nxd3+ 10 Qxd3 Bd7 11 Nbc3 e6 12 0-0 Ne7
13 Rac1 Rc8 =) 9...Bf5 10 Na3 should be met by the simple 10...e6, when
11 Nf3 Nc6 12 Nb5 Bb4 13 Nc3 is an odd transposition to a position where
Black is comfortable and has scored well in a few games, including Ilijin-
Zelčić, Nova Gorica 2003.

We now return to 6 Ne2 (D):

Black to play

White is trying to prevent Black from activating his light-squared bishop, but
his play hasn’t been particularly energetic, and it is tempting for Black to
play à la Gurgenidze, with moves like ...g6, ...f6, ...Nh6, ...Bf5, ...g6 and
...Nf7/f5 on Black’s agenda, though the question is how to implement this.
Black can also settle into a position akin to lines of the French, while
seeking to make use of the fact that he hasn’t yet blocked in his queen’s
bishop.

6...Bg4

Or:

a) The immediate 6...g6?! seems too slow as the strategically desirable 7 b4!?
(Black is also hard-pressed to equalize after 7 0-0, though the danger is
less imminent) works well tactically because after 7...cxb4 (7...c4 keeps
things closed, but is grim for Black) 8 cxb4 Nxb4 9 Qa4+ Nc6 10 Nbc3 e6
11 Nb5 Qd8 12 Ba3 White will establish a piece on d6.

b) 6...f6?! actually attempts to slow down the play by encouraging White to

block lines with 7 f4. Then 7...Nh6 8 0-0 Bf5!? may be playable for Black
as the sharp 9 Bxf5 Nxf5 10 g4 Nh6 11 e6 Nxg4 12 h3 Nh6 13 dxc5 Rd8
14 Na3 Nf5 is far from clear. However, keeping lines open with 7 0-0!
seems more dangerous, even if 7...fxe5 8 dxe5 Bg4 9 c4 d4 10 f3 Bd7
might not be too bad for Black.

7 f3

Otherwise Black will achieve his goal of reaching a French-type

position with the bad bishop either outside the pawn-chain or
exchanged off.

7...Bd7 8 Be3 (D)

8 0-0 Qb6 9 Kh1 cxd4 10 cxd4 (10 f4!?) 10...Nb4 eliminates

White’s important light-squared bishop.
 Black to play
8...Qb6

This is a critical line, though Black can also play the quiet 8...e6, with
a reasonable form of French (e.g., 9 f4 Qb6!?), or go bishop-hunting
with 8...cxd4 9 cxd4 Nb4.

9 e6

Or 9 dxc5 Qxb2, when 10 e6 Bxe6 transposes to our main line, while 10

f4!? Qxa1 11 Qd2 e6 12 0-0 f6! will lead to an unclear middlegame with Q
vs 2R.

9...Bxe6 10 dxc5 Qxb2

Now:

a) White may offer his rook for a few moves with 11 0-0 Bd7 12 Nd4
Nxd4 13 cxd4, but after 13...g6 14 Nd2 Qa3 15 Qe2 Bg7 there doesn’t
seem to be any great danger for Black.

b) 11 Nd2 Qa3 12 Nb3 a6 13 0-0 g6 14 Nf4 Bg7 looks OK for Black too.
His queen is exposed and there are lines where White gets a draw by
perpetually attacking it.

18C: 4 c4

(1 e4 c6 2 d4 d5 3 e5 Qc7)

4 c4

This is what we feel may be a strong player’s instinctive reaction.

4...Bf5 (D)

Immediately allowing the light-squared bishop to c4 with 4...dxc4? 5

Bxc4 Bf5 is too risky because of 6 e6.
 White to play

Now Black just needs to fortify his centre with ...e6 and he can be happy
with his prospects.

5 h4

This leads to a position known from the move-order 3...Bf5 4 h4 Qc7 5 c4.
After the following untested alternatives, Black seems to maintain a
playable game:

a) 5 Bd3 Bxd3 6 Qxd3 dxc4 7 Qxc4 e6 8 Nc3 Ne7 9 Nf3 +=.

b) 5 Nf3 e6 6 Nc3 Bb4 7 Be2 dxc4 8 Bxc4 Ne7 9 0-0 +=.

c) 5 cxd5 cxd5 6 Nc3 e6 transposes to line ‘d’.

d) 5 Nc3 e6 6 cxd5 cxd5 could be best. White has ideas with Nb5 and
Rc1, which initially look a bit scary. Black may soon have to play a
prophylactic
...a6. Still, that’s rarely a wasted move in the French and the fact that
Stockfish ranks 7 a3 as its top choice at moderate search depths indicates
that Black may look forward to playing a French Defence with his light-
squared
bishop outside his pawn-chain. An alternative continuation is 7 Be3 Nc6 8
Rc1 a6 9 Nf3 and now perhaps 9...h6!? 10 Be2 Nge7 11 0-0 Bh7, when
Black seems solid and ready to complete kingside development with ...Nf5
and either ...Be7 or ...g5 and ...Bg7.

5...e6 (D)

White to play
6 Nc3

Or 6 cxd5 cxd5 7 Nc3 h6, and then:

a) 8 Nge2 Nc6 9 Ng3 Bh7 10 Bd3 Bxd3 11 Qxd3 (Westphal-M.Arnold,

Internet blitz 2020) and now 11...Bb4 offers Black sufficient counterplay by
either bringing the knight to a5 and c4, or by playing ...f6; e.g., 12 Bd2 (12
0- 0 Qd8 makes the earlier h4 advance look a little silly) 12...Bxc3 13 Bxc3
f6 14 0-0 (14 exf6 Nxf6 15 Qg6+ Qf7 =) 14...fxe5 15 dxe5 Nge7 with an
unbalanced game.

b) 8 Bd3 Bxd3 9 Qxd3 Nc6 10 Nge2 (A.Chandra-Margvelashvili,

Richardson juniors 2015) and now 10...Bb4 equalizes.
6...h6

6...dxc4 7 Bxc4 h6 and now:

a) 8 g4 Bh7 9 f4 Nd7 10 Nge2 Ne7 11 Ng3?! 0-0-0 12 0-0?! (after 12 h5

the result is open) 12...h5 –/+ Pokazanjev-Dreev, Russian Team Ch,
Olginka 2011.

b) 8 Nge2 Nd7 9 Ng3 Bh7 10 Nge4 Nb6 11 Bb3 Nd5 (11...Rd8 12 Be3
Ne7 13 Rc1 Nf5 14 g4 Nxe3 15 fxe3 Nd5 16 Qf3 Nxc3 17 Nxc3 Bb4 18 0-
0 +=) 12 Nxd5 exd5 13 Nc3 and in Kadrić-V.Dobrov, Split 2013 Black
could have tried 13...Ne7, when White is only slightly better after 14 h5.

We now return to 6...h6 (D):

White to play

7 Nge2

Alternatively:

a) 7 g4 Bh7 8 Bd3 Bxd3 9 Qxd3 dxc4 10 Qxc4 Nd7 =.

b) 7 Be3 Ne7 8 Rc1 dxc4 9 Bxc4 Nd7 10 Nge2 a6 11 Ng3 Bh7 12 f4 Nb6
13 Bb3 0-0-0 14 Qf3 Kb8 15 0-0 += T.Schulze-D.Abel, Verden an der Aller
2020.

c) 7 Bd3 Bxd3 8 Qxd3 dxc4 9 Qxc4 Ne7 (9...Nd7 10 Nf3 Ne7 11 0-0 Nb6 12
Qb3 Ned5 13 Ne4 Be7 = Sengupta-T.Tao, Canberra 2018) 10 Bd2 Na6
11 Nf3 Qb6 12 Ne4 Nd5 13 0-0 Qb5 is equal as Black is for choice after
14 Qxb5?! cxb5 15 Rfc1 Kd7.

7...Ne7 8 Ng3 Bh7 9 Be3 Nd7 10 Rc1

Black is somewhat cramped but has a grip on d5. Now:

a) 10...Qa5 11 c5 f6?! 12 exf6 gxf6 13 Be2 0-0-0 14 0-0 was difficult

for Black in Havelka-Solozhenkin, Marianske Lazne 2017.

b) 10...dxc4 11 Bxc4 Nb6 12 Bb3 Ned5 13 Qg4 0-0-0 14 0-0 Kb8 looks
fully playable for Black.

18D: 5th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 Qc7)

4 Nc3 Bf5 (D)

 White to play

In this position we saw 5 h4 in Model Game 18.

5 g4

This asks Black to demonstrate his idea. Otherwise:

a) 5 f4 is probably too slow. After, e.g., 5...e6 6 g4 Bg6 7 Bd3 Bxd3 8

Qxd3 h5 White must be careful not to become overextended (as 9 f5
hxg4 doesn’t seem to lead anywhere).

b) 5 Be3 a6 6 g4 Bd7 7 h4, as in Lorenzo de la Riva-Peralta, Montcada

2013, should be met by 7...e6. Then White may be objectively better after
8 f4 c5 9 Nf3 cxd4 10 Nxd4 Nc6 11 Qd2, but Black’s more compact
position looks fine after 11...h5 or even 11...f6.

5...Bd7 6 Be3

Or:

a) 6 f4 h5 7 f5 hxg4 8 Qxg4 (8 e6? fxe6 9 Qxg4 exf5 10 Qg6+ Kd8 11 Nf3

Qd6 12 Ne5 {Vlassov-Dreev, Internet blitz 2004} looks messy but 12...Be8!
13 Qxf5 Qf6 gives Black a clear plus) 8...Nh6 9 Bxh6 Rxh6 followed by ...e6
leads to unclear play; e.g., 10 0-0-0 (10 Bd3 and 10 h4 give Black the extra
option of 10...c5!?) 10...e6 11 Bd3 (11 fxe6 Bxe6 12 Qg3 can be met by
12...c5!? =, while 11 h4 exf5 12 Qf3 Na6 13 Nh3 Rxh4 14 Bxa6 bxa6 15
Ng5 Rg4 is rather unclear) 11...Na6 12 a3 0-0-0 with interesting play for
both sides.

b) 6 Bd3 e6 is untested. Some plausible lines: 7 f4 c5 8 Be3 cxd4 9 Bxd4

Nc6 10 Nb5 Qd8 11 Bf2 a6 looks OK for Black; 7 Nce2 c5 8 c3 Nc6 9 Be3
(9 Nf3 invites Black to break up the centre with 9...f6) 9...h5 10 g5 g6 11
Nf3 Nge7 12 0-0 +=.

6...e6 7 f4 c5 8 Nf3 a6

Objectively this should be slightly better for White but I wouldn’t mind
playing Black, and after 9 a3 Qb6 10 Bg2?! (10 dxc5 may be best) 10...h5
11 gxh5 Qxb2 12 Bd2?! cxd4 13 Nxd4 Qb6 Black already had a clear plus
in S.Hoolt-J.van Foreest, Lüneburg 2015.

18E: 6th-Move Alternatives

(1 e4 c6 2 d4 d5 3 e5 Qc7 4 Nc3 Bf5)

5 h4 h5 6 Bd3

Or:

a) 6 Nce2 c5 7 dxc5 Nc6 should be fine for Black: 8 c3 (8 Nf3 e6 9 Ned4?!

Nxd4 10 Nxd4?! Qxe5+ 11 Be3 Bxc5 12 Bb5+ Kf8 –/+ T.R.Hansen-Turov,
Fagernes 2014) 8...e6 9 b4 a5 10 Qa4 Qc8! 11 Qb5?! g6 12 Nd4 Ne7 13
Bg5 Bg7 =+.

b) 6 Be2 e6 shouldn’t trouble Black either:

b1) 7 Bxh5?! c5 8 Be2 Nc6 (it seems Black even has time for 8...a6) 9
Nf3 Bg4 10 dxc5 Bxc5 =+.
b2) 7 Be3 c5!? (7...a6 8 Nf3 c5 =+ looks simpler) 8 Nb5 Qa5+ 9 b4?!
(White should settle for 9 Qd2 Qxd2+ 10 Kxd2, but 10...cxd4! 11 Nxd4 Ne7
12 Nb5 Kd7 13 Nd6 f6 14 Nf3 Nbc6 looks equal) 9...cxb4 10 Nf3 Nc6 11 0-
0 a6 12 Nd6+ Bxd6 13 exd6 Nf6 14 Ne5 Rc8 –/+ G.Jones-Baldauf,
Internet blitz 2020.

6...Bxd3 7 Qxd3 e6 (D)

White to play

Black is behind in development, but the position is closed and he has

the better bishop.

8 Nf3 a6

8...Ne7 seems a sound alternative and 8...c5!? could be worth a try.

9 Ne2 c5 10 0-0!?

This looks like the most dangerous move. In contrast 10 c3 seems

harmless after 10...Nc6:
a) 11 Bf4 cxd4 12 Nexd4 Bc5 13 0-0 Nge7 14 Nb3 Ba7 15 Rfe1 Nf5 16
Re2 Rc8 =+ Coro-Erenburg, Montevideo 2018.

b) 11 0-0 cxd4 (11...Nh6?! 12 Bxh6 Rxh6 13 Nf4 cxd4 14 cxd4 Be7 15 a3

Kf8 16 b4 g6 17 Rfc1 += Samant-Turov, Taleigao 2019) 12 cxd4
Nb4 followed by ...Qc2 will simplify into a queenless middlegame
that is comfortable for Black.

10...Nc6 11 c4

This may be a critical position. Now:

a) 11...Nb4 12 Qb3 dxc4 13 Qxc4 Rc8 (13...b5 14 Qc3 Nd5 15 Qc2 c4 16 a4

+=) 14 Bg5 b5 15 Qc1 += J.Christiansen-P.Neuman, Prague 2020.

b) 11...Nge7 12 dxc5 Nxe5 13 Nxe5 Qxe5 may improve: 14 Bg5 (or 14

Rd1 Qe4 15 Qxe4 dxe4 16 Nc3 f5 17 Bg5 +=) 14...f6 15 Be3 Nf5 16 cxd5
Qxd5 17 Qxd5 exd5; e.g., 18 Rac1 Nxh4 19 Nf4 Nf5 20 Nxd5 Nxe3 21
fxe3 Rc8 22 b4 Kf7 23 e4 +=.
4: Exchange Variation and Panov Attack

Lesson 19: Exchange: Introduction to the Carls Variation

By: TRH

Do not bring your Queen out too early.

Francisco Bernardina Calogno (On the Game of Chess, circa 1500)

1 e4 c6 2 d4 d5 3 exd5

This introduces the Exchange Complex, which consists of two

rather different sub-systems – the Classical Exchange and the
Panov Attack.

3...cxd5

Many ‘Exchange Variations’ are rather boring, but in the Caro-Kann the
pawn-structure immediately becomes asymmetrical and Black doesn’t
have to worry about how to create counterchances.

4 Bd3

White controls the f5-square. Black is now struggling to find a good square
for his light-squared bishop. Instead of going for forced lines, White is
happy to play a certain structure. I have always found this line a little
annoying, as I found it harder to get the dynamic play I like. However, in
this lesson you will see that it’s possible to play something fun, and at the
same time sound.

4...Nc6 5 c3 Qc7! (D)

 White to play

This move forces White to find a Plan B, as Bf4 needs extra preparation.
It was introduced with great success by Carl Carls at the Baden-Baden
tournament in 1925.

It is undoubtedly the most combative choice. With his early queen

development, Black aims to prevent White from finding the best squares
for his pieces. Ideally, White would like to put the bishop on f4 and the
knights on f3 and d2.

White should avoid an early Nf3, as this allows Black to play ...Bg4.
Therefore, White should focus on the development of his queenside, and
that’s what our last move is directed against. Now 6 Bf4 is impossible,
while 6 Nd2 would block the dark-squared bishop.

In this lesson we shall take a look at White’s alternatives to the main move,
6 Ne2, which is the subject of Lesson 20, as well as White’s other
development plans on move 4.

Model Game 19
 Zviagintsev –
Aronian

Berlin rapid 2015

1 e4 c6 2
d4

2 Nf3 d5 3 exd5 cxd5 tends to transpose into lines we discuss under

‘Minor 4th Moves’ in 19A. The independent and quite original 4 Ne5 is
seen from time to time, and has even been used by some very strong
players. After 4...Nc6, 5 d4 e6 6 Bb5 Bd7 transposes to 19A.

2...d5 3 exd5 cxd5 (D)

White to play
4 Bd3

This is a popular line and a common suggestion in books attempting to

offer a complete 1 e4 repertoire for White in a limited space. Maybe as a
consequence of this, it’s also a surprisingly popular choice in online blitz.
White can develop naturally without much theoretical knowledge, and in a
fair number of lines apply some pressure to Black’s kingside. Instead, 4 c4
is
the Panov Attack (Lessons 21-24), and for rarer 4th moves, see 19A.

4...Nc6 5 c3

5 Nf3 is exactly what Black wants and is met by 5...Bg4.

5...Qc7!?

5...Nf6 is most common and should lead to an equal game. But I

recommend this different and in my opinion more fighting interpretation.

6 h3

White prevents all ...Bg4 lines. We shall return to the main move, 6 Ne2, in
Lesson 20. For the quite logical 6 Na3 and some minor alternatives, see
19B.

6...g6 (D)

White to play
Black is intending ...Bf5.
7 Nf3

Having spent time playing h3, this is the only consistent move.

7...Bf5 8 0-0

White can exchange on f5, but he will then have a hard time putting
pressure on Black’s doubled pawns or making use of the semi-outpost on
e5, as his bishop can’t reach f4, and his queen’s knight will have to do a
lot of manoeuvring to find a good square. 8 Bxf5 gxf5 9 0-0 (9 g3 e6 10
Bf4 Bd6 11 Bxd6 Qxd6 12 Nbd2 Nf6 13 Qe2 Ne4 14 Nxe4?! fxe4 15 Nh4
0-0-0 =+
Kulhanek-Velička, Czech Team Ch 2010/11) 9...Nf6 and then:

a) After 10 Re1 e6 Black is already comfortable. If White doesn’t come up

with something special, Black will continue with ...0-0-0, ...Ne4 and
...Rg8. 11 Ne5 (11 c4 is similar to line ‘b’) 11...Nxe5 12 dxe5?! (the
nuance 12 Bf4 avoids a disadvantage) 12...Ne4 left Black better in
Saveliev-Rodshtein, Russian Team Ch, Sochi 2019.

b) 10 c4! is probably best, and forces Black to adjust to the change in

structure. 10...dxc4 11 Nc3 0-0-0 leads to a highly complicated situation,
with chances for both sides. 12 Qa4 Rg8 13 Qxc4 e6 14 a4 a6 15 Bg5 Be7
16 Rac1 Kb8 17 Bxf6 Bxf6 18 d5 Bxc3 19 dxc6 Bf6 20 Qe2 Rd6 21 cxb7
Qxb7 22 Rc4 Bxb2 23 Rb1 Rb6 and Wiersma-Windecker, corr. 2014 was
eventually drawn.

8...Nf6

There is nothing wrong with 8...Bxd3 as it will most likely transpose to

the game. But if you want to give White a second chance to take on f5,
you should go for the game move.

9 Re1

Or:

a) 9 Bxf5 gxf5 transposes to the note to White’s 8th move.

b) Black was fine after 9 Be2 Bg7 10 Nbd2 0-0 11 a3 Rad8 12 Re1 Rfe8 in
Giri-Cheparinov, St Petersburg blitz 2018.

9...Bxd3 10 Qxd3 (D)

Black to play

The exchange of light-squared bishops can be considered a success for

Black. However, White has pressure down the e-file, explaining Black’s
next move.

10...e6

Another option is 10...Bg7 11 Be3 0-0, as in L.van Foreest-Tari, Wijk

aan Zee 2018.

11 Nbd2

11 g3 with the idea 12 Bf4 is best met by 11...Bd6.

11...Bg7 12 Nb3

White achieves nothing by 12 Nf1 0-0 13 Ng3 – with a pawn on g6, g3

is typically not a good square for a white knight. 13...a6 14 a4 Rab8 15
Ne2
Rbe8 was fine for Black in Kevlishvili-Oparin, St Petersburg blitz 2018.

12...0-0 13 Be3

I don’t really see the need for 13 a4. After 13...Rfe8 14 g3 Ne4 15 Bf4
Black can slowly expand on the kingside; e.g., 15...Qd8 16 h4 h6 17 Kg2
g5 18 hxg5 hxg5 19 Be3 f5 gave him a dangerous kingside initiative in
Ponomariov-Le Quang, Dortmund 2010.

13...Ne4 14 Ng5 (D)

Black to play
14...Nd6

Black could also play 14...Nf6 with a silent draw offer, or exchange
with 14...Nxg5. Either way, Black is fine.

15 Bf4 b6

Stifling the knight on b3.

16 Qg3 Rad8 17 Qh4 h6 18 Nf3 g5

Black is not afraid of ghosts.

19 Bxd6

19 Bxg5? hxg5 20 Nxg5 Rfe8 gives White nothing like enough for the
piece. It is well-known that it’s hard to mate an opponent with a bishop on
g7.

19...Rxd6 20 Qg3 Rd7 21 Ne5

21 Qxc7 Rxc7 is slightly more comfortable for Black, as it’s difficult for
White to come up with a decent plan. Black, on the other hand, looks
forward to launching a queenside minority attack.

21...Nxe5 22 dxe5 f6 23 Nd4 Re7 24 Qg4 Qc4 25 Nf3 Qxg4 26 hxg4 Rc8
27 Rad1 Kf7 28 exf6 Bxf6 29 Ne5+ Bxe5 30 Rxe5 (D)

Black to play

This rook endgame is objectively equal, but Aronian managed to outplay

his opponent.
30...Rc4 31 f3 b5

Finally, the minority attack starts! White is bound to end up with a

weakness. One cannot expect perfect endgame play in a rapid game, but
clearly it is Black who has all the chances.

32 a3 a5 33 Kf2 b4 34 axb4 axb4 35 Rh1 Kg6 36 Re3 Rc6 37 Kg3 bxc3 38

bxc3 Rec7 39 Re2 Ra6 40 Re3 Rcc6 41 Rhe1 Rab6 42 Kf2 Kf6 43 Rh1
Rb2+ 44 Kg3 Kg6 45 Rhe1 Rbb6 46 Kf2 h5 47 gxh5+ Kxh5 48 g4+ Kg6
49 Kg3 Kf7 50 Rh1?! Rb3 51 Re5? Rcxc3 52 Rf1 Kf6 53 Re2 e5 54 Kg2
d4? 55 Ra1 Ra3 56 Rb1 Rcb3 57 Rc1 Ke6 58 Rc5? Re3 59 Rxe3 Rxe3
60 Kf2 Rc3 61 Ra5 Rc2+ 62 Ke1 Kd6 63 Rb5 Rc5 64 Rb6+ Kd5 65 Rf6
Rc2 66 Kd1 d3 67 Rf5 Kd4 68 Rxg5 Ra2 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 exd5 cxd5 (4 Bd3 Nc6 5 c3 Qc7 6 h3 g6 7 Qc2)

19A: (4 Nf3)

19B: (6 Na3)

19C: (7 Qc2)

The Classical Exchange lines are basically shortcuts for White, frequently
recommended to complete a repertoire more focused on the Sicilian and
the Open Games. Therefore you are not likely to stumble into deep and
dangerous analysis in any of the magnifiers of this lesson. Going through
19A will make you more familiar with move-order issues, which are quite
important in this Lesson.

Both 19B and 19C are logical attempts by White to challenge Black’s
idea behind placing the queen on c7. Facing these lines unprepared
wouldn’t be the end of the world, but they both have some sting, so it
might be mildly annoying.
19A: Minor 4th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5)

4 Nf3

This is the kind of common-sense move that you frequently encounter at

club level. The same can be said about some of White’s other 4th-move
alternatives:

a) 4 Nc3 looks wrong. If White wants the knight on c3, he should make
sure to advance the pawn to c4 first. Now the knight is misplaced and it’s
hard to challenge Black’s d-pawn. After 4...Nc6 5 Nf3 Bg4 6 Bb5 e6
White’s position is not that bad, but it’s worth noting that Black scores
massively. 7 h3 (7 0-0 Bd6 8 h3 Bh5 =+) 7...Bxf3 8 Qxf3 Bd6 =.

b) 4 c3 has the idea of not revealing the light-squared bishop’s intentions.

The downside of White’s move is that the f5-square is available to Black.
White’s only attempt to punish the natural reply 4...Bf5 is to target the b-
pawn: 5 Qb3 (5 Bf4 is line ‘c’, while 5 Nf3 Nc6 gives White nothing better
than transposing with 6 Qb3 Qd7) 5...Qd7 6 Nf3 Nc6 7 Bb5 at first sight
looks mildly uncomfortable for Black, but it’s nothing to worry about. After
7...f6 (simple chess, covering the e5-square) 8 0-0 e6 the position is
interesting with approximately equal chances, Bieganowski-D.Sikorski,
Polish Team Ch 2008/9.

c) As I will recommend a set-up with the queen on c7, a well-

prepared opponent might try to trick us with 4 Bf4 (D).
 Black to play

Incidentally, this is a position that can also be reached from the move-
order 1 d4 d5 2 Bf4 c5 3 e3 cxd4 4 exd4, so this is where Win with the
London System meets Win with the Caro-Kann! Once again we use the
fact that the f5-square is available: 4...Bf5 5 c3 Nc6 6 Qb3 Qd7, and now:

c1) 7 Nf3 f6 8 Nbd2?! (8 h4 e6 9 Nbd2 transposes to line ‘c2’) 8...g5! 9

Be3 (or 9 Bg3 h5! –/+ Wang Yue-Bartholomew, Internet rapid 2019) 9...e6,
as in Hamark-Engqvist, Swedish Team Ch 2008/9, looks very attractive for
Black.

c2) 7 Nd2 e6 8 Ngf3 (neither 8 Bb5 nor 8 Ne2 looks troublesome for Black)
8...f6 and here 9 h4, as played in Naroditsky-Erdos, Saint Louis 2019, is
intended to prevent Black from playing ...g5. I suggest 9...Nh6 as the
knight can later go to g4 or be rerouted into the game via f7 and d6. 10
Bxh6?! gxh6 is better for Black, thanks to the bishop-pair.

4...Nc6 (D)

4...Bg4 is also possible. Probably White should reply 5 h3 and obtain

the bishop-pair, although this is not the end of the world in this line.
Anyway, let’s play traditionally and develop knights before bishops.
 White to play
5 Bb5
e6

5...Qa5+ is an interesting alternative, but this fits well into our repertoire.

6 Ne5 Bd7

As mentioned in Model Game 19, an oddball line starting with 2 Nf3 (and
used by some of the world’s finest players, albeit largely in blitz and rapid
games) can lead to this position. White’s idea is to gain the bishop-pair,
but as you will see, that’s not the end of the world.

7 Nxd7 Qxd7 8 c3

Sometimes this is played one move later.

8...Bd6 9 0-0 Nf6 (D)

 White to play
10 Nd2

It’s notable that 10 Re1 0-0 11 Bd3 was played in Carlsen-Vidit, Doha
rapid 2016, but 11...e5! is a good response.

10...0-0

White has the bishops while Black has a slight lead in development. If
White could slowly develop, he could hope for a long-term advantage, but
it’s easier said than done:

a) 11 Nf3 a6 12 Bd3 Ne4 and Black intends to follow up with ...f5 with
a good game.

b) White can try to keep e4 protected with 11 Bd3. However, after

11...Rae8 it’s not obvious how White should continue. On the next move,
Black can play ...e5 and if White moves the knight, Black replies ...Ne4
followed by
...f5.

c) 11 Re1 was played in the high-level game Kramnik-Caruana, Wijk aan Zee
2018, where Black replied 11...a6. I suggest instead 11...Rae8, forcing
White to make a decision. For instance, 12 Nf3 Ne4 13 Ng5 f5 looks fine
for Black.

19B: Minor 6th Moves (1

e4 c6 2 d4 d5 3 exd5 cxd5)

4 Bd3 Nc6 5 c3 Qc7 (D)

White to play
6 Na3

This is a logical attempt to punish Black’s last move. More minor options:

a) Once again 6 Nf3 is met by 6...Bg4.

b) 6 Qb3? is a semi-blunder as after 6...Nxd4! 7 Qxd5 e5 Black will gain

time attacking the queen, leaving White much worse.

c) 6 Bg5 should be of no concern to us, as 6...Nf6 7 Bxf6?! (after 7 Nf3

Ne4 8 Bh4 Bf5 9 Na3 a6 10 0-0 e6 11 Nc2, as in H.Mortensen-Svidler,
Copenhagen simul 2010, I suggest 11...Bd6, when there is not a
trace of opening problems in Black’s camp) 7...gxf6 followed by ...e5
is very promising for Black.

d) 6 Nd2 Nf6 7 Ngf3 (the exotic 7 Ndf3 Bg4 8 h3 Bh5 9 Bg5 Ne4 10 Be3
e6, as in Alexikov-Maksimenko, Nikolaev 2001, is fine for Black) 7...Bg4
and we have a split:

d1) 8 Qa4 is best met by 8...Nd7 9 0-0 e6 10 h3 Bh5 11 Re1 Bd6 12

Qc2 Rc8, when Black had no problems whatsoever in Kratochvil-
Stigar, corr. 2012.

d2) The slow 8 Nf1 was tried in Aronian-Anand, Paris blitz 2018. I think
Black should fight for the initiative with the new move 8...0-0-0; e.g., 9
Ne3 h5 10 h3 Bxf3 11 Qxf3 e5 and I believe Black has every reason to be
happy.

6...a6

There is no need to allow a knight jump to b5.

7 Nc2 Nf6 (D)

 White to play

Here White has a choice.

8 h3

Or:

a) 8 Ne2 plans Bf4, which should be prevented in a way typical for this
variation: 8...Bg4 9 f3 (9 Bf4 Qxf4 10 Nxf4 Bxd1 11 Rxd1 e6 leaves White
with no trace of an opening advantage) 9...Bd7 10 Bf4 e5 11 dxe5 Nxe5 12
0- 0 Bd6 13 Kh1 0-0 14 Ne3 Rfe8 = Zinchenko-Hawkins, London rapid
2016.

b) 8 Bg5 Bg4 9 Qd2 (9 Ne2 is best met by 9...Bh5 intending ...Bg6) 9...Bh5
10 Ne3 e6 11 Ne2 Bd6 12 h3 Bg6 13 Bxf6 gxf6 14 0-0-0 0-0-0 was better
for Black in Papasimakopoulos-T.Kasparova, Vrachati 2011.

8...e6

When playing h3, White ruled out the bishop landing on g4. However,
White’s slow development lets Black develop successfully in another
way.

9 Nf3 Bd6 10 0-0

In a blitz game, Dreev chose the immediate 10...b5 here, but I prefer a
slower approach: 10...0-0 11 Re1 Ne7 with the knight heading for g6 and
possibly f4. A possible continuation is 12 Qe2 Ng6 13 a4 b6, intending
...Bb7 and possibly ...Ne4. An alternative approach is ...Nh5 with the idea
of ...Nf4.

19C: 6 h3 g6: Minor 7th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7)

6 h3 g6 7 Qc2

7 Ne2!?, with the idea of playing Bf4, makes sense. After 7...Bf5 (D)
White has the following options:
 White to play

a) 8 Bf4 doesn’t look critical, but is a decent try: 8...Qd7 9 Bb5 a6 10 Ba4
Bg7 11 0-0 Nf6 12 Ng3 was played in Kvisvik-C.Ekeberg, Norwegian
Team Ch 2011/12, and here I suggest 12...b5 13 Bb3 0-0 14 Nxf5 Qxf5
with roughly equal chances.

b) 8 Bxf5 gxf5 and now:

b1) 9 g4 is a typical way to get access to Black’s weak h-pawn. Not

surprisingly, there are downsides to White’s last move as well. I suggest
9...Qd7 10 gxf5 Qxf5 11 Ng3 Qg6 and as White has just as many
weaknesses as Black, we can’t be worse.

b2) 9 Bf4 Qd7 10 Nd2 e6 11 Nf3 Bd6 (now Black makes all the standard
moves in this structure and equalizes easily) 12 0-0 Nf6 13 Ne5 Qc7 14
Nd3 Bxf4 15 Nexf4 0-0-0 16 Re1 Rhg8 17 Qe2 Kb8 18 a4 Rg5 19 Ne5
Rdg8 ½-½ Degraeve-Medvegy, Bundesliga 2008/9.

We now return to 7 Qc2 (D):

 Black to play

This move stops Black’s initial idea, but it comes at a price.

7...f6!

Black plans ...e5, and there is no way to stop it.

8 Ne2

After 8 Bxg6+? hxg6 9 Qxg6+ Kd8 10 Qf7 Nh6 11 Qxd5+ Qd7, three
pawns are not enough for the piece, and Black is close to winning.

8...e5 9 0-0 Nge7 10 dxe5 fxe5

Facing Black’s centre, White is in danger of being worse. After his next
move, things started to go quickly downhill for the world-class player: 11
c4? (better is 11 Qa4 or 11 Bg5) 11...e4 12 Bxe4 dxe4 13 Nbc3 Bg7 14
Nd5 Qd6 15 Bf4 Be5 16 Rad1 Bxf4 17 Ndxf4 Qe5 –+ 18 b4 Nxb4 19 Qb3
Nbc6 20 c5
Qxc5 21 Nc3 Qb4 22 Qc2 0-0 23 Ncd5 Nxd5 24 Nxd5 Qa3 25 Rfe1 Bf5 26
Re3 Qa5 27 Qc4 Rf7 28 Nf6+ Kg7 29 Rd5 Be6 0-1 Aronian-D.Howell,
Caleta 2019.
Lesson 20: Classical Exchange Main Line 6 Ne2

By: TRH

Facts are stubborn things, but statistics are pliable.

Mark Twain

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7 6 Ne2 (D)

Black to play

Now Bf4 is a direct threat. As you will see, in this lesson the play tends to
get sharp.

6...Bg4

As we saw in 19B, this more or less prevents the bishop development

thanks to the pin. When White nowadays plays this line, he is usually
aiming for the following sequence:
7 0-0 Nf6 8 Qe1 e6 9 f3 Bh5 10 Qh4 (D)

Black to play

In this double-edged position, Black has tried several ideas, but can
you guess the move that gives the big boys a plus score with Black
here? (See below!)

Model Game 20

Vaishali – Indjić

Sitges 2018

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7 6 Ne2 Bg4

Black wants to prevent White from playing Bf4.

7 0-0 (D)

Nowadays this is considered the most critical continuation. For

White’s alternatives, including the tense 7 f3 Bd7 8 Bf4 e5, see 20A.
 Black to play
7...Nf6

There is no reason to delay this move any longer. In B.Haldorsen-

T.R.Hansen, Drammen rapid 2020 I was ready to test my repertoire in a
rapid game only to find myself in a bad position straight out of the opening.
That’s because in this position I mixed up the moves and played 7...e6 8
Qe1 Bd6.
The game went 9 f3 Bh5?! 10 Qh4 Nf6 11 Bg5 and Black had serious
problems. I was lucky to turn the tables but my ‘slip’ indicates that
accuracy is needed in this line.

8 Qe1

This is White’s main idea. The queen is aiming at Black’s king and ready
to pop out to h4 after a subsequent f3. For White’s 8th-move alternatives,
see 20B.

8...e6 (D)
 White to play

9 f3

Or:

a) Another advantage of placing the queen on e1 is that White can now

safely play 9 Bf4 Bd6 10 Bxd6 Qxd6. This position has occurred in a few
games, such as Ter Sahakyan-A.Donchenko, Riga 2018. Black can
continue with
...0-0 and play ...Bh5-g6 to exchange bishops.

b) 9 Ng3 is possible, the idea being to take the h5-square away from
the bishop. 9...Bd6 10 h3 was played in Santos Latasa-Gorshtein,
Stockholm 2018/19. Instead of taking on g3, I recommend 10...Bh5 11
Nxh5 Nxh5, which is approximately equal.

c) 9 Bg5 is a poisonous continuation I have actually prepared for White.

After 9...Bh5 (one of White’s ideas is 9...Be7 10 f3 Bh5 11 Qh4 Bg6 12
Bxg6 fxg6, when he is at least slightly better due to Black’s damaged
structure) 10 f4, Black decided to play the immediate 10...0-0-0 in
Predojević- A.Rasmussen, Bundesliga 2018/19, but I think it’s better to
start by exchanging on e2: after 10...Bxe2 11 Qxe2 Bd6 Black will continue
with
...h6, ...0-0-0 and claim that White’s f-pawn is a weakness. For instance,
12 Qf3 Ne7 13 Nd2 0-0-0 with a sharp and interesting fight ahead.

9...Bh5 10 Qh4

We have reached the diagram position from the introduction to this lesson.

10...Rg8! (D)

White to play

I like this somewhat non-conforming move and believe it assures Black

interesting equality. A main point is to be able to recapture with the h-pawn
if White initiates a bishop exchange on g6.

11 Bg5

Other moves aren’t too troublesome:

a) 11 Nf4 Bg6 12 Nxg6 hxg6 13 Na3 a6 = Alekseev-Arutinian, European

Ch, Batumi 2018.
b) The greedy 11 Bxh7 gives Black good compensation after 11...Rh8.

c) 11 Bf4 Qb6 12 Nd2 Bg6 13 Bxg6 hxg6 14 Qf2 Be7 15 Rfb1 Nh5 16
Be3 Qc7 was fine for Black in Nihal-Shimanov, Moscow blitz 2019.

11...Bg6 12 Nf4 (D)

Black to play
12...Bd6

If the chaos in the game is not to your taste, a more solid approach is
available: 12...Be7 13 Nxg6 hxg6 14 Na3 a6 15 Bf4?! (Tarjan-Dreev,
Internet rapid 2017), when Black can get the advantage by playing 15...e5.
After 16 dxe5 Nxe5, Black is ready to castle queenside and attack White’s
weakened light squares. Because of the fact that White has played f3, a
check on the a7- g1 diagonal might turn out useful for Black.

13 Na3 a6 14 c4? Bxd3?

A strange oversight by both players. 14...Bxa3! is a better move-order as

15 cxd5 Bxb2 ends up in Black’s favour, while Black can also pick off a
pawn
with the simple 14...Nxd4!.

15 Nxd3 Bxa3 16 cxd5 Bxb2 17 Nxb2 Nxd5 18 Qxh7 Kf8 (D)

White to play

Not an everyday isolated queen’s pawn position!

19 Nd3 Nxd4 20 Kh1 Nf5 21 Rac1 Qd6 22 Rfd1?! f6 23 Nf4 Nfe7 24 Qe4?
Kf7 25 Re1

25 Nxd5 Nxd5 26 Bd2 Rgc8 is much better for Black, thanks to his
strong knight on d5.

25...Rh8

The bishop on g5 is going nowhere and White is forced into an

incorrect sacrifice.

26 h3 fxg5 27 Nxe6 Qg3 28 Re2 Rh4 29 f4 gxf4 30 Nc7 Rah8 31 Qe6+ Kf8
0-1
Theory Magnifiers

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7 6 Ne2 Bg4 (7 0-0 Nf6 8 f3)

20A: (7 f3)

20B: (8 f3)

Even though we have come to the main lines of the Classical Exchange,
play isn’t overly theoretical. You will probably do OK if you focus on Model
Game 20 and just skim through the magnifiers until the rest of your Caro-
Kann repertoire is in place.

20A: Minor 7th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7 6 Ne2 Bg4)

7 f3

Or 7 Bf4 Qxf4 8 Nxf4 Bxd1 9 Kxd1 = M.Novikov-Riazantsev, Kolomna

2016.

7...Bd7 (D)
 White to play

8 Bf4

Or:

a) 8 0-0 Nf6 transposes to 20B.

b) 8 Na3 a6 9 Nc2 and now 9...Nf6 transposes to note ‘a’ to White’s

8th move in 19B, while 9...e5!? could certainly be considered.

8...e5 9 Bg3 Bd6 10 Na3 a6 11 Nc2 Nf6

I prefer this over the more common 11...Nge7, as White is now more or
less forced to take the e-pawn, as otherwise Black would have the option
of advancing it.

12 dxe5 Nxe5 13 Ncd4

Or 13 Qd2 0-0 14 0-0-0 Rfe8 =.

13...0-0 14 0-0 Rfe8

Black has slightly the better position, although the desperate 15 Bxe5?
was not what the position needed, and Black quickly won in Liebhart-
M.Hahn, Schwäbisch Gmünd 2015.

20B: 8th-Move Alternatives

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 Bd3 Nc6 5 c3 Qc7 6 Ne2 Bg4)

7 0-0 Nf6 (D)

White to play

8 f3

White has a few other moves:

a) 8 Bf4 Qxf4 9 Nxf4 Bxd1 10 Rxd1 e6 = has occurred in a handful

of games.

b) After 8 Qc2 e6 9 Bf4 Bd6 10 Bxd6 Qxd6 11 Ng3 (O.Mari-Ravi Lanka,

Dubai 2008) I suggest 11...g6 intending to answer 12 h3 with 12...Bf5
=.
c) After 8 h3 Bh5 9 Qb3 e6 10 Bf4 (W.Salamon-W.Schweitzer, Austrian
Team Ch 2015/16) Black should respond 10...Bd6 as 11 Bxd6 Qxd6 12
Qxb7 Rb8 13 Qa6 Rxb2 is more than fine for him.

8...Bd7 9 Bf4

Or:

a) 9 f4 resembles a reversed Stonewall Dutch, and leads to interesting

strategic battles. I like 9...g6, when 10 Nd2 was played in Y.Zhou-
G.Hanssen, Hastings 2010/11. Here I suggest 10...Bf5 11 Bxf5 gxf5 12
Ng3 e6. Black has the half-open g-file, a grip on the e4-square and soon
his king will find safety on the queenside.

b) 9 Re1 is an independent try, but it should be of no concern to Black:

9...e6 10 Bf4 Qb6 (this new move is more combative than the previously
played 10...Bd6) 11 Nd2 Nh5 12 Be3 Bd6 is complicated and Black has no
reason to complain.

We now return to 9 Bf4 (D):

Black to play
Black is well prepared to meet this developing move:

9...e5 10 Bg3 Bd6 11 Bb5 0-0

Black is better; e.g., 12 Bxc6 bxc6 13 Nd2 c5 (13...Rfe8 is natural and

possibly even stronger) 14 dxe5 Bxe5 15 Bxe5 Qxe5 16 Re1 Qe3+ 17
Kh1 Rfe8 =+ Nevednichy-Nisipeanu, Bucharest 2001.

Lesson 21: Panov Attack: Introduction and 5 Nf3 g6

By: SJ

The passed pawn is a criminal, who should be kept under lock and key.
Mild measures, such as police surveillance, are not sufficient.

Aron Nimzowitsch

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 (D)

Black to play
This is the Panov Attack, which is generally considered White’s main
attempt for an advantage in the Exchange Complex. White attacks the
centre and heads for a standard position with an isolated queen’s pawn
where White’s extra central space and freer development typically
compensate (at least) for the potential weakness of his d-pawn.

4...Nf6

This developing move is very natural, and so clearly Black’s main option
that any other move will come as a shock to most Panov players. With that
in mind, in Lesson 24 we shall also take a look at 4...dxc4, which I believe
may be a good surprise move.

5 Nf3 (D)

This normal-looking move is a little too sedate to promise an advantage.

The main move is 5 Nc3, which is the subject of Lessons 22 and 23.

Black to play
5...g6!
?
If possible, this is how you should organize your position against an
isolated queen’s pawn. Black’s king will soon be safe and the dark-
squared bishop will start exerting pressure on the (hopefully) vulnerable d-
pawn.

Q: Isn’t that a rather abstract claim? I can see from the statistics that 5...e6
and 5...Nc6 are more popular moves (both here and in the 5 Nc3 main
lines). Surely the choice between these three moves must come down to
personal preference?

The popularity of Black’s alternatives is probably due to the fact that the
set- up with ...g6 is quite demanding for Black – perhaps not in this exact
position where White has played an early Nf3, but clearly in the lines we
shall see in the next two Lessons. If Black is careless he will lose in the
opening. If he plays well, the reward may only come in the late middlegame
or in the endgame.

Model Game 21

Winants – Bosch

Dutch Team Ch 1997

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nf3

Except for this natural but rather slow developing move and the main move
5 Nc3, White has a few minor options that are discussed in 21A.

5...g6 6 cxd5

For 6 Nc3, see 21B and 21C. The only other move of any interest seems to
be the rare 6 c5, when 6...Bg7 7 Bb5+ Bd7 8 Bxd7+ Qxd7 is probably fairly
equal:

a) 9 0-0 0-0 10 Nc3 Nc6 11 Re1 Qf5 12 a3 (D.Johnston-J.Cooper, Welsh

Ch 1981) 12...Ne4 gives Black an edge.

b) 9 Ne5 Qf5 10 0-0 Nc6 11 Nc3 Nxe5 (11...0-0 is a simpler equalizer)

12 dxe5 Qxe5 13 Qa4+ and White has reasonable compensation for a
pawn.
6...Bg7 (D)

Curiously, the game’s actual move-order was 1 e4 c5 2 Nf3 g6 3 c3 Bg7 4

d4 cxd4 5 cxd4 d5 6 exd5 Nf6.

White to play

Given time, Black will castle and then pick up at least one d-pawn
in complete safety.

7 Bb5+

This is White’s most critical try. 7 Nc3 transposes to note ‘a’ to White’s
7th move in 21B, where after 7...Nxd5 Black can safely meet both 8 Bc4
and 8 Qb3 with 8...Nxc3 followed by 9...0-0.

7...Nbd7 8 d6

After 8 Nc3 0-0 White can head for the same pawn-structure with the
queens on by 9 d6 exd6 10 0-0. However, 10...a6 seems very comfortable
for Black; e.g., 11 Bxd7 (11 Bd3 Re8 =) 11...Bxd7 12 Bg5 Qa5 =+
Reinderman- Feuerstack, Amsterdam 2020.
8...exd6 (D)

White to play

The pawn-structure is basically symmetrical and the isolated d-pawns are

relatively safe as they are placed on a closed file. However, the seemingly
insignificant difference that Black’s pawn is less far advanced actually
offers some interesting possibilities.

9 Qe2+ Qe7 10 Qxe7+

After 10 Bf4 Qxe2+ 11 Bxe2, 11...Ne4!? seems best:

a) 12 Nbd2 Ndf6 13 0-0 0-0 14 Nxe4 Nxe4 15 Rac1 Be6 16 Bd3 Bd5 =
M.Gruber-A.Nussbächer, Austrian Team Ch 2003/4.

b) After 12 Bd3 Ndf6 13 0-0 0-0 14 Re1 Bf5 15 Na3, 15...d5 seems
dead equal. Instead in Leupold-Teichmeister, corr. 2008 Black chose
15...Rfe8, presumably trying to keep some life in the game, but after 16
h3 Rac8 17 Nb5 a6 18 Bxe4 Rxe4 19 Rxe4 Nxe4 20 Nxd6 Nxd6 21
Bxd6 Be6 he was still equal but without any real winning chances.
10...Kxe7 11 0-0 Nb6 (D)

White to play

Black is trying to control d5.

12 Nc3 a6 13 Re1+ Be6

It’s still about the d5-square.

14 Bf1 h6 15 g3 Rac8 16 Bg2 Kd7 17 Re2?! Nfd5

A nice square for a knight, but what about Black’s other knight?

18 Bd2 Nxc3

A well-known idea – exchanging the ‘superfluous’ knight.

19 Bxc3

Also after 19 bxc3 Nd5 Black has somewhat the better long-term
chances thanks to his control of the c4-square.
19...Nd5 (D)

White to play

Black’s strategy has succeeded and his remaining knight is now very strong.

20 Be1 Rc4 21 Rd1 Rhc8 22 h3?!

White can try 22 Ne5+!? dxe5 23 dxe5, but after the aesthetically
pleasing 23...Rc1! 24 Red2 R8c2!, major problems remain for White, as
he has nothing better than 25 Bxd5 Rxd2 26 Bxe6+ Kxe6 27 Rxc1 Rxb2
–/+.

22...Rc1

All Black’s pieces, including his king, are more active than their white
counterparts. Despite the symmetrical pawn-structure he is very much
better.

23 Rxc1 Rxc1 24 Kh2 Nc7

24...g5 is strong too.

25 a3?!
After this point it’s clear that Black is winning, but 25 b3 Nd5 is also very
unpleasant for White.

25...Bc4

26 Rd2 Bd5 27 Re2 Bc4 28 Rd2 Ne6 29 b4

If White tries to prevent Black’s next move with 29 h4, Black’s most
straightforward win is 29...Bd5 30 Re2 Bxf3 31 Bxf3 Nxd4 32 Re3 Nxf3+ 33
Rxf3 Rxe1 34 Rxf7+ Re7.

29...Ng5 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 (5 Nf3 g6 6 Nc3 Bg7 7 Be2)

21A: (5 c5?!)

21B: (7 Qb3)

21C: 7 Be2

The lines discussed in these magnifiers are not particularly dangerous

and have been included mostly for completeness.

21A: Rare 5th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6)

5 c5?!

White hopes to put a bind on the queenside but this isn’t realistic. Other
rare moves:

a) 5 Bg5 Ne4 6 Be3 (6 Bf4? e5 –+) 6...e5 7 Nc3 Qa5 leaves Black at
least equal.
b) After 5 cxd5 Nxd5 (D) White has a choice:

White to play

b1) 6 Nc3 g6 transposes to Lesson

22. b2) 6 Bc4 Nb6 7 Bb3 and then:

b21) After 7...g6, 8 Nc3 transposes to Lesson 22, while 8 Qf3!? takes a
mildly irritating stab at f7; Black can then choose the solid 8...e6 or the
pawn sacrifice 8...Bf5!? – compare the note to Black’s 7th move in 28C.

b22) Perhaps 7...Nc6 is Black’s safest bet. Then 8 Ne2 doesn’t look
frightening, and after 8 d5 Na5, it seems likely that the knight will end
up ideally placed on d6.

5...e5! (D)
 White to play

This thematic counter-stroke more or less refutes White’s over-

optimistic advance.

6 Nc3?!

With 6 dxe5 White acknowledges that his idea has failed. His position
looks ugly, but with some care he may get a playable game: 6...Ne4 7
Nc3! (weakening but essential; 7 Bd3?! Qa5+ 8 Bd2 Qxc5 9 Bxe4 dxe4 10
Bc3 e3 11 fxe3 Qxe3+ 12 Ne2 Nc6 –/+ Mariotti-Pfleger, Olot 1972)
7...Nxc3 8 bxc3 Bxc5 9 Bd3 (9 Nf3 Qb6 10 Qc2 also avoids any serious
disadvantage) 9...Nc6 10 Nf3! (not 10 Bf4? 0-0 11 Nf3 f6! –/+; e.g., 12 exf6
Qxf6 13 Be2
Be6 14 Bd2 d4!) 10...0-0 (10...d4!? 11 0-0 dxc3 12 Rb1 is far from clear)
11 0-0 Bg4 12 Bf4 f6 13 exf6 Qxf6 and the surprising resource 14 Be2!
enabled White to maintain an uneasy equality in Relyea-Perrin, corr. 2017.

6...exd4 7 Qxd4 Nc6

Stockfish likes 7...a6

too. 8 Bb5 Be7 (D)

 White to play

9 Nf3 0-0 10 Bxc6 bxc6

Black is obviously better:

a) 11 0-0 Nd7 –/+.

b) 11 Ne5 Re8 12 0-0 Qc7 13 Re1 and in Apryshko-Zabolotnov, St

Petersburg 2016, Black should have played 13...Nd7! 14 Nxd7 Bxd7 and
now 15 Bf4 Bxc5 16 Qxc5 Qxf4 –/+ or 15 Be3 Bf6 16 Qd2 Qe5 –/+.

21B: 6 Nc3

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6)

5 Nf3 g6 6 Nc3

This position could also arise from the 5 Nc3 g6 6 Nf3 move-order. As we
shall see, it probably isn’t energetic enough to justify White’s acceptance
of an isolated d-pawn.
6...Bg7 (D)

White to play

7 Qb3

Or:

a) 7 cxd5 Nxd5 seems mostly to help Black:

a1) 8 Bc4 Nxc3 9 bxc3 (9 Qb3? Ne4 –+) 9...0-0 10 0-0 Qc7 11 Bb3 (11 Qb3
Nc6 12 Be2 Be6 13 Qa3 Rac8 14 Rb1 Rfd8 =+ Hebden-Anand, London
rapid 2014) 11...Qxc3 12 Bg5 Bg4 13 Bxe7 Bxf3 14 gxf3 (Kalinitschew-
Gisbrecht, 2nd Bundesliga 2001/2) and now Black’s safest path is 14...Rc8
15 Bc5 Nd7 16 Rc1 Qa5 =.

a2) 8 Qb3 Nxc3 9 bxc3 0-0 10 h4 (10 Be2 Nc6 11 0-0 Qc7 is a well-
tested position where Black has a comfortable game) 10...Nc6 11 Ba3?
Be6 12 Qxb7?! (Nakamura-Giri, Leuven rapid 2016) 12...Qa5! 13 Bb4
Nxb4 14 Qxb4 Qf5 –/+.

a3) 8 Be2 0-0 9 0-0 Nc6 10 h3 Be6 11 Ne4?! (11 Re1 Rc8 12 Bf1 Nxc3 13
bxc3 Bd5 = Z.Kovacs-Kustar, Hungarian Team Ch 2003/4) 11...Qb6 12
Nc5 Bf5 –/+ Skomorokhin-Shaposhnikov, Naberezhnye Chelny 2010.

b) The active 7 Bg5 has been tried by some very strong players, but
Black has a number of ways to equalize or complicate according to taste.
I suggest 7...Ne4 (D):

White to play

b1) 8 Bf4 (this and the equally harmless line ‘b2’ can also arise from the
Grünfeld) 8...Nxc3 9 bxc3 0-0 10 cxd5 Qxd5 11 Be2 Nc6 12 0-0 Bf5 13
Qb3 Be4 14 Rac1 Rac8 15 Rfd1 Na5 = F.Bruno-Delchev, Cento 2011.

b2) 8 Bh4 Nxc3 9 bxc3 0-0 10 cxd5 Qxd5 11 Be2 Nc6 12 0-0 e5 13 c4 Qe4
14 Bd3 Qg4 15 h3 Qh5 16 d5 (16 g4? Bxg4 17 hxg4 Qxg4+ 18 Bg3 e4 –
/+) 16...Nd4 17 Be7 Nxf3+ 18 Qxf3 Qxf3 19 gxf3 Re8 =+ Van Wely-
Ragger, French Team Ch 2018.

b3) 8 Be3 Nxc3 9 bxc3 Nc6 10 cxd5 Qxd5 11 Be2 0-0 12 0-0 Na5 (12...Rd8
=) 13 Nd2 Qd8 14 Nb3 Be6 15 Rb1 and rather than 15...Rc8?
(Grachev- Boruchovsky, European Ch, Jerusalem 2015) 16 d5! +/–,
15...Bd5 leaves Black comfortably equal.
b4) 8 cxd5 Nxg5 9 Nxg5 Qb6 10 Bb5+ Bd7 and now:

b41) 11 Bxd7+ Nxd7 12 0-0 0-0 13 Re1 Qxb2 14 Re3 Nb6 15 Rb1 Qa3 16
Rb3 Qd6 =+ Stefansson-Salmensuu, Torshavn 2000.

b42) 11 Qf3 0-0 12 a4 (12 Bxd7 Nxd7 13 0-0 h6 14 Nge4 Qxd4 =+

Rodriguez Vargas-Lopez Martinez, Catalonian Team Ch 2003) 12...Qxd4
13 0-0 h6 14 d6 Qxd6 15 Rad1 Qf6 16 Bxd7 (16 Qxb7? Bc6 17 Bxc6
Qxc6 18 Qxc6 Nxc6 –/+) 16...Nxd7 17 Rxd7 Qxg5 18 Qxb7 Rab8 19 Qxa7
Rxb2 =.

b43) 11 Qe2 0-0 12 0-0 (12 Rd1 Bxd4 13 a4 Bxb5 14 axb5 h6 15 Nge4 a6 =)
12...Bxb5 13 Nxb5 Nd7 14 Nf3 Nf6 15 Nc3 Rad8 16 Qc4 Rd7 17 Rfe1 Rfd8
= Fominykh-Yandemirov, Minsk 1995.

We now return to 7 Qb3 (D):

Black to play
7...0-0 8
cxd5

This position can arise from other move-orders, as we see in Lesson 23.
8...Nbd7

Black intends ...Nb6 in most cases. There have been quite a lot of
games from this position but there is no sign of any theoretical edge
for White:

a) 9 Bf4?! seems rather pointless after 9...Nb6:

a1) 10 Bc4 Bf5 11 Be5 Rc8 12 Be2 Nfxd5 13 Bxg7 Kxg7 14 0-0
(Duneas- Ker, Tauranga 2008) offers Black many good options
including the simple 14...Nxc3 15 bxc3 Be6 with heavy pressure on
White's weak pawns.

a2) 10 d6 exd6 11 Be2 Be6 is better for Black, who controls d5. After 12
Qb4?! Nbd5 13 Qxd6 (Tommiska-Nevanlinna, Finnish Team Ch
2018/19) 13...Qxd6 14 Bxd6 Rfe8 he has a big plus.

b) 9 Bc4 Nb6 10 0-0 Bf5 (10...Nxc4 11 Qxc4 b6 is also fine) and Black
is already comfortable:

b1) 11 Ne5 Rc8 12 Re1 Nxc4 13 Nxc4 Nxd5 = 14 Bg5?

(K.Novikov- A.Zaitsev, Tula 2013) 14...Be6 –+.

b2) 11 Re1 Rc8 12 Bf1 Nfxd5 should offer Black few problems: 13 a4
(13 Nxd5 Nxd5 14 Bg5 {Yanev-Barbosa, Varna 2014} can be met by
14...h6; e.g., 15 Bh4 a5 16 Qxb7 Rb8 17 Qa7 g5 18 Bg3 Rxb2 =+) and
now Black should avoid 13...Nxc3?! 14 bxc3 Be6?! 15 Rxe6, as in
Czerniak-Parma, Polanica Zdroj 1964. Instead 13...a5! looks extremely
solid for Black; note that ...Nb4 is then a threat.

c) 9 d6 exd6 10 Be2 Nb6 11 0-0 Be6 looks nice for Black, who controls d5:

c1) 12 Qd1?! Rc8 13 Bf4 Nfd5 14 Nxd5 Nxd5 15 Bg3 Qb6 16 Qd2 Nb4 17
a3 Nc2 18 Rad1 Bd5 –/+ Dolzhikova-Gandrud, Drammen tt 2019.

c2) 12 Qa3! (Gomez Sanchez-Martinez Duany, Cuban Ch, Villa Clara

2017) 12...Nbd5 13 Ng5 Bf5 14 Qb3 Nxc3 15 bxc3 Rc8 is close to equal,
but Black has whatever edge there is.

d) 9 a4!? (D) may be White’s best try for a slight pull.

 Black to play

After 9...Nb6 10 a5 Nbxd5 11 Bc4 e6 (11...Nxc3 12 bxc3 Ne4 13 0-0 Nd6 is

solid) 12 0-0 (I.Balog-Černoušek, Slovakian Team Ch 2016/17) Black can
keep the game rather unclear with 12...a6 13 Bg5 h6 14 Bh4 g5 15 Bg3 b5
16 axb6 Nxb6 17 Be2 Nh5 18 Be5 Nf4!?, but White seems on firmer
ground.
Black can also play 9...a5 followed by ...Nb6, which looks completely
solid and reliable.

e) 9 Bg5 is White’s most popular try, but 9...Nb6 (D) is fine for Black:
 White to play

e1) 10 Bc4 Bf5 11 Rd1?! Ne4 (11...Rc8!?) 12 0-0 Nxc3 13 bxc3 (Tal-
Bronstein, USSR Ch, Baku 1961) and now 13...Nxc4 14 Qxc4 Be4
leaves Black very comfortable.

e2) 10 Bxf6 Bxf6 11 Be2 e6! 12 dxe6 Bxe6 logically breaks open the game
to Black’s benefit. 13 Qd1 (13 Qc2 Rc8 14 0-0 Nd5 15 Rac1 Re8 is
pleasant for
Black) 13...Nc4 14 b3 and after 14...Na5 15 Rc1 Nc6 16 0-0 Nxd4 17
Nxd4 Bxd4 = Black’s bishop-pair had no significance in S.Drolet-
F.Leveille, Quebec 1991. Perhaps 14...Nd6 15 0-0 Rc8 offers more play.

21C: The Mid-Kent Variation

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nf3 g6 6 Nc3 Bg7)

7 Be2

While not theoretically challenging, this move is an interesting illustration

of what happens when you try to play a theoretical line with reversed
colours and an extra move.
7...0-0 8 0-0 Nc6 (D)

White to play

Giddins tells that this position used to be dubbed the ‘Mid-Kent Variation’
after an area in England where it was explored by some strong amateurs.
It can arise from a number of different openings, including the Grünfeld
and is identical to the Tarrasch main line, except for the reversed colours
and the fact that here it’s White to move. The position has been quite well
tested, and White scores rather miserably. As a matter of fact it seems that
Giddins is right when claiming that White’s extra move is of little value as
Black can use the extra information to adapt to the new situation. If it were
Black to move (as it indeed is in the Tarrasch version), his main
continuations would be ...Bg4 and ...dxc4. Let’s see some attempts to
make use of the extra move:

9 Re1

This seems the overall most useful move as the rook will be well placed
on the half-open e-file and makes 9...Bg4 less tempting.

Other moves:
a) 9 Be3 dxc4 10 Bxc4 Na5 11 Be2 Be6 12 Ng5?! Bc4 13 Nge4?! Nxe4
14 Nxe4 Bxe2 15 Qxe2 Nc6 –/+ Eberlein-Gheorghiu, Biel 1991.

b) 9 c5 goes for another pawn-structure, but after 9...Ne4 10 Be3 Nxc3

11 bxc3 b6 Black is at least equal:

b1) 12 cxb6 axb6 13 Bf4 Ba6 14 Re1 e6 15 Qb3?! Na5 16 Qb4 Bxe2 17
Rxe2 Nc4 18 Nd2 (Rausis-Kanovsky, Czech Team Ch 2014/15) 18...Na3 –
/+.

b2) 12 Qa4 Na5 13 Rfd1 Qc7 14 Rac1 Be6 15 c4 Bd7 16 Qb4 Nxc4 17
Bxc4 dxc4 18 Qxc4 Rac8 = S.Kapnisis-Lupulescu, Greek Team Ch,
Peristeri 2010.

c) 9 h3 prevents ...Bg4 but is largely irrelevant after 9...dxc4 10 Bxc4 Na5:

c1) 11 Bd3 Be6 12 Re1 Rc8 13 Bg5 Bc4 14 Bxc4 Nxc4 15 Qe2 Re8 16 Rac1
Nb6 = M.Petrosian-Kovalenko, Dubai 2015.

c2) 11 Be2 Be6 12 Ne5 (12 Re1 Rc8 13 Bg5 h6 14 Bh4 Nd5 15 Rc1 Nxc3 16
bxc3 Bf6 17 Bxf6 exf6 18 Qa4 Bc4 19 Bxc4 Rxc4 = Hebden-Hodgson,
British Ch, Plymouth 1992) 12...Rc8 13 Bf3 (13 Qa4 a6 14 Rd1 Nd5 15
Bd2 Nc6 16 Bf3? b5 17 Qxa6 Nxd4 18 Nxd5 Bxe5 –/+ Van Elst-Shirazi,
French Team Ch 2007/8) 13...Nd5 14 Ne4 Nb4 15 Nc3 Bf5 =+ Heberla-
Vehi Bach, European Ch, Plovdiv 2008.

9...Bf5 (D)
 White to play

This move is typical for a lot of reversed-colour variations. Black goes for a
relatively harmless continuation which is nevertheless sufficient for
equality.

10 a3

Or:

a) 10 Bf4 dxc4 11 Bxc4 Rc8 12 Rc1 Bg4 13 d5 and in Lammers-P.Nielsen,

Bundesliga 2017/18, Black could have played 13...Nh5! 14 Bg5 Bxf3 15
gxf3 Ne5, with a comfortable game.

b) 10 Bg5 dxc4 11 Bxc4 Rc8 is already pleasant for Black: 12 Be2?

(Tadić- Gliksman, Novi Sad 1972) 12...Qb6! is good for Black; or 12 Rc1
h6 13 Be3 Ng4 =+ Yuneev-Agrest, Miedzybrodzie 1991.

10...Rc8

Black is perfectly fine despite being a tempo down on some minor

Tarrasch variation:
a) 11 h3 Ne4!? 12 cxd5 (12 Nxd5 e6 13 Ne3 Nxd4 14 Nxd4 Bxd4 15 Bf3
h5 16 Qe2 Qh4 =+) 12...Nxc3 13 bxc3 Qxd5 14 Be3 Na5 =.

b) 11 c5 Ne4 and now:

b1) 12 h3 e5 13 Be3 (13 dxe5 Nxc3 14 bxc3 Nxe5 15 Nd4 Rxc5 16

Nxf5 gxf5 17 Rb1 b6 =) 13...Nxc3 14 bxc3 e4 15 Nd2 Be6 =+.

b2) 12 Bf4 Bg4 13 Be3 Nxc5 14 dxc5 Bxf3 15 Bxf3 d4 16 Na4 dxe3
=+ Balashov-Timofeev, Vladivostok 2012.

Lesson 22: Panov Attack Euwe System: 6 cxd5 and Other Moves

By: SJ

When everything’s coming your way, you’re in the wrong lane.

Steven Wright

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3

Applying pressure to the d-pawn.

5...g6

This ambitious move was popularized by Euwe. In the previous lesson

we saw this set-up working perfectly against quiet play by White. In this
lesson we’ll see Black withstand a tougher test.

6 cxd5 (D)
 Black to play
6...Nxd5!

I used to be under the impression that this was inaccurate and that Black
should delay the recapture and castle before trying to regain the pawn.
Now White has two related moves which both force Black’s knight to the
seemingly unattractive b6-square. However, we shall see that the knight
can usually do a very good job from b6, frequently supporting a colleague
on c4.

Model Game 22

Tkachev –

Gunina Caleta

2013

1 c4

The English Opening. What this is doing in a Caro-Kann book will

soon become clear!

1...c6
If you combine the Caro-Kann with the Slav, it takes very little extra effort
to include this move in your ...c6 repertoire.

2 e4

This is the Steiner Variation, which is the subject of Lesson 28, while 2 d4
d5 is the Slav. White has a few more flexible alternatives like 2 Nf3 d5 3 b3
but they are somewhat lacking in sting.

2...d5 3 exd5 cxd5 4 d4

This transposes back to a position we would normally reach via the

move- order 1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4. Instead White can try 4
cxd5 and after 4...Nf6 we shall look at the non-transposing moves 5
Bb5+, 5 Qa4+, 5 Nf3 and 5 Nc3 in Lesson 28.

4...Nf6 5 Nc3 g6 (D)

White to play
6 cxd5
The main alternative is 6 Qb3 (Lesson 23). For 6 Nf3, see 21B. For other
6th moves, including 6 Bg5, see 22A.

6...Nxd5!

One of the slightly annoying lines after 6...Bg7 is 7 Bb5+ (7 Bc4 is the
traditional main line) 7...Nbd7 8 d6 exd6 9 Qe2+ Qe7 10 Bf4 Qxe2+ 11
Bxe2, when compared to the note to White’s 10th move in Model Game
21, Black doesn’t have the easy solution with ...Ne4.

7 Bc4

This active move is White’s most common try. The real alternative is 7 Qb3
(D), which actually scores better.

Black to play

Then 7...Nb6 is recommended (as 7...Nxc3?! 8 Bc4! leads to the line

below with 7 Bc4 Nxc3?! 8 Qb3!), when White has these options:

a) 8 a4 Bg7 9 a5 isn’t particularly dangerous:

a1) 9...Be6 10 Qb5+ Bd7 11 Qb3 Be6 12 Qb5+ Bd7 ½-½ is a reasonable
repetition which concluded the game Orzechowski-Swicarz, Wroclaw
2007.

a2) 9...N6d7 10 Be3 0-0 11 Nf3 Nc6 12 Qa4 Nf6 13 Bb5 a6 14 Bxc6 bxc6 15
Ne5 (L.Vajda-Erenburg, Budapest 2002) and now 15...c5 is perhaps
the clearest path to equality; e.g., 16 dxc5 Nd5 17 Rd1 Nxc3 18 bxc3
Qc7 =.

b) Against 8 Bb5+ we shall examine Smerdon’s suggestion 8...Nc6!?

– exactly the move that White was trying to prevent – in 22B.

c) 8 d5 Bg7 9 Be3 0-0 10 Rd1 Na6 looks fine for Black; e.g., 11 Be2?!
Qd6 12 Nf3 Nc5 =+. For a closer look at 8 d5, consult 22C.

We now return to 7 Bc4 (D):

Black to play
7...Nb6!

This isn’t the ideal square for the knight but it wins a tempo by attacking
the bishop. In contrast, 7...Nxc3?! hopes for a pleasant pawn-structure but
is met by the annoying 8 Qb3!, forcing Black to weaken his dark squares,
one way
or another, with the move ...e6.

8 Bb3 Bg7 9 Nf3

This is very natural but a little compliant, as it allows not only a pin but
also the developing move ...Nc6. For White’s alternatives, including the
rare but quite challenging 9 d5, see 22D.

9...0-0 (D)

The immediate 9...Bg4?? obviously fails to 10 Bxf7+, but playing

9...Nc6 before White prevents it is an alternative. In any case, we shall
see Black getting this knight into play easily enough after the text-
move.

White to play
10 d5

This advance is rare but scores well. White’s main moves are 10 0-0
(22E) and 10 h3 (22F).

Of White’s minor moves, 10 Be3 Nc6 seems fine for Black:

a) 11 d5 Na5 12 Bd4 Nxb3 13 axb3 e5 14 dxe6 Bxe6 15 0-0 Bg4 16
Bxg7 Kxg7 17 h3 Qxd1 18 Rfxd1 Bxf3 19 gxf3 Rfc8 = Za.Varga-
Negulescu, Romanian Team Ch, Calarasi 1995.

b) 11 h3 Na5 12 Bc2 Nac4 13 Bc1 (13 0-0?! Nxb2 14 Qe2 N2c4 15 Bg5 Be6
–/+ Šulskis-Turov, Finnish Team Ch 2017/18) 13...Be6 14 0-0 Rc8 15
Qe2 Nd5 16 Re1 Nxc3 17 bxc3 Qa5 18 Bg5 Bd5 =+ Ådnøy-K.Lie,
Norwegian Ch, Hamar 2007.

10...Bg4

Black is willing to give up the light-squared bishop for a safe blockade

of White’s d-pawn.

11 h3 Bxf3 12 Qxf3 N8d7

From here the knight has a nice view of e5 and c5.

13 0-0 (D)

13 Bg5 Nc5 14 Rd1 Nxb3 15 axb3 Nc8 16 0-0 Nd6 17 Rfe1 Re8 18 g4 Qd7
19 Re3 (Beliavsky-Peresipkin, Minsk 1976) and now Black can equalize
with 19...Rad8 =.
 Black to play

White’s bishop-pair and space advantage are balanced out by Black’s

active knights and pressure against the isolated pawn.

13...Nc5!

Black’s knights are coming to their ideal squares.

14 Bg5

White must play actively to keep the balance:

a) 14 Bc2 Rc8 15 Rd1 Nc4 16 Rb1 Nd6 17 Bf4 Re8 18 Rbc1 Qa5 and
Black may have some long-term pressure.

b) 14 Rd1 Nc8 15 Be3 Nxb3 16 axb3 a6 and now:

b1) 17 Rd2 Nd6 18 Re1 Re8 19 Na4 Qa5 (19...Rc8?! 20 Nb6 += Renet-
Atanasijadis, Vrnjačka Banja 1987) 20 Qd1 Rac8 =+.

b2) 17 Rac1 Nd6 18 Na4 Rc8 19 Rxc8 Qxc8 20 Rc1 Qd7 =+ Dolzhikova-
Kislinsky, Kiev 2004.

b3) 17 Qf4 Nd6 18 Bd4 Bxd4 19 Rxd4 Rc8 20 Re1 Re8 21 Re5 Qc7 =
Poulheim-Leite, corr. 2012.

14...Nxb3

Could this be the moment for 14...h6, putting the question to the bishop?

15 axb3 Qd7 (D)

15...h6 16 Be3 Nc8 17 Rfd1 Nd6 18 Rxa7 Rxa7 19 Bxa7 Qa5 20 Bd4
Bxd4 21 Rxd4 Qa1+ 22 Rd1 Qxb2 = Frolianov-Tomashevsky, Russian
Team Ch, Sochi 2013.

White to play

16 Rad1

Or:

a) 16 Rfe1 Rfe8 (½-½ Gipslis-Gligorić, Zagreb 1965) 17 Rad1 Nc8 18 d6

Nxd6 19 Nd5 f6 20 Bd2 Rad8 21 Bc3 with compensation.

b) 16 Rfd1 is probably the main challenge:

b1) 16...Nc8 17 Qe3 (17 d6 Nxd6 18 Nd5 +=) 17...Re8 18 Bf4 a6 19 Na4
(Nunn-Seirawan, London 1984) 19...Qb5 20 Rac1 Nd6 21 Nb6 Rad8 22
g4! Bxb2 23 Rc7 =.

b2) 16...Rfe8 is little-tested, but after 17 d6 exd6 I can offer the sample
lines 18 Bf4 Be5 19 Ne4 Qf5 20 Bxe5 Qxf3 21 gxf3 Rxe5 22 Nxd6 Re7 =
and 18
Ne4 Bxb2 19 Rxd6 Bxa1 20 Rxd7 Nxd7 21 Nd6 Re1+ 22 Kh2 Be5+ 23
Bf4 Bxf4+ 24 Qxf4 Re7 =.

16...Nc8

Destination: d6.

17 Rfe1 Re8 18 Re2 Nd6 (D)

White to play

From an opening theoretical point of view it’s time to conclude that Black’s
play has been a success. Chances are balanced in a position full of life.

19 g4 Rac8 20 Rde1 Bf8 21 Be3 Nb5 22 Ne4 Bg7 23 Ng5 Rf8

Black’s sounder pawn-structure is compensated by White’s piece activity.

24 Bxa7??

24 Rd2 b6 is roughly equal.

24...Bf6!

This sensible reaction is quite clearly winning. Actually it’s not clear
what White was hoping for, as the obvious 24...Nxa7 25 Rxe7 Qd6 is
probably also winning for Black.

25 Ne4 Bd4!

This was Black’s point.

26 Ng5 e5 27 Ne4 f5 28 gxf5 Bxa7 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 (6 cxd5 Nxd5 7 Bc4 Nb6 8

Bb3 Bg7 9 Nf3 0-0 10 h3)

22A: (6 Bg5)

22B: 7 Qb3 Nb6 (8 Bb5+)

22C: 7 Qb3 Nb6 8 d5

22D: (9 d5)

22E: 10 0-0

22F: 10 h3
These magnifiers are a diverse bunch: 22E and 22F are the main lines
and should be studied in close connection to the model game. 22C is
critical but probably no problem for a well-prepared player as Black (so
make sure to be that). 22A and 22D are harmless and mainly included for
completeness.

22A: Rare 6th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6)

6 Bg5

This doesn’t actually threaten much. Even less dangerous are:

a) 6 Bf4 Bg7 7 Nf3 0-0 8 Be5 Be6 9 cxd5 Nxd5 10 Be2 Nc6 11 Bxg7 Kxg7
=+ Bany-Sulypa, Polish Team Ch, Bydgoszcz 1990.

b) 6 c5 Bg7 7 Bb5+ Nc6 8 Nge2 0-0 9 0-0 and now 9...e5!? seems fine
but very complicated. In Mariotti-Meduna, Biel 1982 Black chose the
simpler 9...a6 10 Bxc6 bxc6 11 b4 Nd7 (11...Bf5!?) 12 a4 e5 =.

6...Bg7 (D)
 White to play

Black offers a temporary pawn sacrifice.

7 Bxf6

This seems consistent, but ineffective. 7 Nf3 transposes to note ‘b’ to

White’s 7th move in 21B.

7...Bxf6 8 Nxd5?!

This is the point, but there is no way White can protect his d-pawn without
his dark-squared bishop. 8 cxd5 can be answered with 8...Qb6, though
Black needn’t even rush to win his pawn back.

8...Bg7 9 Nf3

9 Ne2 Nc6 10 Qd2 Be6 11 Ne3 Nxd4 12 Nxd4 Qxd4 13 Qxd4 Bxd4 –
/+ Arnaudov-Delchev, Bulgarian Ch, Sofia 1996.

9...Nc6

Black will win his pawn back with the better position:

a) 10 h3 e6 11 Nc3 Nxd4 12 Nxd4 Qxd4 –/+ Ramon Pita-

Belakovskaya, Elista Women’s Olympiad 1998.

b) 10 Be2 0-0 11 0-0 Bg4 12 Ne5 Bxe2 13 Qxe2 Nxd4 14 Qe4

(Bregadze- S.Kasparov, Internet blitz 2007) 14...Nf5 –/+.

22B: 7 Qb3 Nb6 8 Bb5+

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6)

6 cxd5 Nxd5 7 Qb3 Nb6 (D)

 White to play
8 Bb5+

After the slow 8 Nf3, Black has 8...Be6!, winning time to control d5; e.g.:

a) 9 Qb5+?! Nc6 10 Ne5 (Pirttimäki-Pyhälä, Järvenpää 1988) and

now 10...Qd6 is a safe path to a small plus.

b) 9 Bb5+ Nc6 doesn’t change much: 10 Qb4 (10 Qd1 Bg7 11 0-0 Rc8 12
Ne5 0-0 13 Bxc6 bxc6 =+ N.Kosintseva-Rambaldi, Dallas 2015) 10...Bg7
11 Bg5 h6 12 Bh4 Nd5 13 Bxc6+ bxc6 14 Qc5 0-0 15 0-0 Qb6 16 Nxd5
Bxd5 17 Bxe7 Rfe8 18 Qa3 Rab8 =+ Korneev-A.Graf, Spanish Team Ch,
Sanxenxo 2004.

8...Nc6!? (D)
 White to play

This is Smerdon’s recommendation, and below I have only added

marginally to his analysis in Smerdon’s Scandinavian.

9 d5

Only this can be critical as 9 Nf3 should again be met by 9...Be6, and 9
Be3 a6 10 Bxc6+ bxc6 11 Nf3 Bg7 12 Rc1 Be6 13 Qc2 0-0 must be at
least equal for Black.

9...a6 10 Ba4

Or:

a) 10 Bxc6+ bxc6 11 dxc6 Be6 12 Qc2 Qc8 =+.

b) 10 dxc6 axb5 11 cxb7 Bxb7 12 Nf3 (12 Qxb5+ Qd7 13 Nf3 Qxb5
14 Nxb5 Bxf3 15 gxf3 Ra5 =+) 12...Bxf3 13 gxf3 Bg7 14 Qxb5+ Nd7
with compensation.

10...Nxa4 11 Qxa4 b5!

This exchange sacrifice is Black’s real idea.

12 Nxb5

Or 12 Qf4? e5 –/+.

12...axb5 13 Qxa8 Nd4! (D)

White to play

Thanks to the fork threat, Black’s activity and White’s lack of

harmony provide full long-term compensation for the exchange.

14 Kf1

White cannot really allow the fork:

a) 14 Rb1? Bf5 –+.

b) 14 Bg5?! Nc2+ 15 Kd2 (15 Ke2?? Bg4+ –+) 15...Nxa1 16 Nf3 Bg7 17
Re1 (17 Rxa1 0-0 –/+) 17...0-0! 18 Bxe7 Qc7 19 Qc6 Qf4+! 20 Re3 (20
Ke2 Bb7!) 20...Bf5! –/+.
c) 14 Nf3 Nc2+ 15 Ke2 Nxa1 16 Bg5 Bg7 17 Rxa1 0-0 =+.

14...Bg7 15 Be3 0-0 16 Bxd4

Or:

a) 16 Rc1? Qd7 –/+.

b) 16 Rd1 Qc7 17 Bxd4 Qc4+ 18 Ne2 Bg4 19 Qxf8+ Kxf8 20 Bxg7+ Kxg7
21 f3 Bf5 =+.

16...Bxd4

White still needs a lot of time to organize his camp. It seems Black has
an advantage:

a) 17 Ne2 Bxb2 18 Rb1 Be5 19 Rxb5 Qc7! 20 Qa5 Qc2! =+.

b) 17 Nf3 Bxb2 18 Rd1 b4 19 Kg1 Qd6 20 Qc6 Rd8 21 Qxd6 Rxd6 =+.

22C: 7 Qb3 Nb6 8 d5

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 cxd5 Nxd5 7 Qb3 Nb6)

8 d5 Bg7 (D)
 White to play

While White’s advanced d-pawn may be a little annoying, taking the c6-
and e6-squares away from Black’s minor pieces, it’s actually less useful
than it was on d4, fighting for c5 and e5. Now a black knight will be ideally
placed on d6.

9 Be3

9 Bb5+ Bd7 10 Nge2 Bxb5 11 Qxb5+ Qd7 12 Qb3 0-0 13 0-0 Rc8 14 Rd1
Na6 15 Be3 Nc4 (15...Nc5 16 Qb4 Na6 17 Qb3 Nc5 is a reasonable
repetition) 16 Bd4 Nd6 17 Bxg7 Kxg7 = Blaskowski-Medvegy, Austrian
Team Ch 2006/7.

9...0-0 10 Rd1 Na6 (D)

 White to play
11 Bxa6

This gives Black a half-open b-file and light-square play but has been
White’s choice in some high-level correspondence games.
Alternatively:

a) 11 Be2?! Qd6 12 Nf3 (12 Bf3? Nc5 13 Qb5 Nca4 14 Nge2 Bd7 was
agreed drawn in Cherniaev-Gipslis, Gausdal 1993 but Black is better; e.g.,
15 Qb3 Nc5 16 Bxc5 Qxc5 17 0-0 Rac8 –/+) 12...Nc5 favours Black:

a1) After 13 Qa3 Nca4 14 Qxd6 exd6 15 Nxa4 Nxa4 16 b3 Nc3 17 Rd2, as
in Upleger-W.Becke, Germany tt 1988, 17...Bd7! is very strong as the
bishop enters play via b5 in many lines.

a2) 13 Qb5 Nca4 14 Ne4 Qc7 15 Rc1 (15 Rd2?, as in Erwich-Spirin,

Leiden 2011, loses to 15...Qc1+ 16 Bd1 Nxb2 17 0-0 Qc4 –+) 15...Qd8 16
b3?! Bf5
17 Nfd2 Nb2 18 Rc5 Bd7 19 Qb4 Bg4 20 f3 a5 –+ Bäckström-Stigar,
corr. 2010.

b) 11 Nf3 Qd6 12 Bxa6 bxa6 13 0-0 Bb7 and now:

b1) 14 Rfe1 Rfd8 15 Bg5 Rd7 16 h3 (S.Chekhov-A.Evdokimov, Moscow
2012) and here Black can play 16...a5 as 17 Ne4 is met by 17...Bxd5 18
Nxd6 Bxb3 19 axb3 Rxd6 =+.

b2) 14 Ng5 (S.Chekhov-Linchevsky, Voronezh 2013) 14...a5!? with the

idea 15 Nge4 Qb4 seems roughly equal; e.g., 16 Bc5 Qxb3 17 axb3 Rfe8
=.

11...bxa6 12 Nge2 a5

12...Bb7 13 0-0 Qc7 14 Rc1 Qc4 15 Qxc4 Nxc4 16 Bc5 Rfe8 17 b3 Nb6 18
Rfd1 Rad8 =+ M.Zaitsev-Balduan, Germany tt 2000/1.

13 0-0 Ba6 14 Bd4 Qd6 15 Bxg7 Kxg7 16 Rd4 Rac8 17 Qd1 Rfd8 (D)

White to play

Black appears to have the upper hand but judging from the
correspondence games of Dolgov, it’s well within the drawing zone after
18 Re1 Bb7:

a) 19 Qd2 Nxd5 20 Nxd5 Bxd5 21 Nc3 e6 22 Rd1 Qe5 23 h3 Rd7 24

Nxd5 Rxd5 25 Rxd5 ½-½ Miakutin-Dolgov, corr. 2005.
b) 19 h4 Bxd5 20 Nf4 e6 21 h5 Qc5 22 Nfxd5 Nxd5 23 Qd2 Rd7 24 Ne4
Qb6 25 Rd3 Qb4 26 Nc3 Rcd8 27 Re4 Qb8 28 Red4 Qe5 29 h6+ Kf8 30
Ne4 (30 a4 a6 31 g3 Qf5 32 Nd1 Ke8 33 Qxa5 Nf6 34 Qc3 Rxd4 ½-½
Strautinš- Dolgov, corr. 2004) 30...Qc7 31 Qg5 e5 32 Ra4 Qc2 33 Qxe5
Qxd3 34 Qh8+
½-½ Strautinš-Dolgov, corr. 2005.

22D: Rare 9th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 cxd5 Nxd5)

7 Bc4 Nb6 8 Bb3 Bg7 9 d5

White has also tried:

a) 9 Nge2 0-0 10 0-0 (after 10 Be3 Nc6, 11 0-0 comes to the same
thing, while 11 d5 is similar to line ‘b’) 10...Nc6 11 Be3 Na5 12 Bc2
Nac4 = Witkowski-P.Thompson, corr. 1997 (and numerous other
games).

b) 9 Be3 0-0 10 h4 Nc6 11 Nge2 h5 12 d5 Na5 13 Bd4 Nxb3 14 axb3 e5

=+ J.Polgar-Grishchuk, Almaty blitz 2008.

9...0-0 (D)
 White to play
10
Nge2

10 Be3 Na6 11 Nge2 comes to the same thing, while 10 Nf3 transposes
to Model Game 22.

10...Na6 11 Be3

11 0-0 Nc5 is comfortable for Black.

11...Bg4 12 f3

After 12 0-0 Black must be a little careful:

a) Not 12...Nc7?!, as in M.Zaitsev-S.Kasparov, 2nd Bundesliga 2008/9,

since after 13 h3! grabbing the d5-pawn will cost Black the exchange:
13...Bxe2 14 Qxe2 Nbxd5 15 Nxd5 Nxd5 16 Rfd1 e6 17 Bc5 Re8 18 Ba4.

b) 12...Rc8 is better:

b1) 13 Bd4 Bxd4 14 Qxd4 Bxe2 15 Nxe2 Qd6 16 Rad1 Nc5 17 Nc3 Rfd8 =
Schmenger-Lytchak, Pinneberg juniors 1996.

b2) 13 f3 Bf5 14 Ng3 (14 Bd4 Nc5 15 Bxg7 Kxg7 16 Qd4+ Kg8 17 Ng3 e6
=) 14...e6!? 15 dxe6 Bxe6 16 Bxe6 fxe6 gave Black enough play in
V.Khan- Schenning, corr. 2013.

b3) 13 h3 Bxe2 14 Qxe2 Nc5 15 Rfd1 and now 15...Nxb3? 16 axb3 left
Black in deep trouble in M.Zaitsev-A.Kartsev, Dortmund 2011, so I
suggest 15...a6; e.g., 16 Bg5 Re8 17 Bc2 (17 d6 Bf6 =) 17...Qd6 18 Rac1
Bxc3!? 19
bxc3 Nxd5 20 Qf3 e6 21 c4 Qe5 with interesting play.

12...Bd7 (D)

White to play

White’s space advantage is no problem as Black can easily activate all

his pieces. Chances look equal:

a) 13 0-0 Nc8 14 Ne4!? (14 Ng3 = is safer) and now 14...Bxb2?! (Rechel-
Bosch, Groningen 1997) 15 Nd4! offers White very dangerous play,
though objectively Black should probably be OK. Instead 14...Nd6 leads to
a calmer but interesting game; e.g., 15 N2c3 Rc8 16 Qd2 f5 17 Ng5 Nc5
18 Rad1 (not
18 Bc2?! Nc4 –/+) 18...Nxb3 19 axb3 Qa5 =.

b) 13 Bd4 Bxd4 (13...Nc8?! 14 h4!? could be tricky) 14 Qxd4 Nc8 15 0-0

(now Black can meet 15 h4 with 15...Qb6) 15...Qb6 16 Ne4 Bb5 17 Rfe1
Nb4 (17...Rd8 is probably equal) 18 N4c3 Bxe2 19 Qxb6 axb6 20 Rxe2
Rd8 21 Rd1 += Sermek-Zelčić, Solin/Split semi-rapid 2002.

22E: 10 0-0

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 cxd5 Nxd5 7 Bc4

Nb6 8 Bb3 Bg7)

9 Nf3 0-0 10 0-0 (D)

Black to play

White develops in straightforward fashion, not fearing a pin of his knight.

10...Nc6

The equally natural 10...Bg4, unsurprisingly, is equally playable.

11 d5

This must be critical. After the normal developing move 11 Be3, Black
is already for choice following 11...Na5 12 Bc2 Nac4 =+.

11...Na5 (D)

White to play

This is yet another example of how Black’s knights create

queenside counterplay.

12 Re1

None of White’s alternatives have given Black any problems:

a) 12 h3 Re8 13 Re1 Bf5 14 g4 (14 Bg5 Bxc3 15 bxc3 Nxb3 16 axb3 Qxd5
17 Qxd5 Nxd5 = Mons-Ragger, Bundesliga 2015/16) 14...Bd7 15 Bg5
Nxb3 16 axb3 h6 17 Bh4 g5 18 Bg3 e6 19 d6 Bc6 = Navara-Wei Yi,
Yancheng (4) 2018.

b) 12 Bg5 h6 13 Bh4 (13 Be3 Bg4 14 h3 Bxf3 15 Qxf3 Nbc4 16 Rac1?! Rc8
17 Bxc4 Nxc4 18 b3?! Nxe3 19 Qxe3 Qa5 –/+ Molner-Gareyev, Los
Angeles 2012) 13...g5 14 Bg3 Bg4 15 Re1 Rc8 16 Qd3 e6 17 d6 Nac4 =
T.Taylor-
Antal, Budapest 2003.

12...Re8!

This is flexible, and ...e6 and ...e5 are important resources.

13 Ne5

Alternatively:

a) 13 Bg5 Bxc3 14 bxc3 Nxb3 15 axb3 Qxd5 16 Rxe7 Rxe7 17 Bxe7 Bg4
= Hamdouchi-Edouard, Spanish Team Ch, Leon 2012.

b) 13 Be3 Bg4 14 h3 Bxf3 15 Qxf3 Bxc3 16 bxc3 Nxb3 17 Bxb6 Nd2

18 Qd3 Qxb6 19 Qxd2 = Vachier-Lagrave – Nepomniashchy, Wijk aan
Zee 2011.

13...Qd6 14 Nb5

14 Bf4 Bf5 15 Nd3 is a shorter route to the same position.

14...Qc5 15 Nc3 Qd6 16 Bf4 (D)

 Black to play

Black has equalized but if he wants to avoid a draw by repetition there

is some fancy footwork needed.

16...Bf5 17 Nd3 Qd8

17...Qd7 18 Ne5 Qd6 is a repetition.

18 Ne5 Nxb3 19 axb3 Qd6

Now:

a) 20 g4?! probably isn’t quite sound: 20...Qb4! 21 Bg5 Bxe5 22 gxf5 f6

23 fxg6 hxg6 24 Qf3 Qd6 =+.

b) 20 Nd3 e5 is equal but not drawn.

c) 20 Qd2 Qb4 21 g4 Bc2 22 Qxc2 Qxf4 =+ Croberson-Ciron, Internet

rapid (computer assisted) 2007.

22F: 10 h3
(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 cxd5 Nxd5 7 Bc4 Nb6 8
Bb3 Bg7 9 Nf3 0-0)

10 h3

This is White’s most popular move. Now it may be difficult for Black to find
a good square for his light-squared bishop.

10...Nc6 (D)

White to play
11 Be3

11 d5 seems illogical in combination with h3. More interesting is 11 0-

0 Nxd4 12 Nxd4 Qxd4 13 Bg5, though White probably only wins back
the pawn with equality.

11...Na5

This seemingly decentralizing move is an essential part of Black’s

strategy. The knight fights for control over c4 from the edge.
12 Bc2

White has also tried 12 0-0 Nxb3 13 Qxb3 (13 axb3 Nd5 14 Bg5 Be6 15
Re1 Re8 16 Ne4 b6 17 Qd2 Qd7 =+ Hjartarson-P.Roberson, Internet rapid
2017) 13...Be6 14 Qd1 (14 Qc2 Rc8 15 Rfd1 Nc4 16 d5?? Bf5 17 Qb3
Nxb2 –+
Tokranovs-Stupak, Katowice blitz 2017) 14...Rc8 15 Re1 Nd5 16 Bd2
Nxc3 17 bxc3 Bd5 18 Ne5 Re8 19 Qa4 a6 =+ Roy Chowdhury-Astaneh
Lopez, London 2010.

12...Nac4 (D)

White to play

13 Bc1

Alternatively:

a) After 13 Rb1?!, 13...e5 gave Black a nice plus in M.Biriukov-

Quiñonez Garcia, Internet blitz 2020, but actually 13...Be6! 14 Qe2 Rc8
–/+ is even stronger.

b) 13 Qe2 Be6 14 Bd3 (14 Bc1 Bxd4 15 Nxd4 Qxd4 16 0-0 Rfd8 =+)
14...Rc8 15 0-0 Nxe3 16 fxe3 Nd5 17 Nxd5 Bxd5 18 e4 Bc6 19 d5 Bd7 =+.

13...Nd5 14 0-0 Nxc3 15 bxc3 Be6

Black’s light-square control at least balances out White’s extra

space: a) 16 Rb1 Bd5 17 Re1 Rc8 18 Bg5 f6 19 Bc1 e6 20 a4 Qd6

21 Nd2 =
Kekelidze-Kacheishvili, Mesa 2010.

b) 16 Re1 Bd5 17 Bg5 Re8 18 Bb3 Rc8 19 Ne5 (19 Qd3?! Qc7 20 Re2?! e6
–/+ Wohl-Zelčić, Schwarzach 2011) 19...Qa5 20 Qd3 h6 21 Nxc4 Bxc4
22 Bxc4 Qxg5 =.

Lesson 23: Panov Attack Euwe System: 6 Qb3

By: SJ

The essential disadvantage of the isolated pawn ... lies not in the pawn
itself, but in the square in front of the pawn.

Richard Réti

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 (D)

 Black to play

White exerts pressure on the centre, hoping to provoke a capture on c4.

6...Bg7

This obviously costs a pawn, but if Black wanted to play a line involving
...e6, he should have done so without including ...g6.

7 cxd5

The proof of the pudding lies in the eating.

7...0-0

This is where White decides the nature of the struggle:

a) 8 Nf3 transposes to 21B.

b) 8 g3 is the subject of 23A.

c) 8 Nge2?!, intending Nf4, seems very logical but is actually very risky for
White. We look at it in Model Game 23.

d) 8 Be2 is the undisputed main move. After 8...a5! (D), White has
the following options:

White to play
d1) 9 a4?! Na6 and
now:

d11) 10 Bf3?! Nb4 11 Be3? (11 Nge2 Nd3+ is just =+) 11...Bf5 and in
Faure- Skembris, Geneva 1989 Black was already winning. The game was
over after 12 Kd2 Ra6! 13 Nge2 Rb6 14 Rag1 Nbxd5 15 Nb5 e5 –+ 16 g4
Be6 17 Qd1
e4 18 Bg2 Nxg4 19 h3 Ndxe3 0-1.

d12) 10 Nf3 Nb4 11 0-0 Nfxd5 (Smerdon recommends 11...Bf5, which

also looks fine) 12 Nxd5 Qxd5 13 Bc4 (Kjetzae-Stelting, Hamburg Ch
1987) 13...Qd6 =+.

d2) 9 Bf3 seems consistent and is examined in

23B. d3) 9 Nf3! is probably best, and the subject

of 23C.
 Model Game 23

Frolianov –

Siugirov Tiumen

2012

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 Bg7 7 cxd5 0-

0 8 Nge2?! (D)

Black to play

In order to keep his pawn, White has played some slightly artificial
moves, and can quickly get into trouble.

8...a5!

Surprisingly, this seems to be the way to exploit White’s passive play.

9 Nf4

White tries to get his camp in order by freeing his light-squared

bishop. Otherwise:
a) 9 a4 Na6 and Smerdon seems right in assessing that Black is
already better.

b) 9 g3 a4! 10 Nxa4 Qxd5 11 Qxd5 Nxd5 12 Bg2 Rxa4 13 Bxd5 Nc6!

leaves Black at least equal.

9...a4!?

This advance is the key idea. However, 9...g5! (D) is a more accurate
move- order (and therefore my recommendation), forcing the knight to
choose a retreat-square immediately. Then:

White to play

a) After 10 Nd3 a4!, White has nothing better than transposing to the
game with 11 Nxa4.

b) In the case of 10 Nfe2 a4, 11 Qc4 is superior to taking the pawn, but
Black can be happy after 11...Ra5!; e.g., 12 Be3 (12 Bxg5? Nxd5 13 Nxd5
Rxd5 14 h4 Nc6 is desperately unpleasant for White, M.Horvath-Van de
Ploeg, corr. 2013) 12...Nxd5 (12...Na6!?) 13 Nxd5 Be6 and Black regains
the piece with some advantage.
10 Nxa4?!

Black’s move-order allows White the extra possibility of 10 Qd1,

enabling him to meet 10...g5?! with 11 Nh5!. Black should be equal after
10...Nbd7, but it is more promising to sidestep this by playing 9...g5! first.

10...g5

How often do you see Black execute a plan involving the pawn moves
...a4 and ...g5?

11 Nd3 (D)

11 Ne2 may be somewhat better but after 11...Nxd5 12 Nac3 Be6 13

Bxg5 Nc6 White isn’t really any closer to getting his king to safety.

Black to play
11...Nxd5

Black already has an edge and, perhaps more importantly, White must
play very accurately just to survive.
12 Nc3

After the greedy 12 Bxg5? Nc6 13 Nc3 Be6, White is too far behind in
development; e.g., 14 Qxb7 (14 Qd1 Nxd4 is even worse) 14...Nxd4 15
Rc1 Rb8 16 Qa6 Bf5 and White won’t get his king safely out of the centre.
A typical line is 17 Qc4 Nxc3 18 bxc3 Rc8 19 Qb4 Ne6 20 Bxe7 Qd5 21
Qd6 Qe4+ 22 Be2 Rfe8 and Black is winning easily.

12...e6

Black is spoilt for choice here, with 12...Ra5! probably even better.

13 Ne5 Nc6 14 Nxc6 bxc6 15 Bc4

15 Be2 isn’t sufficient for equality: 15...Bxd4 16 0-0 Bg7 (16...Qf6 17 a4 h6

=+ may be more accurate) 17 Rd1 h6 18 Ne4 Ba6 19 Bxa6 Rxa6 20 a3 Qb8
=+ Šimaček-A.Kim, Pardubice 2019.

15...Bxd4 16 0-0 (D)

Black to play
White has survived the direct attack.

16...Bg7 17 Bd2 h6 18 Rad1 Ba6 19 Bxa6 Rxa6

Black is only marginally better.

20 Qc4 Qa5 21 b4 Qb6 22 Rb1?!

After 22 h4, nothing is clear.

22...Ra3

Black’s queenside pressure gives him a clear advantage.

23 Rb3 Rfa8 24 Rxa3 Rxa3 25 Nb1 Ra6 26 a3 Qb5 27 Qe4 c5

Now White will have the only pawn weakness.

28 bxc5 Ra4 29 Qc2 Rc4 30 Qd3 Qxc5

Black has the sounder pawn-structure as well as the more active pieces.

31 h3 Rc2 32 Qe4 Bf6

Even better is 32...Bd4.

33 Qf3 Rb2 34 Be3?

Black should also win after the more stubborn 34 Bb4 Qc2; e.g., 35
Qd1 Nxb4 36 axb4 Qf5 –+.

34...Qc2

Now Black simply picks up a piece.

35 Nd2 Nxe3 36 Ne4 Nxf1 37 Nxf6+ Kf8 38 Nd7+ Ke7 39 Ne5 Qf5 40
Nc6+ Kd6 0-1

Theory Magnifiers
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 Bg7 7 cxd5 0-0 (8 Be2
a5! 9 Nf3!)

23A: 8 g3

23B: 9 Bf3

23C: 9 Nf3!

These magnifiers are all important for your understanding of the Euwe
Variation. If you are a casual player or have limited time for study, you
may limit yourself to playing through the moves in bold, but don’t skip
them entirely.

23A: 8 g3

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 Bg7 7 cxd5 0-0)

8 g3

This would be fine for White if allowed a little peace to

complete development.

8...e6! (D)
 White to play

Black offers a couple of pawns in order to get at White’s uncastled king.

9 dxe6

After 9 Bg2 Nxd5 White has nothing to show for his pawn weaknesses.
The game Delemarre-Armas, Wijk aan Zee 1995 was drawn after 10 Nge2
Nc6 11 0-0 Nxc3 12 bxc3 Na5 13 Qd1 Bd7 14 Rb1, but 14...Rc8 (and a
few other moves) would have given Black an obvious edge.

9...Bxe6 10 Qxb7 Nbd7 11 Bg2 Rb8 12 Qxa7 Re8 13 Nge2 Bc4! (D)
 White to play

Black has full compensation for his three pawns thanks to his piece
activity and the white king stranded in the centre:

a) Stockfish recommends 14 Qa4 with the idea of giving an exchange back

by 14...Nb6 15 Qc2 Nfd5 16 0-0. After 16...Nxc3 17 Nxc3 Bxf1 18 Bxf1 Rc8
19 h4 Bxd4 20 Bg5 Qd6 21 Qd3 Nc4 chances are equal.

b) 14 b3 Nd5 and then:

b1) 15 Be3?! Nxe3 16 fxe3 Bxe2 17 Kxe2 (W.Stern-Lepsenyi, European

corr. Ch 1972) 17...Nf6 18 Rhe1 Ng4 19 Kd2 Rc8 20 Bh3 Ra8 21 Qxa8
Qxa8 22 Bxg4 Qc6 =+.

b2) 15 0-0 Nxc3 16 Nxc3 Bxf1 17 Kxf1 Rc8 (= Smerdon) 18 Nd5 Rc2
19 Be3 Nf6 20 Nxf6+ Qxf6 21 Kg1 Bh6 22 Rf1 Bxe3 23 fxe3 Qe6 24 e4
= Blake-Kappe, corr. 2010.

c) 14 Bf3 Nd5 15 0-0 (15 Qa4? N7b6 16 Qc2 Qf6 17 Bxd5 Nxd5 18 Nxd5
Bxd5 19 Rf1 Qxd4 –+ Õim-Shereshevsky, USSR Ch, Kharkov 1967)
15...Nxc3 16 Nxc3 Bxf1 17 Kxf1 Nf8 18 Nd5 Ne6 19 Be3 Rxb2 was hard
to
evaluate in H.Hagesæther-Evseev, Norwegian Team Ch 1998.

23B: 8 Be2 a5! 9 Bf3

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 Bg7 7 cxd5 0-0)

8 Be2 a5! 9 Bf3 Na6 (D)

White to play

This line is the recommendation of David Smerdon in his book on the

Scandinavian. A lot of what follows is his analysis but it’s not entirely
clear to me how much of it is based on 2010 analysis by Gawain Jones.

10 a3

Or 10 Nge2 Nb4 and now:

a) 11 a3?! Nd3+ 12 Kf1 Nxc1 13 Rxc1 Bg4?! (13...Ra6!) 14 Bxg4 Nxg4 15

Ne4 Ra6 16 h3 Nf6 =+ Licznerski-Dors, Polish Team Ch, Poronin 2013.

b) 11 0-0 Bf5 and now:

b1) 12 Bf4 Bc2 13 Qa3 Bd3 14 Rac1 Re8 15 Rfd1 Bf8?! (15...Bc4 =) 16 Qa4
+= Santa Maria-L.B.Hansen, World Junior Ch, Gausdal 1986.

b2) 12 Bg5 Bc2 13 Qa3 h6 14 Be3 Rc8 15 Rac1 b6 16 h3 Qd7 (after

16...g5?! 17 Rxc2! Nxc2 18 Qb3 Nb4?!, as in Bellin-Bronstein,
Hastings 1975/6, 19 Ng3 gives White more than enough for the
exchange) 17 Bf4 Rfd8 18 Bg3 Bd3 =.

10...b5! (D)

White to play

11 Qxb5

Or:

a) 11 Nge2 b4 12 Ne4 Nxe4 13 Bxe4 Bb7 14 Bf3 Qd7 15 0-0 a4 16 Qe3

Bxd5 =+ Monnisha-Purtseladze, Chennai women 2013.

b) After 11 Nxb5, 11...a4!? may be best:

b1) 12 Qc4 e5! 13 dxe5 Nd7 14 Bf4 Qa5+ 15 Bd2 Nxe5 16 Bxa5 Nxc4 =+.
b2) After 12 Qxa4 (Balakrishnan-Ronka, Internet blitz 2020) Black should
go for the natural 12...Bd7 13 Qb3 Rb8 14 a4 Nc7; e.g., 15 Be2 Ne4 16 Nf3
Nd6 17 Bf4 Bxb5 18 axb5 Ncxb5 19 Qd1 Qb6 20 Be5 Nc7 =.

c) 11 d6 Rb8 12 dxe7 Qxe7+ 13 Nge2 Be6 14 Qd1 Rfe8 (14...Nc7 15 0-0 b4

16 axb4 axb4 17 Ne4 Nxe4 18 Bxe4 Rbd8 = Smerdon) and now:

c1) 15 Bg5 h6 16 Bh4 (16 Bf4 Rbd8 17 0-0 b4 18 axb4 axb4 19 Rxa6 bxc3
20 Nxc3 Bc4 21 Ra4 Bxf1 22 Kxf1 Qd7 =+) 16...Bc4 17 b3 Bxe2 18
Nxe2 b4 =.

c2) 15 0-0 b4 16 axb4 axb4 17 Ne4 Nxe4 18 Rxa6 (18 Bxe4?! Bc4 –/+)
18...Bc4 19 Ra4 Qe6! 20 Ra1 Rbd8 21 Be3 Nxf2! 22 Bxf2 Bxe2 23
Bxe2 Qxe2 =+.

11...Nc7 (D)

White to play
12 Qb3

12 Qc6 Bd7 13 Qc4 Rc8 and now:

a) 14 Qe2?! Bf5 (14...Be8 =+ Smerdon) 15 g4 Bd7 16 Bg5 e6 17 d6 Ncd5
18 Qd2 Nxc3 19 bxc3 Bb5 20 Rb1 Ba6 21 Ne2 Qxd6 –/+.

b) 14 Qa2 e6! 15 Bg5 h6 16 Bxf6 Qxf6 17 Nge2 Nxd5 18 Nxd5 exd5 19 0-

0 Bb5 =.

12...Ba6! (D)

Much better than 12...a4?! 13 Nxa4 Ncxd5?! 14 Ne2 Qa5+ 15 Nac3

+/– Naiditsch-B.Vučković, European Ch, Rijeka 2010.

White to play

With this move we leave GM practice. Some years ago that would have
meant that the continuation from this point is unreliable. That isn’t the
case any more. The following lines are far more accurate than anything
humans could produce in an over-the-board encounter. Still they say little
about how a real game between strong players could go. My guess is that
White would have a far tougher task defending with his two extra pawns
than Black keeping up his initiative.

13 Nge2 Rb8 14 Qa2 Qd7 15 0-0

Or 15 d6 exd6 16 0-0 a4 17 Re1 Rb3 18 Qb1 Bc4 with compensation.

15...a4 16 Bf4

Stockfish considers the position after 16 Bd2!? Rb3 17 Nc1! Bxf1 18

Kxf1 Rbb8 19 Nd3 equal. That’s no doubt correct but for most humans
‘unclear’ describes the position better.

16...Rb3! 17 Bxc7 Qxc7 (D)

White to play

18 Rfe1!

Not:

a) 18 Nxa4?? Rxf3! –+.

b) 18 Rac1?! Rfb8 19 Nxa4 Qa5 20 Nac3? Bh6 –/+ Smerdon.

18...Rfb8 19 Nxa4 Qa5 20 Nec3 Bc4

With computer-like care White may be able to hold

this: a) 21 Be2?! Rxc3! 22 Bxc4 Rc2 23 d6 e6 –/+.

b) 21 Qb1! Nxd5 22 Nxd5 Bxd5 23 Bxd5 Qxd5 24 Qe4! (24 Rxe7?! Qc4!
=+) 24...Qxd4 25 Rad1 Qxe4 26 Rxe4 e6 =.

23C: 8 Be2 a5! 9 Nf3!

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3 g6 6 Qb3 Bg7 7 cxd5 0-0 8

Be2 a5!)

9 Nf3!

This may seem inconsistent but having seen the activity Black may
generate after the slower 9 Bf3, it’s clear why White should prepare
immediate castling.

9...a4! (D)

White to play
10 Nxa4 Nxd5 11 Nc3

Or 11 Bc4 b5!! 12 Bxb5 Nc7 13 Be2 Nc6 with compensation, as pointed

out by Smerdon.

11...Nb6 12 0-0

Smerdon gave 12 d5 e6! with full compensation:

a) After 13 dxe6 Bxe6 14 Qc2 Nc6 15 0-0 Nb4 16 Qe4 N4d5 17 Nxd5
Bxd5 Black may pick up the a-pawn with equality, or play for more. The
repetition 18 Qb4 Ra4 19 Qb5 Bc6 20 Qb3 Bd5 21 Qb5 Bc6 is worth
noting.

b) 13 Bg5 Qd6 14 0-0 Nxd5 15 Rfd1 Qc7! seems fairly balanced: 16

Rxd5!? (16 Nxd5 exd5 17 Qxd5 Bxb2 =) 16...exd5 17 Nxd5 Qa5 18 Ne7+
Kh8 19 Nxc8 Rxc8 20 Qxf7 Qc7 and chances are level.

12...Be6 13 Qd1

Smerdon gave 13 Qc2 Nc6 with reasonable compensation for the pawn.
That assessment isn’t challenged by the Stockfish line 14 Bg5 Nxd4 15
Nxd4 Qxd4 16 Bxe7 Rfc8 17 Rfd1 Qe5 18 Bd6 Qa5 =.

13...Nc6 14 Be3 (D)

 Black to play
14...Nd5

Black’s piece activity provides fairly stable compensation for the pawn so
he doesn’t have to remember or find ‘only’ moves. Also 14...Qd7 15 Ne4
Bd5 16 Nc3 Bxf3 17 Bxf3 Nxd4 18 Bxd4 Bxd4 19 Nd5 Nxd5 20 Qxd4 (Li
Ruifeng-Ostrovsky, Linthicum 2016) is just slightly better for White, and
Black may also consider other sensible moves like 14...Nb4.

15 Ng5

Or:

a) 15 h3 (15 Kh1 Qb6 16 Qd2 Rfd8 17 h3 comes to the same thing)

15...Qb6 16 Qd2 Rfd8 17 Kh1 Kh8 18 Ng5 Bxd4 19 Nxe6 fxe6 20 Bxd4+
Qxd4 =.

b) 15 Qd2 Qb6 16 Bh6 Bxh6 17 Qxh6 Nf6 18 Qd2 Rfd8 19 Rfd1 +=.

15...Qb6!? (D)

An interesting concept: White will not be able to attack the pawn weakness in
view of Black’s superior activity. 15...Bf5 also seems fully playable.

White to play

16 Nxe6 fxe6 17 Nb5 Kh8 18 Qb3 Na5 19 Qd3 Nc6

Inviting a repetition, but 19...e5 could well be

stronger. 20 a4

White decides to play on, but Black has compensation for a pawn.
Stockfish gives full equality after some complicated lines starting 20...Nxe3
21 fxe3 e5. However, an easier move-order for a human to play is 20...e5
21 dxe5 and only now 21...Nxe3 22 Qxe3 (22 fxe3 Rfd8 =) 22...Qxe3 23
fxe3 Rxf1+ 24 Kxf1 Bxe5 25 Nc3 Bxh2 with practical equality.

Lesson 24: Panov Attack: Miles’s 4...dxc4

By: SJ

He who fears an isolated queen’s pawn should give up chess.

Siegbert Tarrasch

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4!? (D)

White to play

Black immediately inflicts some damage on White’s pawn-structure. Ideally

he would like to do so only after White’s light-squared bishop has moved,
but this is in general hard to achieve. The move nevertheless has its points,
and it’s surprising that it occurs in less than 2% of all Panov games.

5 Bxc4

This position arises much more frequently from the Queen’s

Gambit Accepted move-order 1 d4 d5 2 c4 dxc4 3 e3 c5 4 Bxc4
cxd4 5 exd4.

5...Qc7

The first point of Black’s early exchange: the light-squared bishop

is vulnerable on the c-file.
6 Bb3 Bg4

The subsidiary point: thanks to the exposed bishop on c1, Black sows
further confusion.

Model Game 24

A. Jakab – Varga

Hungarian Team Ch 2015/16

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4 5 Bxc4

In the game this position was reached from the QGA move-order, but
here there are two other important options. In each case Black has no
need to transpose back to standard lines:

a) 5 Nf3 Be6! forces White to waste time if he is to regain the pawn. 6 Nc3
(6 Na3 Bd5 followed by ...e6) 6...Nf6 7 Be2 (7 Bxc4?? is a surprisingly
common blunder, while 7 Ne5 is well met by 7...Nc6!, and 7 Be3!? Nc6 8
Rc1 Nd5 is solid for Black) 7...Nc6 8 0-0 g6 and now 9 Ng5 Nxd4 10 Nxe6
Nxe6 11 Bf3 merely offers White sufficient compensation for the pawns.

b) 5 Nc3 Qc7!? (a rare move, but reliable) 6 Nf3 (after 6 Qa4+ Bd7 7 Qxc4
Qxc4 8 Bxc4 Nf6 9 Nf3 e6 Black has nothing to fear, while 6 g3 e6 7 Bg2
targets the black queen but is artificial, with 7...Nc6 8 Bf4 Bd6 one of
several good replies) 6...Nf6 7 Ne5 e6 8 Bxc4 (Artemiev-B.Savchenko,
Russia Cup, Khanty-Mansiisk 2014) 8...Nc6! nicely neutralizes White’s
activity. After 9 Nb5 Bb4+ (9...Qb8?? 10 Bf4 +–) 10 Bd2 Bxd2+ 11 Qxd2
Qb8 Black is secure, while 9 Nxc6 bxc6 10 0-0 (10 Qe2 Bd6 11 Bg5 can
be met by 11...Nd5, due to 12 Nxd5 cxd5 13 Bxd5?? Qa5+ –+), 10...Bd6
gains time by hitting h2. Then 11 h3 0-0 12 Bg5 (12 Na4 e5) 12...Be7
intending ...Rd8 is balanced.

5...Qc7 6 Bb3

For alternatives, see 24A and 24B.

6...Bg4 7 f3 (D)

This attack on the bishop gains some time but the weakening of
White’s kingside pawn-chain is felt in some lines and his king’s knight
loses an important option.

Other moves are less theoretically critical. Black maybe shouldn’t expect
too much from 7 Ne2?! Bxe2! but White’s position will be difficult to play.
From a practical viewpoint it’s much more important to be aware that 7 Nf3
Bxf3 8 Qxf3!? is a more or less forced draw. See 24C for analysis.

Black to play
7...Bd7

The bishop is quite well placed here as it will frequently continue to

c6, helping to control the important d5-square.

8 Ne2

With the normal square for the knight occupied by a pawn, this seems
rather obvious. Rarer options:
a) 8 Nc3 e6 9 f4?! (9 Nge2 transposes back to the main game) 9...Nf6 10
Nf3 Be7 11 0-0 Nc6 12 Kh1 0-0 13 Ne5 Rfd8 is equal as 14 Nxf7? Kxf7 15
f5 Kg8 16 fxe6 Be8 doesn’t give White enough for the piece, Liogky-Kallai,
French Team Ch 1998.

b) 8 d5 g6 9 Ne2 Bg7 10 Nbc3 Nf6 11 Bf4 (11 Be3 0-0 12 0-0 ½-½
Mikhalchishin-Sermek, Slovenian Team Ch, Ljubljana 1997) 11...Qb6
12 Qd4 Qxd4?! 13 Nxd4 0-0 14 0-0-0 += Mikhalchishin-Sermek,
Slovenian Team Ch, Bled 1999.

8...e6 (D)

White to play
9 Nbc3

White has also tried 9 0-0 Bd6 10 f4 Nf6 11 Nbc3 0-0 12 Ng3?! Bc6
(12...Nc6!?) 13 Be3 Nbd7 14 Rc1 = Sebag-Ushenina, European
Women’s Ch, Kusadasi 2006.

9...Nf6 10 Bg5
Alternatively 10 Bf4 Bd6 11 Qd2, and now:

a) 11...Bc6 12 Rc1 Bxf4 13 Nxf4 0-0 14 0-0 Nbd7 15 d5 exd5?! (15...Qd6

keeps White’s edge minimal) 16 Nfxd5 Nxd5 17 Nxd5 += Jelen-
I.Almasi, Vienna 1996.

b) 11...0-0 12 0-0 Bc6 13 Rfd1 Rd8 14 Qe3 (14 Rac1 Nbd7 15 Bg3 a6 16
Kh1 Bxg3 17 Nxg3 Nb6 18 Nce4 Qe7 19 Nc5 Bd5 =+ Naumann-Grund,
German Junior Ch, Halle 1996) 14...Nbd7 15 Rac1 (15 Bxe6!? Bxf4! is
sharp but roughly balanced) 15...Nb6 = Henriquez Villagra-Perez Gormaz,
Chilean Ch, Iquique 2013.

We now return to 10 Bg5 (D):

Black to play

10...Be7 11 0-0

11 Rc1 may be more challenging:

a) 11...Bc6 12 Bxf6 Bxf6 13 d5 exd5 14 Nxd5 is probably a bit better for

White; e.g., 14...Qa5+ 15 Nec3 Bxc3+ 16 Rxc3 0-0 17 0-0 Bxd5 18
Qxd5
Nc6 19 Qxa5 Nxa5 (Goldin-Altounian, USA Ch, San Diego 2004) 20 Bd5
+=.

b) The new move 11...Qb6 may improve; e.g., 12 0-0 (against 12 Qd2,
Black can consider 12...Nc6!?, planning ...Na5) 12...0-0 13 Kh1 Na6 14 d5
Rad8 =.

11...Bc6?!

This seems inaccurate. Probably Black should play 11...0-0, allowing 12

d5, as it’s not even clear that White is fully equal after 12...Rd8 13 Rc1?!
Bc6; e.g., 14 Qe1 Nxd5 15 Nxd5 exd5 16 Bxe7 Qxe7 17 Nd4 Qf6 18 Qc3
Be8 19 Rfe1 Nc6 20 Nxc6 bxc6 21 Qxf6 gxf6 22 Ba4 Rab8 =+.

12 Bxf6?!

12 f4! is better; e.g., 12...Ng4 13 Bxe7 Qxe7 14 Qd3 0-0 15 h3 +/–

when White threatens f5 (15...Nh6 16 Ng3 doesn’t change much).

12...Bxf6 13 d5 exd5 14 Nxd5 Bxd5 15 Qxd5 0-0 (D)

White to play
The position is a bit drawish and probably equal, but if given a choice
I would take the black pieces.

16 Nc3 Bxc3

Trading in the strong bishop for a slight pawn weakness. 16...Nc6 is

good too, but there is no way to get anything more than a slight pull.

17 bxc3 Nd7 18 Rae1 Nc5 19 Re5 Nd3 20 Re4

20 Re3 is more solid but Black has some

pressure. 20...Rad8 21 Qc4 Nc5 22 Rd4 Rde8

23 Qd5?!

After 23 Rfd1 Qe5 24 h4 b6 Black has a plus but White should be able
to hold.

23...Re3 24 Rg4?!

It’s hard to pinpoint the decisive inaccuracy but this may be it. After 24
Qc4 Qe7 Black has an advantage but White may be able to hold.

24...g6

Suddenly White is too uncoordinated and there seems to be no way to

reduce the pressure against the weak c-pawn.

25 h4?!

25 Bc4 Ne6 26 Qd2 Qe5 27 Bxe6 Qxe6 28 a4 Rc8 –/+.

25...Rxc3 26 h5?!

26 Rc4 is somewhat better, though 26...Rd8 27 Qg5 Rxc4 28 Bxc4

Qb6 leaves White in deep trouble.

26...Nxb3 27 axb3 Qb6+ 28 Kh1?

White also loses after 28 Qd4 Rxb3, if more slowly.

28...Rc5 0-1

Theory Magnifiers

1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4 5 Bxc4 Qc7 (6 Bb3 Bg4 7 Nf3)

24A: (6 Qd3)

24B: 6 Qb3

24C: (7 Nf3)

Black’s 4th and 5th moves challenge conventional opening advice by

activating White’s bishop and bringing the queen out early. The
resulting position needs care in order not to collapse, so these
magnifiers are mandatory reading if you want to reach a playable
middlegame.

24A: Rare 6th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4 5 Bxc4 Qc7)

6 Qd3 (D)

Or:

a) 6 Qe2 Be6! 7 Bb5+ (A.Biro-D.Biro, Hungarian Team Ch 1998/9; 7 b3

may keep a tiny edge) 7...Nc6 8 Nf3 a6 9 Nc3! axb5 10 d5 Bxd5 11
Nxd5 Qd8 12 Nc3 e6 =.

b) 6 Be2 Nf6 7 Nf3 e6 8 0-0 Be7 9 Nc3 0-0 (9...Nc6! is solid) 10 Ne5
Nc6 (10...Rd8?! 11 Nb5 += Vaganian-Gabriel, Bundesliga 2004/5) 11
Bf4 Qb6 with chances for both sides.
 Black to play

This looks slightly peculiar but demonstrating full equality for Black proves
challenging.

6...Nf6 7 Nc3 e6

7...a6!? may be fine for Black but the resulting positions seem too
demanding for a peripheral line.

8 Nf3 Nc6 9 0-0 Be7 10 Bg5

10 Nb5 Qd8 11 Bf4 and now 11...Nb4! 12 Qe2 Nbd5 = seems to improve
on Lautier’s 11...0-0.

10...0-0 11 a3

11 Rfe1 Rd8 12 Rad1 Na5 13 Bb5 a6 14 Ba4 Nc4 15 Qe2 b5 16 Bb3 Bb7 17
Ne5 Rac8 18 a4 Qb6 = T.Georgescu-Grecescu, Romanian Team Ch,
Predeal 2007.

11...Rd8 12 Rfd1 h6 13 Bh4 Nh5 14 Bxe7 Nxe7!?

This offers Black a more comfortable game than 14...Qxe7, which
was played in Lautier-Kallai, French Cup 2001.

15 Ne5 Nf6 16 Rac1 (D)

Black to play

Lautier evaluates this as +=, which seems reasonable. However, it is

probably one of those small advantages that evaporate after a few accurate
moves.

16...Nf5 17 Nb5 Qe7 18 Bb3

Or:

a) 18 g4?! Nxg4! 19 Nxg4 Qg5 20 h3 h5 gives Black the more

comfortable position.

b) 18 Ba2 seems quite similar to our main line, but is probably slightly
weaker as some tactical ideas don’t work unless the light-squared
bishop is protected by the queen.

18...a6
Now:

a) 19 Nc7 Rb8 20 Qc3 Rd6 21 Bc2 Qd8 22 Bxf5 exf5 23 h3 Be6 24

Nxe6 Rxe6 =.

b) 19 Rc7 is more testing:

b1) 19...Nd7? 20 Nxf7! +/–; e.g., 20...Qxf7 21 d5.

b2) 19...Bd7 20 Nc3 Qd6 21 Rxb7 Rab8 and Black is fine:

b21) 22 Rxb8 Qxb8 23 Nxd7?! Rxd7 24 Ne2?! (24 Ba4 Rxd4 is only
marginally better for Black) 24...Ng4 25 Qh3 Nfe3! 26 fxe3 Qxb3 =+.

b22) 22 Nc4 Qxd4 23 Rxb8 Rxb8 24 Qxd4 Nxd4 25 Rxd4 Rxb3 =.

24B: 6 Qb3

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4 5 Bxc4 Qc7)

6 Qb3 (D)
 Black to play
6...e
6

Now we return to more explored territory. 6...Be6?! is complicated after

both 7 d5 Bxd5 8 Bxd5! and 7 Na3 Bxc4 8 Nxc4 Nc6, but the risk seems
greater than the potential benefits.

7 Nf3

White has the following more independent options:

a) 7 Nc3 Nc6 8 Nb5 Qa5+ 9 Bd2 Bb4 10 0-0-0?! Nf6 11 Bxb4 Nxb4 12 a3
Nbd5 13 Bxd5 Nxd5 14 Nd6+ Ke7 15 Nc4 Qa6 =+ Istratescu-
Wiesinger, Schwarzach 2016.

b) After 7 Ne2 Nf6 8 Nbc3 Be7 (covering b5 with 8...a6!? seems safer) 9
Bf4 Qa5 10 0-0 0-0 11 Rad1, Black should probably play 11...a6 as
11...Nc6?! (Movsziszian-Miles, Hamburg 1995) can be met by 12 d5 with
annoying pressure.

7...Nc6 8 0-0 Nf6 9 Nc3 a6 (D)

 White to play

10 Bd3

Or 10 Be2 Be7, and now:

a) 11 Bg5 0-0 12 Rac1 Ng4 13 Ne4 Bxg5 14 Nexg5 h6 15 Ne4 Qf4 16

Qd3 Nf6 = Lputian-Kallai, West Berlin 1989.

b) 11 Be3 0-0 12 Rac1 Qd6 13 Rfd1 (13 Bg5 Nxd4 14 Nxd4 Qxd4 15 Bf3 =)
13...b5 14 d5 Na5 15 Qc2 exd5 16 Ng5 Bb7 17 Bf3 Rac8 18 Qb1 h6 19 Nge4
Nxe4 20 Bxe4 Nc4 21 Nxd5 Nxe3 22 Nxe3 Bxe4 23 Qxe4 Rxc1 24 Rxc1 =.

10...Be7

Black will get his king to safety, but White may keep a slight plus:

a) 11 h3 0-0 12 Bg5 Rd8 13 Rad1?! (13 Rac1 =) 13...Nxd4 14 Nxd4 Rxd4 15

Be3?! Rh4! 16 Be2 b5 17 Rc1 Qb8 18 f4 Bb7 19 Kh2 g5 –+
Drozdovsky- Ivanchuk, Mainz rapid 2007.

b) 11 Be3 0-0 12 Rac1 and White has an edge:

b1) 12...Bd7 13 Bb1 Qd6 14 Rfd1 Nb4 15 d5 (15 Ne5!?) 15...Rac8 16 dxe6
Qxe6 17 Qxe6 Bxe6 18 Ng5 Bg4 19 f3 Bh5 20 a3 += Vaganian-
Lukin, Telavi 1982.

b2) 12...b5 13 Ne4 Bb7 14 Rfe1 Rfc8 15 Ne5 Nxe4 16 Bxe4 Qd6 17 Rcd1
Na5 18 Qd3 Bxe4 19 Qxe4 Nc6 20 Nxc6 Qxc6 21 d5 Qd7 22 Bd4 +=
P.Ax- W.Löffler, corr. 2007.

24C: Rare 7th Moves

(1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 dxc4 5 Bxc4 Qc7)

6 Bb3 Bg4 7 Nf3

This is quite rare but still dangerous.

7 Ne2?! Bxe2! 8 Kxe2 e6 (8...Nc6 9 d5 0-0-0 is also interesting) 9 Nd2
Nf6 10 Nf3 Nc6 11 Re1 Bb4 was more comfortable for Black in P.Pascal-
Mellado Trivino, Andorra 2000.

7...Bxf3

Not 7...e6? 8 Nc3 Nf6 9 0-0 +/–, when one important line is 9...Be7 10 h3
Bh5 11 d5 exd5 12 g4 Bg6 13 g5 Nh5 14 Nxd5 Qd8 15 Re1 Nc6 16 Nd4
+–.

8 Qxf3!? (D)

After 8 gxf3 e6 9 Nc3 Bb4 Black has equal chances in an interesting position.

Black to play
8...Qxc1+

Black cannot even contemplate moves like 8...e6?? 9 Bf4, when White’s
lead in development is too great.

9 Ke2 Qxh1!
Black must force events. While 9...Qxb2+? 10 Nd2 Nc6 11 Qxf7+ Kd7 has
actually scored quite well for Black in practice, it is extremely risky for him,
and probably objectively lost after 12 Rhd1!.

10 Qxf7+

Or:

a) 10 Qxb7 e6 11 Qc8+ Ke7 12 Qc5+ Ke8 13 Qc8+, with a repetition,

is among the simpler drawing lines.

b) 10 Nc3 Qxa1 11 Qxf7+ Kd8 (11...Kd7?? 12 Be6+ Kc6 13 Qf3+ +–)

12 Qxf8+ Kd7 13 Qf5+ Kd8 14 Qf8+ Kd7 =.

10...Kd8! (D)

White to play

Forced.

11 Nd2
This is more interesting than 11 Nc3 Qxa1 12 Qxf8+ Kd7 and 11 Qxf8+
Kc7 12 Nc3 Qxa1 13 Nd5+ Kd7 14 Qf5+ Kd8 15 Qf8+, with repetitions.

11...Qxa1 12 Qxf8+ Kd7 13 Ba4+

Now 13...Kc7 14 Qf4+ Kd8 15 Qf8+ is another simple draw. More

entertaining is 13...Nc6 14 Qxa8 Qxb2 15 d5 Kc7 16 dxc6 Nf6 17 Qxh8,
although this is also a draw after 17...Ne4 18 Qe8 Qxd2+ 19 Kf3 Qd3+
20 Kf4 g5+ 21 Ke5 Nc5 22 Qxe7+ Kb6 23 cxb7 Qe4+ 24 Kf6 Qg6+ 25
Ke5
Qe4+ 26 Kf6.
5: Early Deviations

Lesson 25: Maroczy’s 3 f3 and Rare Third Moves

By: TRH

There’s something nice about being positionally worse out of the opening
... You don’t have to worry about making concessions, because you’ve
already made concessions.

Magnus Carlsen

1 e4 c6 2 d4 d5 3 f3 (D)

Black to play

This variation is better known as the ‘Fantasy Variation’, but we don’t see
any reason not to give the naming honours to Maroczy, who was among
the first strong players to use it with any success.
Also the question of whether this is a main line is open to debate. It’s
relatively rare but has been played several times by Li Chao, Ivanchuk
and Morozevich, so underestimate it at your peril. Anyhow, we decided
that it didn’t deserve an alternative line for Black, so that probably
groups it with the minor lines.

3...e6

Black has many alternatives, including the aggressive 3...e5 and 3...Qb6,
but I believe this is the simplest and strongest. If I had to pick a colour, I
would already choose Black. We shall keep the game closed, only taking
on e4 if there is a specific purpose, and otherwise highlight the drawbacks
of 3 f3.

Model Game 25

A. Ward – W. Steiger

corr. 2007
1 e4 c6 2 d4 d5 3
f3

The really rare third moves are not much to worry about but are
briefly examined in 25A.

3...e6 (D)

Yes, it’s arguably a French with two additional moves, f3 and ...c6. And
while it’s true that Black normally plays ...c5, the more modest c-pawn
move is also useful in a few cases. At worst it is a loss of one tempo. That
cannot be said about White’s extra f-pawn advance, which downright hurts
his position in several lines.
 White to play
4 Nc3

White has some minor alternatives:

a) 4 e5?! normally leads to a position that usually occurs from the French
after 4...c5 5 c3 Nc6 6 f4. Luckily, it’s dangerous only for White. After
6...Nh6 7 Nf3 Qb6 the pressure on d4 is hard for White to handle. Then 8
Bd3 cxd4 9 cxd4 Bd7 10 Bc2 Nb4 has been played in many games, and is
very comfortable for Black, while after 8 Na3 cxd4 9 cxd4 Bxa3 10 bxa3
Nf5 and the d-pawn is about to fall.

b) 4 Be3?! dxe4 5 Nd2 exf3 6 Ngxf3 is one of several options related to

the Blackmar-Diemer Gambit (see 1A). In general, White has some
compensation in positions like this. However, Black’s position is solid
and free of weaknesses, so against patient play, White’s initiative will
fade out, leaving Black with a good position. Here are some lines after
6...Nf6:

b1) Following 7 Bd3 Be7 8 0-0, Black has scored well with 8...0-0 and
8...Nbd7, but I suggest 8...Nd5 as an improvement. Black will continue
with castling, and at a good moment, exchange on e3. White fails to get
enough
compensation.

b2) Against 7 Nc4, which is recommended by Bezgodov, 7...b5 is the most

direct reply (with 7...Nd5 a good alternative move-order). After 8 Nce5
Nd5! 9 Bd2 f6! 10 Nd3 Bd6 11 g3 0-0 12 Bg2 Qb6 White had insufficient
compensation in P.McCollum-R.Therrien, USA 1998.

4...Bb4 (D)

White to play

Black renews the threat of taking on e4 followed by a check on h4.

5 a3

White is ready to sacrifice a pawn. Other moves are rather toothless. For 5
Ne2 and some even rarer 5th moves, see 25B and for 5 Bf4, see 25C. 5
e5 c5 6 a3 Bxc3+ 7 bxc3 Qc7 is a line where the extra move f3 does
White no favours in comparison with the Winawer.

5...Bxc3+ 6 bxc3 dxe4

Black has a few alternatives, but there is no good reason not to take the pawn.

7 Qe2

This is by far the most popular move. Otherwise:

a) 7 fxe4?! has been played on a few occasions, but Black scores heavily.
For instance, 7...Qh4+ 8 Kd2 Nf6 and it’s clear that something has gone
seriously wrong for White.

b) 7 Nh3 is best met by 7...Qa5, forcing White to make an awkward move

to defend the c-pawn. 8 Qd2 (8 Bb2 Nf6 9 fxe4 e5 is comfortable for Black)
8...Nf6 9 Be2 e5 10 Ng5 exf3 11 Bxf3 h6 12 Ne4 Nxe4 13 Bxe4 f5 14 Bf3
0- 0 was better for Black in Moussard-Diepeveen, Belgian Team Ch
2012/13.

c) 7 a4!? prepares to develop the dark-squared bishop to a3. I suggest

7...Nd7 8 Ba3 Ngf6, when Black shouldn’t have any worries. Sooner or
later the a3- f8 diagonal can be blocked by ...c5.

7...c5! (D)

White to play
This is often the best reaction to a gambit: first take the pawn, and then
return it to gain time for development.

8 Qxe4

After 8 Qe3, 8...cxd4 has only been played a few times, but I believe it’s a
good move. After 9 cxd4 Nc6 10 Bb2 Qb6 11 0-0-0 Black played 11...exf3
in two correspondence games, but I see no reason to help White’s
development, and suggest instead 11...Nf6:

a) 12 Qg5 is best met by 12...Rg8 13 fxe4 Nxe4 14 Qh4 Nf6, when White
has sufficient compensation, but nothing more than that.

b) After 12 fxe4 0-0 13 Nf3 White has the bishops and a strong centre,
while Black has play against White’s king; e.g., 13...Bd7 14 Bd3 Na5 15
Rhe1 Rac8 16 d5 Nb3+ 17 Kb1 Nc5 18 Qd4 Rfd8 and in this sharp
position chances are even.

8...Nf6 9 Qh4

Or:

a) 9 Qe3 cxd4 10 cxd4 Nc6 11 Bd3 Qxd4 12 Qxd4 Nxd4 13 Ne2 Nc6 14
Bb2 b6 15 Nd4 Bb7 left Black a healthy pawn up in D.Cumming-V.Anikeev,
corr. 2015.

b) 9 Bb5+ Nbd7 10 Qe2 a6 11 Bd3 b5 12 Nh3 cxd4 13 cxd4 Nb6 14

c3?! Bb7 =+ M.Maddens-Sagar, Dieren 2016.

9...cxd4 10 cxd4 Nc6 11 c3

White has gained the bishop-pair at the cost of development problems

(which is often the case). Here Black needs to attack before White
consolidates.

11...e5 12 Ne2 exd4 13 Nxd4 0-0 (D)

 White to play
14 Kf2

14 Be2?! has the point that 14...Re8?! 15 Ra2 Bd7 16 0-0 =+ allows White
to defend along his second rank, but 14...Qa5! cuts across this plan as
now 15 Bd2 Re8! keeps the white king in the centre.

14...Ng4+ 15 Kg3 Qa5

15...Nxd4 16 Qxd8 Nf5+ 17 Kxg4 Rxd8 is a risk-free path to a

slight advantage. The move played in the game is more ambitious.

16 Bd3?

With 16 Nxc6 bxc6 17 h3 White would survive, although Black is slightly

better after 17...Nf6.

16...h6 17 Nxc6 bxc6 18 Bf4

After 18 fxg4 Qxc3 Black regains the piece with a decisive advantage.
18...Rd8 19 Bc2 Nf6 20 Rhe1 Rd5 21 Be3 Qxc3 22 Rac1 Qxa3

With two extra pawns Black should be technically winning.

23 Kf2 Be6 24 Qa4 Qd6 25 Kg1 a5 26 Bf2 c5 27 Bd1 c4

One of the advantages of leading by two pawns is that you can return one
to obtain control of the situation.

28 Rxc4 Rd2 29 Rd4 Rxd4 30 Bxd4 Qb4 31 Bf2 Nd5 32 Bb3 Nf4 33 Bg3
Qc5+ 34 Bf2 Qg5 35 g3 Bxb3 36 Qxf4 Qxf4 37 gxf4 a4 38 Re7 a3 39 Ra7
Rxa7 40 Bxa7 f6 41 Kf2 a2 42 Bd4 Kf7 43 Ke3 Ke6 44 Ba1 Kd5 45 Kd3
Bd1 0-1

White’s king can’t defend both sides of the board.

Theory Magnifiers

1 e4 c6 2 d4 d5 (3 f3 e6 4 Nc3 Bb4 5 Bf4)

25A: (3 Bd3)

25B: (5 Ne2)

25C: 5 Bf4

25A contains nothing essential and was included for the sake of
completeness. 25B and 25C are also not dangerous from a theoretical
point of view. However, some of the players employing 3 f3 are quite
dangerous attackers, so if you want to save some effort over the board you
may prefer to have a good look at the lines beforehand!

25A: Really Rare 3rd Moves

(1 e4 c6 2 d4 d5)

3 Bd3
This move doesn’t have much going for it but at least it protects the e-
pawn. In passing we mention two moves that don’t:

a) 3 Nf3? has been tested by die-hard ‘gambiteers’ but it’s no

coincidence that Black has won all games after 3...dxe4 4 Ng5 Nf6 5 Nc3
Bf5 6 Bc4 e6.

b) 3 c4?! is a gambit that may also arise by a Slav move-order. After

3...dxe4 4 Nc3 Bf5 White has been unable to demonstrate full
compensation:

b1) 5 Nge2 e5! 6 dxe5 Qxd1+ 7 Nxd1 Nd7 8 Ng3 Bg6 9 Bf4 Bb4+ –/+.

b2) 5 f3 e5!? 6 fxe4 exd4 7 exf5 dxc3 8 Qxd8+ Kxd8 9 Bg5+?! Be7 10 0-0-
0+ Nd7 –/+.

b3) 5 g4 Bg6 6 Bg2 and once again the central thrust 6...e5! gives Black
an excellent game with no risk.

3...dxe4 4 Bxe4 Nf6 5 Bf3 (D)

Black to play

This set-up is even less dangerous than the similar French variation where
Black (with a pawn on e6 rather than on c6) at least has to find a way
to activate his light-squared bishop.

5...Bf5

Problem solved – Black has equalized. He should just follow up in

standard fashion with ...e6, ...Bd6, ...0-0 and quite soon ...c5. But let’s
look at a few practical examples.

6 Ne2 e6 7 0-0

Or:

a) 7 Ng3 Bg6 8 h4 (8 0-0 probably keeps chances equal) 8...h6 9 h5 Bh7

10 c3 Qc7 11 Nd2 Nbd7 12 Nge4 and now Black has a lot of alternatives
to 12...Bxe4 13 Nxe4 Nxe4 14 Bxe4 Nf6 15 Bf3 Bd6, which only gave him
equality in Nanu-Nisipeanu, Romanian Team Ch, Eforie Nord 1997
(though he won fairly quickly!). Possibly 12...e5 is better for Black already.

b) 7 Nd2 Bd6 8 Nc4 Bc7 9 0-0 Nbd7 10 Bf4 Qb8 11 Bxc7 Qxc7 12 Ng3
Bg6 13 Qe2 0-0 14 Ne5 Nxe5 15 dxe5 Nd5 = Nuber-Svetushkin, Dresden
2008.

7...Bd6 8 Ng3 Bg6

Black is (still) fully equal:

a) 9 c4 Nbd7 10 Nc3 0-0 11 Qb3 Qc7 12 Be3 Rfe8 13 Rac1 e5 =+

Nanu- Cioara, Bucharest 1999.

b) 9 Bg5 Nbd7 10 Nd2 Qc7 11 Nc4 Be7 12 Qd2 Nb6 13 Nxb6 Qxb6 14 c3
h6 15 Be3 Nd5 = San Claudio Gonzalez-Guerrero Olmos, Asturias Ch,
Gijon 2002.

25B: 5 Ne2 and Rare 5th Moves

(1 e4 c6 2 d4 d5)

3 f3 e6 4 Nc3 Bb4 (D)

 White to play

5 Ne2

Or:

a) After 5 Qd3 White typically castles queenside. I suggest the new move
5...b5, which has numerous ideas. The main one is to play ...Be7, ...b4,
...a5 and ...Ba6. This way we are ready to attack White’s queenside, as
well as hitting the queen on d3. Another idea is ...Nd7-b6-c4. As the
position has never been tested, and the variations are non-forcing, I shall
limit myself to a few brief lines. I don’t see an advantage for White:

a1) 6 Bf4 Nf6 7 0-0-0?! Be7 and Black is ready to play ...a5 with an attack.

a2) 6 exd5 exd5 7 Bf4 Nf6 8 0-0-0?! 0-0 9 g4 Nbd7 10 h4 Nb6 11 h5

Nc4 and Black’s position looks much better.

a3) After 6 Nh3 a5 7 Nf2, 7...Nf6 is one of many good moves; e.g., 8
Be2 Ba6 9 0-0 Be7 =.

a4) 6 a3 Be7 7 Be3 a5 8 Qd2 Nf6 9 e5 Nfd7 and Black’s attack looks quite
promising; e.g., 10 f4 Nb6 11 Nf3 b4 12 Nd1 bxa3 13 bxa3 Ba6 and when
the knight lands on c4, Black will be better.

b) 5 Bd2 Ne7 and now:

b1) 6 Nce2 has been played in a number of grandmaster games, with

Black usually capturing on d2. I instead suggest keeping the tension with
6...Ba5. A possible continuation is 7 c3 Bb6 8 Qc2 c5, which resembles a
good French for Black.

b2) 6 a3 should probably be met by 6...Ba5 as playing f3 weakened

White’s dark squares, so it makes sense to keep the dark-squared bishop
on the board. After 7 Bd3 Nd7 (D) we have a split:

White to play

b21) 8 e5 Bb6 is just bad for White. When White defends the d-pawn,
Black plays ...c5 with an excellent position.

b22) After 8 b4 Bb6 9 Nge2 e5 the central tension favours Black.

b23) The aggressive 8 f4 has been played in a handful of games. After 8...c5
9 e5 c4 10 Be2 (10 Bf1!?) 10...Nf5 11 Nf3 Bb6, defending the d-pawn is a
little awkward. I like Black’s way of handling the position in Alonso
Gonzalez-Voveris, corr. 2017: 12 Bc1 f6 13 g4 Nxd4 14 Nxd4 fxe5 15 Nf3
e4 and Black had sufficient compensation for the piece, and White was
soon forced to return the material.

b24) 8 Nge2 0-0 9 0-0 e5 10 exd5 exd4 11 Nxd4 Ne5 12 Qe2 Nxd3 13
Qxd3 Bb6 14 Kh1 Nxd5 and thanks to the bishop-pair, Black was better in
Marcinkiewicz-Raykin, corr. 2010.

5...Ne7 (D)

5...dxe4 is more common. But with White’s knight on e2, I prefer to place
my knight on e7. Putting it on f6 would allow White to play e5 and then f4.
Now these moves wouldn’t make much sense.

White to play
6 Be3

6 a3 Ba5 7 Be3 0-0 8 g3?! (8 Qd2 Nd7 transposes to our main line
below) 8...dxe4 9 fxe4 e5 10 Bg2 Bg4 11 b4 Bb6 12 Qd2 Nd7 =+
Podinić-
Radovanović, Bosnian Team Ch, Banja Vrućica 2013.

6...0-0 7 Qd2 Nd7 8 a3

After 8 Ng3 (Piorun-Tomczak, Polish Team Ch 2015) I suggest 8...dxe4

with the idea of playing 9 fxe4 c5, when White’s centre is about to crumble,
leaving Black with a good position.

8...Ba5

Now:

a) After 9 Nf4, 9...c5 10 dxc5 d4 11 Bxd4 e5 12 Be3 exf4 13 Bxf4, as in

D.Mayers-P.Taylor, London 2000, is hard to evaluate but Black has plenty
of less forcing options.

b) 9 Ng3 f5! 10 exf5 Nxf5 11 Nxf5 exf5 12 Bd3 Nf6 13 0-0 Nh5 = Timofeev-
Riazantsev, Russian Ch, Moscow 2008.

25C: 5 Bf4

(1 e4 c6 2 d4 d5 3 f3 e6 4 Nc3 Bb4)

5 Bf4 (D)
 Black to play

This is the most common move but can hardly be dangerous.

5...Nf6

White’s bishop on f4 means we don’t have to worry about White

advancing the e-pawn.

6 Qd3

This position has been played many times, although I don’t understand why
White would willingly aim for this. It’s hard to imagine how White could
create an attack, while Black’s counterplay comes quickly. Instead 6 e5
Nfd7 gives Black a good version of the French, as the bishop does not
belong on f4. After 7 a3 Bxc3+ 8 bxc3 c5, the pawn on f3 hinders White’s
development, which means White will have to move the bishop and then
the f-pawn. Black is doing fine after moves like ...Nc6 and ...Qa5. Rambaldi
gives 9 a4! and recommends further analysis. As a first take I offer 9...Nc6
10 Be3 Qa5 11 Qd2 Ne7 12 f4 and now the committal 12...c4!?. Black is
intending ...Nb6 and ...Nxa4, and it’s clear that White’s a-pawn is lost.
White’s chances lie on the kingside. A sample line is 13 Nf3 Nb6 14 g3
Nxa4 15 Ra3 b5 16 Bg2
Qb6 when Black is planning ...a5, ...Bd7 and eventually ...b4. White is
under pressure.

6...b5

I believe this is a new way to handle Black’s position. The idea is very
simple: we are going to play ...a5, ...Ba6, ...Be7 and ...b4. Black already
seems better; for instance, 7 Ne2 Be7 (7...a5 8 Ng3 Ba6 also looks OK) 8
a3 a5 9 Ng3 Ba6 10 Qd2 0-0 11 Bd3 b4 12 axb4 Bxb4 with the idea of
playing
...c5.

Lesson 26: 2 Nc3 and Introduction to the Two Knights

By: TRH

The most powerful weapon in chess is to have the next move.

David Bronstein

1 e4 c6 2 Nc3

The other knight move, 2 Nf3, is just as good a way to reach the Two
Knights. Note that the independent line 2...d5 3 exd5 cxd5 4 Ne5
transposes to 19A after 4...Nc6 5 d4 e6 6 Bb5.

2...d5

There is no advantageous way for Black to prevent a transposition to

the Classical Variation if that is what White in fact wants.

3 Nf3 (D)

3 d4 obviously is the main line, while Goldman’s 3 Qf3 is the subject of 26A.
 Black to play

Now Black’s main move has always been 3...Bg4. That’s the subject of
Lesson 27. However, there are interesting alternatives. One of them is to
aim for the same structure as in the Korchnoi Variation.

3...dxe4

I believe the reason why you don’t see this move recommended in
repertoire books about the Caro-Kann is that it doesn’t fit well with the
Capablanca Variation. However, in this book we have more freedom as
we have already seen the advantages of the Korchnoi Variation.

4 Nxe4

The variation I am about to recommend is likely to come as a surprise.

Black takes some risks to reach a position that is demanding for both
players. The surprise effect, and the fact that you (hopefully) have done
your homework, should put you in a favourable situation.

4...Nf6
Sadly for Black, it’s impossible to transpose to the Capablanca Variation
with 4...Bf5?. The problem is that 5 Ng3 Bg6? 6 h4 h6 7 Ne5! Bh7 8 Qh5
g6 9 Bc4! e6 10 Qe2 gives Black a miserable game. In fact, the second
World Champion showed how to handle a position like this: 10...Bg7? 11
Nxf7! 1-0 Em.Lasker-Radsheer, Netherlands simul 1908.

5 Qe2

This was first played by Paul Keres in 1950 (with an unexpected windfall
as Black replied 5...Nbd7??), but was rarely repeated until the 1990s.
However, its popularity exploded during the last decade when White
started to feel the need for an alternative to 5 Nxf6+.

5...Na6 (D)

White to play

Surprise! This aggressive move has only been played a couple of dozen
times, mostly in the last two years, and attracts players willing to take a
risk. Strong grandmasters who have played it include Rapport, Nisipeanu,
Kovalenko and Oparin.
 Model Game 26

Vachier-Lagrave – Rapport

Caleta 2018

1 e4 c6 2 Nf3 d5 3 Nc3 dxe4 4 Nxe4 Nf6 (D)

White to play

Rapport occasionally plays the regular Korchnoi Variation too but maybe
he found it more tempting with White’s knight already committed to f3.

5 Qe2

Or:

a) After 5 Nxf6+ exf6, White normally plays 6 d4 with a transposition to a

somewhat toothless version of the Korchnoi Variation. The knight is slightly
misplaced on f3 and White is not in time to prevent Black’s ideal set-up.
This variation is covered in Model Game 1. For some rather harmless 6th-
move alternatives, see 26B.
b) 5 Ng3 is, like in the Korchnoi Variation, once again possible. Although it
has been played by a few strong players, it’s not challenging. 5...h5 is the
most aggressive option, and it has been played by Rapport. However, I like
5...g6 even more, as it makes White’s last move look silly: with a pawn on
g6, the knight is badly placed on g3. 6 Bc4 Bg7 7 d4 c5 8 dxc5 and here,
instead of 8...Qa5+ as played in G.Mammadova-Ushenina, Turkish Team
Ch, Kocaeli 2015, I like 8...Qxd1+ 9 Kxd1 Nbd7. There is no adequate way
to protect the c5-pawn, and Black is at least equal.

c) 5 d3 is obviously not critical, as Black can take on e4 and reach an

equal endgame. However, if you prefer a more complicated position, I
suggest 5...Nbd7. Moves like 6 Ng3 and 6 Ned2 can’t be dangerous to
Black, and the only logical follow-up I can see is 6 Bf4. A possible line is
6...Nxe4 7 dxe4 Qb6, which comes with a tactical point: 8 Rb1 (8 Bc1 is
the engine’s recommendation, but Black is obviously doing fine after 8...e6
or 8...e5) 8...e5! 9 Nxe5? Qb4+ 10 Qd2 Qxe4+ 11 Qe2 Bb4+ and White’s
only move is 12 Kd1 –/+.

We now return to 5 Qe2 (D):

Black to play
5...Na6!?

Other moves are less appealing, to varying degrees:

a) 5...Nbd7?? has been tried, rather unsuccessfully, in 11 games (no

names mentioned). In all games White replied 6 Nd6#.

b) 5...Nxe4 may seem a simple answer to White’s play, but 6 Qxe4 Nd7 7
Bc4 Nf6 8 Ne5! forces 8...e6 and after 9 Qe2 White has a pleasant
advantage due to Black’s restricted bishop on c8. A few years ago I
looked at this from Black’s perspective and concluded that it’s close to
unplayable even if the engine thinks 9...b5 is viable. I was fortunate
enough to get a chance to use my analysis as White: 10 Bd3! Qc7 11 b3
Bb7 12 Bb2 Bd6 13 0-0 0-0 14 Rae1 a6? 15 f4 Rae8? 16 Rf3 c5 17 Rh3
g6 18 c4 bxc4 19 bxc4 Qe7 20 Qf2 Rd8 21 Qh4 Nh5 22 Qxh5 and Black
soon had to resign in T.R.Hansen- N.Kumar, Reykjavik 2017.

6 d4 (D)

For 6 Nxf6+ gxf6, see 26C.

The modest 6 d3 quickly leads to equality after 6...Nxe4 7 dxe4

(Gajčin- J.Rapport, Belgrade women 2018) 7...e5 =, as the pawn is
poisoned: 8 Nxe5?? Qa5+ –+.
 Black to play
6...Qd5 7 Nc3

This is the most critical, and also the choice of the highest rated players.
Here too White can exchange knights with 7 Nxf6+ gxf6, transposing to
26C.

7...Qa5

7...Qh5!? has been tried by Kovalenko but there is no need to deviate

from the main line.

8 Qe5?!

A natural response when surprised with this line. However, 8 Ne5 is critical
and should be met by 8...Nb4. For the details, see 26D. I spent a full day
trying to make 8...g6?! work. The problem was 9 Qc4 Be6 10 Nxc6. At first
I was convinced there was a way to get good compensation, but I ended
up liking White’s position too much. That being said, the position is
extremely complex and it would be hard for White to navigate correctly.

8...Qxe5+ 9 dxe5 Nb4! (D)

 White to play

If it weren’t for this move, White would be better.

10 Bd3

After 10 exf6 White manages to pick up the knight following 10...Nxc2+ 11

Kd1 Nxa1 12 Bd3 (12 fxe7? Bxe7 13 Bd3 Bg4 14 Ke2 0-0-0 15 Be3 might
look OK for White until Black plays 15...Rxd3 16 Kxd3 Bf5+ 17 Ke2 Nc2
and is close to winning) 12...Bg4 13 Ke2 0-0-0 14 Bd2 exf6 15 Rxa1 Bc5
but Black’s active pieces and open files for the rooks guarantee him a clear
advantage.

10...Nxd3+ 11 cxd3 Nd7

11...Nd5?! 12 Nxd5 cxd5 13 Be3 leaves White slightly better due to his
lead in development and the fact that the d4-square will be useful for his
minor pieces.

12 Be3 (D)

The position is simplified, but I believe Black has every reason to be

ambitious. After all, he has the two bishops and the better pawn-structure.
In the phase that followed, neither player managed to handle the
subtleties very well. Both players had their chances to fight for an
advantage, but let it slip away.

Black to play
12...Nb6?!

The knight was well placed on d7, and it was time to develop the other
pieces instead.

I like 12...g6! 13 Ke2 (13 e6 fxe6 14 Bd4 gives White some compensation,
but with an extra pawn and the bishop-pair, Black seems better after
14...Rg8) 13...Bg7. If Black manages to provoke the d4 advance, White
can no longer use the d4-square for his minor pieces. This is why the
knight belongs on d7 – to put pressure on the e-pawn. After 14 d4 Nb6
Black will castle kingside, and, at the right moment, open the position with
...f6. I prefer Black.

13 Ke2
13 a4 could be enough to give White a slight plus. Ideally, Black would
like to meet this with 13...Be6, but 14 Nd4 is awkward for Black.

13...Be6?! 14 Nd4 Bd5?! 15 Nxd5?!

Once again 15 a4! seems strong, although this time it involves a pawn
sacrifice. 15...Bxg2 16 Rhg1 Bh3 17 a5 Nc8 (17...Nd7 18 Rg3! +/–) 18
a6 and White is in the driving seat.

15...Nxd5

After this move the game never leaves the evaluation of being equal. It
seems to me the players are steering towards a draw from here on.

16 e6 g6 17 exf7+ Kxf7 18 Nf3 Bg7 19 Ng5+ Ke8 20 Rab1 a5 21 Ne4 b6

22 Rhc1 Kd7 23 Nc3 a4 24 Nxd5 cxd5 25 d4 Rhc8 26 Kd3 e6

When the rooks come off the board, the draw is inevitable.

27 Rxc8 Rxc8 28 Rc1 Rxc1 29 Bxc1 Kc6 30 b3 axb3 ½-½

Theory Magnifiers

1 e4 c6 2 Nc3 d5 (3 Nf3 dxe4 4 Nxe4 Nf6 5 Qe2 Na6 6 d4 Qd5 7 Nc3 Qa5
8 Ne5)

26A: 3 Qf3

26B: 5 Nxf6+

26C: 6 Nxf6+

26D: 8 Ne5

None of these magnifiers demand deep study or much memorization but

you should decide whether you like my gambit recommendation in 26A
before you meet it over the board. Similarly you should find out how you
feel about
the positions where Black recaptures on f6 with his g-pawn in 26C.

26A: Goldman’s 3 Qf3

(1 e4 c6 2 Nc3 d5)

3 Qf3 (D)

Black to play

Unsurprisingly, this is not such a good move, although it has been played
by some strong players. I have decided to fight it with aggressive means.
After all, this is how you punish your opponent for developing his queen
on the third move. Against the rare move-order 2 Qf3, Black’s simplest
solution is 2...d5 3 Nc3 with a transposition, even if 2...e5 may be even
better.

3...e5!?

If you don’t like sacrificing pawns with Black as early as move three, I
recommend 3...d4 as a good alternative. The ‘critical’ line goes 4 Bc4 e6
5 Nce2 c5 6 d3 Nc6 and now 7 Qg3 to free f3-square for the knight and
possibly aim to advance the f-pawn. Here I like 7...h5!? 8 h4 (after 8
Nf3?! h4! White may regret his early queen development) 8...Nf6 (the
immediate 8...b5!? may be more accurate, since then 9 Bxb5?? isn’t an
option at all), when White can no longer play f4 due to the weakness on
g4. After 9 Nf3 Black can, if he feels like, spice things up considerably
with 9...b5!?:

a) 10 Bxb5 Qa5+ 11 Nc3 Bd7 12 Bxc6 Bxc6 and suddenly a draw was
agreed in Pinchon-Packroff, corr. 2014. After 13 b4! Qxb4 14 Rb1 Qa5 15
Rb8+ Rxb8 16 Qxb8+ Qd8 chances are equal.

b) 10 Bb3 c4 (10...Ng4 is the move for the faint-hearted, after which Black
is doing OK) 11 dxc4 Nxe4 12 Qf4 f5 and thanks to the threat of playing
13...Bd6, White has to be very careful. 13 Ne5 Bb4+ 14 c3 dxc3 15 bxc3
Qf6 and in this complicated situation, Black has no reason to complain.

We now return to 3...e5!? (D):

White to play
4 exd5

4 Qg3 Nf6 (after 4...f6 Black’s set-up is not only beautifully symmetrical, but
it’s also better for Black according to the engines; it’s not easy to see how
White can successfully attack Black’s big centre) 5 Qxe5+ (Nikolovski-
Andonovski, Skopje 2018) and now 5...Be7 gives Black a lot of
development for the pawn.

4...Nf6 5 dxc6

5 Bc4 is probably best met by 5...e4, when Black is still playing a sound
gambit. After 6 Qe2 cxd5 7 Bxd5, rather than 7...Bc5? 8 Nxe4! +– as in
Brynell-Sta.Hansen, Brøndby 1985, I suggest 7...Nxd5 8 Qb5+ Nc6 9
Qxd5 Qxd5 10 Nxd5 Bd6 (Höxter-Pagenkopf, corr. 2013), when with the
bishop- pair and easy development, Black has good compensation.

5...Nxc6

While this position has only occurred rarely in practice, Black has achieved
a good score from it.

6 Bb5 Bd7 (D)

White to play
7 Nge2

Continuing to break the principles of sound opening play with 7 Nd5?! is

not recommended: 7...Nd4 8 Nxf6+ gxf6 9 Bxd7+ Qxd7 10 Qe4 Bg7
(10...Rg8!?) 11 c3 f5 12 Qb1 Nc6 gives Black more than enough
compensation. After 13 d3 0-0-0 14 Bg5 f6 15 Bd2 Black was too eager to
regain the pawn in Ardeleanu-Zajić, Paraćin 2017 and took on d3,
whereupon the game fizzled out to a draw. A move like 15...Rhg8 –/+
should be preferred.

7...Be7

7...Bc5?! 8 0-0 0-0 9 d3 Nb4 10 Bg5 left White better in Vachier-Lagrave

– Admiraal, Internet rapid 2018.

8 0-0 0-0

Black is ahead in development and enjoys sufficient compensation for

the pawn.

9 Bc4

Or:

a) 9 d3 Nb4 10 Bxd7 Qxd7 and Black regains the pawn with a level position.

b) 9 Ne4 Nxe4 10 Qxe4 a6 and once again, Black has good

compensation – especially after 11 Bxc6 Bxc6 12 Qxe5 Bf6, when
Black’s bishops are extremely powerful.

9...e4

Now:

a) 10 Nxe4 Ne5 11 Nxf6+ Bxf6 12 Qb3 Nxc4 13 Qxc4 Rc8 gives Black
good compensation.

b) 10 Qg3 Nb4 11 Bb3 Be6 12 Bxe6 fxe6 13 Qh3 Nxc2 14 Qxe6+ Kh8
15 Rb1 Bd6 is messy but approximately level.
26B: 5 Nxf6+ Without an Early d4

(1 e4 c6 2 Nc3 d5)

3 Nf3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6 (D)

White to play
6 Bc4

Except for 6 d4 (see Lesson 1), this is the only move that tests
Black. Harmless alternatives include:

a) 6 Be2, although played occasionally, isn’t a serious attempt for an

opening advantage. Sooner or later White should go for d4, which would
take us to positions discussed all the way back in Lesson 1; e.g., 6...Bd6 7
0-0 0-0 and now:

a1) 8 b3 Re8 9 Bb2 Nd7 10 d4 Nf8 11 Re1 (C.Kulkarni-J.Cuartas, Barbera

del Valles 2009) and here 11...Bc7 has the idea of lining up with ...Qd6. I
like Black’s kingside attacking chances.
a2) After 8 Re1 Qc7 White has run out of useful waiting moves, and 9
d4 transposes to the authors’ one and only encounter in this line,
Sv.Johnsen- T.R.Hansen, Gausdal 2000.

b) 6 g3 Bd6 7 Qe2+ (7 Bg2 0-0 8 0-0 Re8 9 d4 Bf5 10 Be3 Nd7 11 c4 Be4 12
Nd2 Bxg2 13 Kxg2 f5 14 Nf3 Nf6 15 Bg5 h6 16 Bxf6 Qxf6 17 Qd3 Re4 18
Rfe1 Rae8 =+ Przybylski-Roiz, Berlin blitz 2015) 7...Be7 8 Bg2 0-0 9 0-
0 Na6 10 Re1 Bd6 11 d4 Be6 12 c4 Re8 13 b3 Qd7 14 Bb2 Bb4 =
Kokarev- D.Howell, Riyadh blitz 2017.

We now return to 6 Bc4 (D):

Black to play
6...Bd6

This is very similar to the Forgacs Variation, and also here we can go for
the most combative continuation. Instead 6...Na6 may become quite
similar to the lines in 2A. But with Nf3 instead of d4, there is no queen
check on a5 after 7 Bxa6, so White can induce some structural damage if
he is willing to give up his light-squared bishop. 6...Qe7+ is also quite
playable and was tested in the epic battle Sv.Johnsen-T.R.Hansen,
Gausdal 2000.
7 Qe2+

For a ‘Win with’ repertoire, this move is quite challenging, as it seeks

an endgame where Black’s winning chances are limited.

7...Be7

As soon as White plays d4 we are back in the lines covered in the lesson
about the Forgacs Variation. There is no way for White to play
successfully without this move.

8 0-0 0-0 9 Re1

This position has been reached a few times, and with the white pieces we
find players like Adams and Shirov. In all these games, Black played
9...Bd6. I believe 9...Re8 is just as strong, and I don’t see any promising
options for White as 10 d4 still takes us back to the Forgacs line (2C, to be
precise).

26C: 6 Nxf6+

(1 e4 c6 2 Nc3 d5 3 Nf3 dxe4 4 Nxe4 Nf6)

5 Qe2 Na6 6 Nxf6+ gxf6 (D)

 White to play

White might claim he benefits from the fact that Black had to recapture
with the g-pawn. The position is reminiscent of the Larsen Variation,
which in its traditional form has a dubious reputation nowadays. Here,
however, White had to pay a price by placing his queen on e2.

7 d4

Alternatively:

a) You can probably blame it on being a blitz game, but in Sankalp-Oparin,

Internet blitz 2020 White played the extremely modest 7 h3?!. After 7...Bf5
it quickly got even worse: 8 d4?? Nb4 (this is one of the main ideas behind
5...Na6). White is already lost, and indeed resigned after 9 g4 Nxc2+ 10
Kd1 Bg6.

b) With the queen on e2, it seems to make sense to fianchetto the bishop
with 7 g3?. However, Black’s response shows why the line starting with
5...Na6 is powerful: after 7...Nb4! the c-pawn turns out to be a huge
problem in White’s camp. For instance, 8 d3 (8 d4?? is again a blunder as
after 8...Bf5 Black is winning) 8...Qd5 9 Bg2 Bg4 and Black threatens to
take on c2, overloading
the queen. The only way to save the pawn is the sad move 10 Kd1, which
was played in Hammer-Kariakin, Internet blitz 2020. With the white king on
d1, Black obviously has a pleasant choice. 10...e5 is strong, and so is
Kariakin’s 10...0-0-0. After 11 Bd2?! e5 12 a3?! he retreated the knight to
a6, although 12...Nxd3 looks stronger. 13 cxd3 (not 13 Qxd3? e4 –+) offers
Black a number of attractive continuations, including 13...Qxd3 14 Qxd3
Rxd3 15 Kc2 Rxf3 16 Bxf3 Bxf3, when he is for choice.

7...Qd5 (D)

White to play

8 c3

Or:

a) After 8 h3 Bf5 9 c3 (Hrebenshchykova-Rakhmangulova, Ukrainian

Women’s Ch, Lutsk 2019) Black should play 9...0-0-0; e.g., 10 Be3 Nc7 11
Nh4 Bg6 after which 12 Nxg6 hxg6 takes us to a pawn-structure I
remember was discussed in an old book by Larsen called Solide Åbninger
(‘Solid Openings’). I am pretty sure he would have argued that Black is for
choice in this position.
b) 8 g3 is better here than on the previous move. Best seems 8...Nb4 9 c3
Nc2+ 10 Qxc2 Qxf3 11 Rg1 Qh5 12 h4 Bg4 with simplification. The
engine gives White a tiny plus, but I wouldn’t mind playing Black.

c) Another possibility is 8 c4 Qh5 9 Be3 Bg4 10 0-0-0 0-0-0 11 h3 and

we have a choice:

c1) 11...Bxf3 is solid. After 12 Qxf3 Qxf3 13 gxf3 e6 the position is

strategically complex and roughly equal.

c2) 11...Rg8 tempts White to sacrifice the exchange. It’s hard for White to
untangle the kingside, and the most promising I can find is 12 hxg4 Qxh1
13 d5 Bh6. You can spend hours analysing this position, but I believe
Black is doing fine. Once again the engine gives ‘triple zero’.

8...Bg4 9 h3 Bxf3 10 Qxf3 Qxf3 11 gxf3 Nc7

Now 12 Bd2 is enough for equality. Instead, after 12 Bf4 Nd5, Black was
slightly better in Holtman-Jablonicky, Lvov 2018, and White eventually paid
the price for his damaged structure.

26D: Main Line with 8 Ne5

(1 e4 c6 2 Nc3 d5 3 Nf3 dxe4 4 Nxe4 Nf6 5 Qe2 Na6)

6 d4 Qd5 7 Nc3 Qa5 8 Ne5 (D)

 Black to play

This move has twice been played by strong players and is White’s
critical choice.

8...Nb4 9 a3

9 g3 is best met by 9...g6 10 Bg2 Bg7 11 0-0 0-0. Here White’s best seems
to be 12 Nd1 Nd7 13 c3 Qa6 14 Qxa6 Nxa6, when he is slightly more
active but Black’s position is solid. I believe that, with some patience, Black
will slowly equalize.

9...Be6 10 Rb1 Nbd5 11 b4

11 Bd2 was tried in G.Papp-Pechac, Slovakian Team Ch 2018/19. Here I

suggest the cold-blooded 11...g6. It seems risky to allow a discovered
knight jump, but it appears to work out fine for Black. After 12 Ne4 Qa4! 13
c4 Qc2 14 Qd3 Qxd3 15 Bxd3 Nxe4 16 Bxe4 Nf6 the position is equal.

11...Qb6

The threat to the d4-pawn is annoying for White. To fight for an advantage,
White can sacrifice the pawn:

12 Na4!?

12 Nxd5 cxd5 13 Be3 g6 =.

12...Qxd4 13 Bb2 Qe4 14 Nc5 Qxe2+ 15 Bxe2 Bc8 16 Bf3 e6

Now:

a) 17 c4 is interesting. Then 17...Nb6 leads to a highly complex situation;

e.g., 18 0-0 Bd6 19 Nxc6 bxc6 20 Bxc6+ Ke7 21 Bxa8 Nxa8 when it’s
very complicated, and the chances are level.

b) 17 0-0 and here 17...Bd6 allowed White an edge in Ragger-Nisipeanu,

Austrian Team Ch 2018/19, so I suggest drastic measures: 17...Bxc5 18
bxc5 0-0. Black is intending the curious knight tour ...Nc7 and ...Na6 to
attack the c5-pawn. Also the idea of playing ...Nfd5 followed by ...f6 and
...e5 is on the agenda. Chances are balanced.

Lesson 27: Two Knights Main Line: 3...Bg4

By: TRH

I can comprehend Alekhine’s combinations well enough; but where he gets

his attacking chances from and how he infuses such life into the very
opening
– that is beyond me.

Rudolf Spielmann

1 e4 c6 2 Nc3 d5 3 Nf3 Bg4

This solid developing move is very clearly Black’s most popular reaction
to the Two Knights. Now both players have a few chances to deviate but
a critical position occurs after the following moves:

4 h3 Bxf3 5 Qxf3 (D)

 Black to play

The evaluation of this position is central in this lesson. On the surface it

seems quite attractive for White, who has the bishop-pair and somewhat
the better development.

Q: This may be solid for Black but why would I want a position that must
be at least slightly better for White?

I assume you refer to White’s two bishops. They may indeed be a long-
term asset but Black is active and his pawn-structure is rock-solid. He has
a dark- squared bishop, and will therefore place his pawns on the light
squares. The resulting positions tend to be of a closed nature, and Black’s
knight is no worse than either of White’s bishops. If White can patiently
open the position under the right circumstances, he might end up better.
However, we shall see that Black is able to create sufficient counterplay –
often taking advantage of White’s exposed queen – and the positions tend
to end up strategically complicated. For instance, after 5...e6, the pawn
sacrifice 6 d4!? may be White’s most principled try. Then 6...Nf6 7 Bd3
dxe4 8 Nxe4 Qxd4, picking up a pawn and forcing White to prove
compensation, is also a way to play for a win.
 Model Game 27

V. Soloviev – Petrosian

Moscow Ch 1950

1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 (D)

White to play
4 h3

White immediately puts the question to Black’s bishop. After 4 d4 e6, 5 h3

Bxf3 6 Qxf3 transposes to the game, while for 5 Bd3 Nf6 6 0-0, see 27A.
There you will also find some analysis of the quite common but
theoretically harmless 4 Be2.

4...Bxf3

4...Bh5 is a fighting option, but in my opinion it’s under a dark cloud after
5 exd5 cxd5 6 g4 Bg6 7 d4 followed by a quick Ne5 and h4-h5.
5 Qxf3

5 gxf3 is an invention by Tal and tried in blitz games more recently by

Grishchuk and Nepomniashchy. White is hoping to strengthen the centre
and attack on the g-file. There is no need to worry – Black found the
antidote in 1960: 5...e6 6 d4 and now the flexible 6...Nd7, waiting to see
White’s set-up:

a) 7 Qd3 g6 (a key move when playing against doubled f-pawns and a

half- open g-file: Black wants a stable position where White’s
weaknesses will count in the long run) 8 Be3 Bg7 9 0-0-0 Ne7 10 Kb1
was played in Grishchuk-Mamedyarov, Beijing blitz 2013, and here I
suggest 10...b5 with the clear intention of starting an attack on the white
king.

b) 7 Bf4 Ngf6 8 Qd3 Nh5 9 Be3 g6 is a typical reaction by Black. In

Nepomniashchy-Wojtaszek, Beijing blitz 2014, the g6-pawn killed
White’s hopes of a g-file attack, while the knight on h5 prevented White
from advancing his h-pawn.

5...e6 (D)

White to play
Having exchanged his light-squared bishop, Black’s centre pawns belong
on light squares.

6 d4

This is White’s most critical continuation. However, there are alternatives:

a) For the traditional main line, 6 d3, see 27C.

b) The flexible 6 Be2 is covered in 27D.

c) The ambitious 6 g4 can be met by 6...h5!. The details can be found in 27B.

d) The quiet 6 g3 shouldn’t worry Black, but you will find some analysis of
this too, along with some other rare 6th moves, in 27B.

6...Nf6

I have struggled against this move with the white pieces, which makes it
logical to start playing it with Black. Instead 6...dxe4 7 Qxe4 (the gambit 7
Nxe4 is possible too) 7...Nf6 8 Qh4 has occurred in a few of my white
games, and I believe in White’s chances here.

7 Bd3

White’s main alternative is Hammer’s recommendation on Chess24, 7

exd5 cxd5 8 Bd3 Nc6 9 Ne2, which is analysed in 27E.

Some minor alternatives:

a) 7 Be3 Bb4 8 Bd3 dxe4 9 Bxe4 Nxe4 10 Qxe4 gives Black no problems.

b) 7 g3 dxe4 8 Nxe4 Qxd4 is better for Black.

c) 7 e5 Nfd7 scores excellently for Black. It clearly resembles a French

Defence where Black has exchanged his bad bishop, and White has
problems securing the d4-square. Now play usually continues 8 Qg3 (there
are a few other tries, but against all White’s sensible moves, Black
continues ...c5 with a good game) 8...Qb6 (practically winning the d-pawn,
and White has to
prove compensation) 9 f4 Qxd4 10 Be3 Qb4 11 0-0-0 g6 12 h4 h5 13 Be2
Qa5 and in Djurić-Groszpeter, Venice 2003, the players agreed to a draw,
but Black seems to be better.

d) After 7 Bg5, the most common reply is 7...Be7, but I believe Black has
a very simple equalizer in 7...h6 (D):

White to play

d1) 8 Bd2 Bb4 9 e5 Bxc3 10 bxc3 Ne4 11 Qg4 Kf8 12 Bd3 Nxd2 13 Kxd2 c5
= K.Honfi-Bilek, Budapest 1963. Compared to a French, Black does not
have a bad light-squared bishop.

d2) 8 Bh4 allows Black to play a good McCutcheon French with 8...Bb4,
as in Piskun-Kharitonov, Moscow 2011. White’s problem is that 9 e5?! g5
10 Bg3 Ne4 followed by ...Qa5 is much better for Black.

d3) 8 Bxf6 Qxf6 9 Qxf6 (9 Qe3 Bb4 10 e5 Qe7 11 a3 Bxc3+ 12 bxc3 c5 =

T.Christensen-J.Richardson, Copenhagen 1997) 9...gxf6 is not the most
lively position I have seen, but the game is equal and Black eventually won
in Garcia Jimenez-Cruz Lledo, Xativa 2016.
7...dxe4 8 Nxe4 Qxd4 (D)

White to play

Grabbing this pawn obviously involves some risk, not only because of the
tempo-losses involved but also because of the opening of the position. On
the other hand there are also long-term risks in playing passive positions
without any obvious compensation.

9 Be3

Or:

a) 9 Bg5?! leaves White without sufficient compensation after 9...Nbd7 10

0- 0-0 Be7.

b) 9 c3 Qd8 10 0-0 Nxe4 11 Bxe4 Bd6 12 Rd1 Qc7 13 Qd3 Be5 14 Bxh7
Nd7 15 Be4 0-0-0 16 Qf3 Nf6 17 Bg5 Qa5 is approximately equal, but
after 18 Bxf6?! Bxf6 Black ended up winning in Yilmazyerli-Svane,
European Team Ch, Iraklion 2017.

c) 9 0-0 Nxe4 10 Bxe4 Nd7 11 Bf4 and White prevailed after the weak
11...Rc8? in A.Liang-Lenderman, Saint Louis 2017. The dangers Black
faces are also shown by the fact that 11...Be7?! 12 Rad1 Qb6? 13 Rxd7!
Kxd7 14 Be3 Qxb2 15 c3! gives White a decisive attack. I prefer 11...Qf6,
and if 12 Rad1, then 12...Rd8 with a solid game.

9...Bb4+! (D)

White to play

Forcing White’s king to stay in the centre.

10 Ke2 Qd8 11 Qg3

Or:

a) 11 Rad1 Nbd7 12 Qg3 Nxe4 13 Bxe4 g6 14 c3 Be7 15 Rd2 Qa5 and

Black was better in Luu Tran-Ngo Duc, Vietnamese Under-15 Ch, Hai
Phong 2018.

b) 11 Nxf6+ Qxf6 12 Qg4 Bd6 13 Rhe1?! h5 14 Qc4 Nd7 and Black had
at least equalized in Melnikov-Kagansky, St Petersburg 2006.

c) 11 Rhd1 Nbd7 12 Qg3 Nh5 (if you don’t want a draw, you can keep the
fight going with 12...Nxe4 13 Bxe4 g6) 13 Qf3 Nhf6 14 Qg3 Nh5 15 Qf3
Nhf6 ½-½ Hernaez Fernandez-Canizares Cuadra, corr. 2008.

11...Nxe4

Rather than 11...Bf8?!, as in Lugovskoi-Yuffa, Sochi 2020, when 12

Nxf6+! gxf6 (12...Qxf6?? 13 Qc7 +–) 13 Rhd1 gives White a strong attack.

12 Qxg7

After 12 Bxe4 0-0 13 Rad1 Qe7 White has the usual compensation,
but nothing more.

12...Qf6 13 Qg4 Bc5 14 Qxe4 Bxe3 15 Qxe3 Nd7

The position is roughly equal, but soon Petrosian started to take over.

16 Rhd1 Nb6 17 c3 Nd5 18 Qe4?!

18 Qf3 would have kept equal chances.

18...Rg8 19 Kf1 Rg5 20 Re1? 0-0-0 21 Qh4 Rdg8

White can’t avoid material losses.

22 Be4 Rxg2 23 Qxf6 Rg1+ 24 Ke2 Rxe1+ 25 Rxe1 Nxf6

Black is a pawn up and won easily.

26 Bf3 Nd5 27 Be4 h6 28 Kf3 Kd7 29 Bc2 f5 30 Rd1 Ke7 31 Bb3 Rd8
32 Bc2 Nb6 33 Re1 Kf6 34 Re2 e5 35 Bb3 Rd3+ 36 Kg2 Nd5 37 Kh2
Rxh3+ 38 Kxh3 Nf4+ 39 Kh4 Nxe2 40 Kh5 Nf4+ 0-1

Theory Magnifiers

1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 (4 h3 Bxf3 5 Qxf3 e6 6 d4 Nf6 7 exd5)

27A: (4 d4)
27B: (6 g3)

27C: 6 d3

27D: 6 Be2

27E: 7 exd5

Black’s system with ...Bg4, ...Bxf3 and ...e6 is extremely solid, so nothing
too bad should happen if you do your calculations and show common
sense. Therefore it may be a good idea to gain a feel for the positions in
online blitz and informal games before spending much time on the
magnifiers. When you feel ready for some study, I would recommend
starting with 27A and 27E.

27A: 4 d4 / 4 Be2

(1 e4 c6 2 Nc3 d5 3 Nf3 Bg4)

4 d4

Despite being played relatively often, 4 Be2 should not pose Black
any problems after 4...e6 (D):
 White to play

a) 5 0-0 Nf6 6 e5 Nfd7 7 d4 Be7 8 Ne1 Bxe2 9 Nxe2 c5 10 c3 Nc6 11 Nd3

0- 0 12 Be3 was played in Muñoz Pantoja-Sanchez Sebastian, Barcelona
2019. Black has a good French, and here I believe 12...c4 13 Ndf4 b5
gives Black the upper hand.

b) 5 d4 Nf6 and now:

b1) After 6 h3 a safe choice is to take on f3, but the alternative is more
ambitious (and probably better): 6...Bh5 7 e5 Nfd7 8 g4 Bg6 9 h4 has
been seen in many games. Black has played 9...h5 in a number of
games, but I favour a rarer move: 9...h6 10 h5 Bh7 11 g5 hxg5
(D.Sebastian-Pitl, Bundesliga 2019/20) 12 Nxg5 c5 and Black begins his
counterplay and is aiming to castle queenside. The position is probably
equal, and I would be happy with the black pieces. White will struggle to
find a safe place for his king.

b2) 6 e5 is a speciality of Miguel Muñoz Pantoja. 6...Nfd7 7 0-0 (7 Ng5

Bxe2 8 Qxe2 Be7 9 Qh5 g6 10 Qh6 Bf8 = Annaberdiev-Kuzubov, Turkish
Team Ch, Kocaeli 2015) 7...Be7 8 Ne1 Bxe2 9 Nxe2 c5 10 c3 has occurred
ten
times, with Muñoz Pantoja as White in nine of them. The position
resembles a French, where Black has exchanged his light-squared
bishop. Black should be doing fine after 10...Nc6:

b21) 11 Nd3 0-0 12 Be3 was tried by Muñoz Pantoja in a couple of

games. Here I suggest 12...c4 13 Ndf4 b5, when White obviously has
some kingside attacking chances, but Black’s queenside play arrives
quickly. Chances are equal.

b22) 11 Be3 Qb6 12 Qd2 cxd4 13 cxd4 Bb4 14 Qc2 Bxe1 15 Rfxe1 0-0 16
Qd2 Rac8 17 Rac1 Nb4 18 Rxc8 Rxc8 19 Rc1 Rxc1+ 20 Nxc1 Nc6 21 b3
Qb5 and in this equal position, Muñoz Pantoja-Kovchan, Palma de
Mallorca 2015 was agreed drawn.

4...e6 (D)

White to play

Black sets up a strong pawn-centre on the light squares while keeping all
his development options.

5 Bd3
This developing move has been played in a number of games, but Black
can easily solve all his opening problems. After 5 h3 we can transpose
back to the main lines (i.e. Model Game 27 and 27E) via 5...Bxf3 6 Qxf3.

5...Nf6 6 0-0 Be7 7 Re1 0-0 8 h3 Bxf3

8...Bh5 is also possible, but Black risks getting a cramped game.

Besides, accepting the pawn sacrifice is a far greater challenge to
White’s opening play.

9 Qxf3 dxe4 10 Nxe4 Qxd4

White has compensation thanks to his bishop-pair. However, Black has

no weaknesses and should be able to neutralize White’s initiative.

11 Rd1

After 11 c3 Qd8 12 Bf4 Nbd7 13 Rad1 Nd5 14 Bg3 Qb6 Black is

slightly better.

11...Nxe4 12 Bxe4 Qf6 13 Be3

Not 13 Qb3? Na6 14 Qxb7 Nc5 15 Qb4 Qh4, when White is in trouble.

13...Qxf3 14 Bxf3 Rd8 15 Rxd8+ Bxd8 16 Rd1 Bc7

Black wants to play ...Kf8 and ...Ke8 to control the d7-square, and eventually
...Na6 or ...Bb6. Chances are equal.

27B: Rare 6th Moves

(1 e4 c6 2 Nc3 d5 3 Nf3 Bg4)

4 h3 Bxf3 5 Qxf3 e6 (D)

 White to play

6 g3

Or:

a) The ambitious 6 g4 can be met by 6...h5! (or 6...Bc5 7 Bg2 h5), and now:

a1) 7 gxh5 Nf6 8 d4 dxe4 9 Nxe4 Qxd4 gives White

insufficient compensation.

a2) 7 Bg2 hxg4 8 hxg4 Rxh1+ 9 Bxh1 and now 9...d4 (to prevent d4) 10
Ne2 Ne7 is probably equal, but Black has potential on the dark squares.

b) 6 Qg3 is a speciality of McShane and Bartel. The former tested the

move in the 2019 World Cup. It makes sense for Black to step away from
the e5 advance with 6...Ne7. Additionally, Black intends ...g6 and ...Bg7,
making White’s queen look a bit misplaced. Then:

b1) 7 exd5 cxd5 8 Nb5 Nf5 9 Qc7 Qd7 10 Bd3 (Z.Bašagić-Mazej,

Slovenian Ch, Dobrna 2002) 10...Nc6 gives Black a good game.
b2) 7 d3 Nd7 8 Be2 g6 9 0-0 Bg7 10 Kh1 0-0 11 f4 and in McShane-
Rodshtein, Batumi Olympiad 2018 Black played the unnecessary
11...Kh8?!, and White won a long game. However, Black’s position was
actually very good as it’s hard for White to continue the attack. He should
prefer 11...b5 as 12 f5? simply doesn’t work due to 12...exf5 13 exf5 Nxf5
14 Rxf5 gxf5 15 Bh6 Qf6 –+. 12 Bf3 is better, but Black is already more
comfortable.

c) The slow 6 a3 is aimed at playing d3 without being disturbed by ...Bb4.

However, this is unproblematic for us. Strong players have used it, and in
the 2019 FIDE World Cup McShane played it in a dramatic match against
Yuffa. I recommend 6...Nd7 (D) as I like to keep the kingside development
flexible, whereas I already know this is the best square for the queen’s
knight.

White to play
Then:

c1) 7 d4 is possible as with the knight on d7, the black queen is no longer
indirectly attacking d4. Now 7...Qb6 leads to a position that has only been
tested 7 times, with Black scoring 5½ points. 8 Qd3 was played twice
against Asrian. The talented Armenian passed away much too early at the
age of 28.
He successfully defended Black’s position against De Vreugt and
Grachev with 8...g6. However, I believe Black can do even better with
8...Rd8, when Black is piling up on the d4-pawn and White is practically
forced to make a decision in the centre. The position after 9 exd5 (9 e5?!
c5 looks like a good French for Black) 9...cxd5 has never occurred, but I
believe Black should experience no problems if he continues with ...Ngf6,
...Bd6 and ...0-0.

c2) 7 d3 is the consistent follow-up but with 7...Qb6 we pin down

White’s bishop to the b2-square. This position is little tested but Black
seems fine:

c21) 8 Be2 Bd6 9 0-0 Ne7 10 Na4 Qc7 11 d4 dxe4 12 Qxe4 0-0 = Milde-
Schreurs, corr. 2016.

c22) 8 g3 is the main line, but here I suggest the new move 8...h5. This is
not a start of an attack, but the beginning of a dark-square strategy. After,
e.g., 9 Bg2?! d4 10 Nb1 (after 10 Ne2 Qa5+ 11 Bd2?! Qc5 Black is
suddenly threatening both to take on c2 and to play ...Ne5) 10...h4 11 g4
Ne7 12 0-0 e5 Black is better. Imagine moves like ...Ng6 and ...Nc5-e6.

6...Nf6 7 Bg2 Bc5 (D)

White to play
The main idea is to prevent d4. Now:

a) 8 0-0 0-0 9 d3 Nbd7 transposes to note ‘b’ to White’s 7th move in 27C.

b) 8 e5 Nfd7 9 Qg4 Nxe5 10 Qxg7 Ng6 11 Ne2 (Van der Wiel-

Dautov, Venlo tt 2000) 11...h5 threatens to trap the queen by ...e5
and ...Bf8.

27C: 6 d3

(1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 4 h3 Bxf3 5 Qxf3 e6)

6 d3 (D)

This is closely related to 6 g3 (27B) and may easily transpose.

Black to play
6...Nd7

This is the most flexible. 6...Nf6 is also possible but I prefer to delay
committing this knight, waiting to see how White continues his
development. If White goes for queenside castling, a knight on f6 will be
exposed to g4-g5,
while this will not be the case with kingside castling.

7 Bd2

White’s logical follow-up is to go for queenside castling, but I believe this

is mostly dangerous for White. Other moves:

a) 7 Be2 is possible but combining d3 and Be2 is not testing for Black;
e.g., 7...g6 (I don’t see any problems with 7...d4 8 Nb1 Bd6 either; for
instance, 9 0-0 Ne7 10 Nd2 0-0 with a complex but equal position) 8 Qg3
Bg7 9 exd5?! cxd5 10 Bf4?! (Mamedov-Ponomariov, Hengshui blitz 2019)
10...Ne7 –/+.

b) 7 g3 is relatively popular. By fianchettoing his light-squared bishop, it’s

clear that White aims to castle kingside, which is why we play 7...Ngf6.
After 8 Bg2 Bc5 9 0-0 0-0 (D) we have reached a topical position. Now:

White to play

b1) 10 g4?! could easily backfire on White. Rather than 10...b5?!

(Maltsevskaya-Giria, Russian Women’s Team Ch, Moscow 2019), I
suggest 10...Ne5 11 Qg3 Ng6, when White’s problem is that f4 is not
legal: 12 Kh1 (12 h4? is consistent, but too committal: 12...Bd6 13 f4 h5
14 g5 Ng4 and
next up is 15...f5 –/+) 12...dxe4 13 dxe4 e5 =+.

b2) The position after 10 Qe2 has occurred in many games, but here I
suggest 10...a5 to gain space on the queenside, and to meet Na4 by
...Ba7. 11 Kh1 transposes to line ‘b3’, and it’s not obvious how else White
should continue.

b3) 10 Kh1 a5 11 Qe2 was played in J.van Foreest-Tari, Internet blitz

2019. Here Black decided to exchange on e4, but I see no reason to do
this. Instead Black should play 11...a4 (D), grabbing more space on the
queenside, while White has to be prepared to meet ...a3.

White to play

The move ...d4 may also be on Black’s agenda, as the queenside

belongs to him. However, White’s plan is to play f4 and it’s important that
we know how to meet this:

b31) 12 e5 Ne8 13 f4 f6 is unproblematic for Black. If White takes on f6,

Black will recapture with the queen. Best is probably 14 Bd2 fxe5 15
fxe5 Nc7 16 Rxf8+ Qxf8 17 Rf1 Qe7 =.

b32) 12 f4 Ne8 doesn’t allow White to play both f5 and e5 (we can allow one
of these moves, but not both):

b321) 13 f5?! d4 14 Nd1?! (after 14 fxe6 fxe6 15 Rxf8+ Bxf8 followed by

...Bd6, Black is better due to White’s weakened kingside) 14...exf5 15
exf5 Nef6 and Black is ready to play ...Bd6 and ...Re8, resulting in a
better position for Black.

b322) 13 g4 a3 14 b3 and now giving White a free hand on the kingside

with 14...d4 may look wrong, but Black is actually doing fine; e.g., 15 Nd1
e5 16 f5 (16 g5?! exf4 17 Bxf4 Qe7 is positionally promising for Black as
...Bd6 and ...f6 will give White problems on the dark squares) 16...Be7 and
White has problems arranging g5, while Black is ready to advance on the
queenside and eventually play ...c5-c4. Everything depends on whether
White can break through. I believe Black is better.

We now return to 7 Bd2 (D):

Black to play

7...Qb6

The idea is to force White’s hand.

8 0-0-0

Black has many different set-ups in positions like this. I suggest a rare but,
in my opinion, very poisonous continuation:

8...d4 9 Ne2

9 Nb1?! Qa6 10 a3 b5 looks very dangerous for White.

9...a5

One of the advantages of Black’s position is that the knight is still on g8.
This means that White can’t gain time with g4-g5, and I believe Black’s
attack is quicker; e.g.:

10 Qg3 a4 11 f4

White has to create counterplay on the kingside, but there is a problem

that the queen is in the way. After 11 a3?! c5 Black will quickly play
...Qa6 followed by ...b5.

11...Ne7 12 Qf3 c5 13 g4 Nc6 (D)

 White to play

This is a typical position in this variation. White’s problem is that his

attack has reached a standstill, and the position must be treated with
great care:

a) 14 f5? loses to 14...c4 15 dxc4 Nde5.

b) 14 Bg2? c4! 15 dxc4 Nc5 and Black is close to winning. For instance,
16 Kb1 a3 17 b3 d3 –+.

c) 14 a3? Qa6 15 e5 b5 16 Bg2 Rc8 and I don’t see how White can
stop Black’s attack.

d) After 14 h4?! a3 15 b3 c4 16 dxc4 Nc5 Black is threatening ...d3 and

it’s hard to prevent; e.g., 17 Ng3 d3 18 Bxd3 Nd4! gives Black a huge
attack.

e) As a matter of fact, 14 Kb1 is one of White’s few playable moves. Here

I suggest the Najdorf-inspired 14...g5 to win control of the e5-square. The
following moves show how easily this can go wrong for White: 15 fxg5?
c4 16 dxc4 Nde5 17 Qg3 Nxc4 –+.

f) 14 e5 is best met by 14...g5 15 Bg2 gxf4 16 Nxf4 (16 Qxf4 Bg7 17

Rdf1?! Rf8 =+) 16...Ndxe5 17 Qe2 Bd6, when White has compensation for
the pawn, but after castling queenside Black should be OK.

27D: 6 Be2

(1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 4 h3 Bxf3 5 Qxf3 e6)

6 Be2

With this move, White wants to develop the kingside before pushing d4.

6...Bc5 (D)
 White to play

I believe this is an accurate move as it takes control of the d4-square.

7 0-0

7 Qg3 is best met by 7...Bd4 8 exd5 exd5 9 0-0 Ne7 10 Nd1 Nd7 11 c3
Be5 12 f4 Bf6 13 d4 0-0 14 Nf2 c5. White won in Ibragimov-Dreev,
Russian Team Ch, Sochi 2012, but Black is not worse at this point.

7...Nd7 8 Qg3

Or:

a) 8 exd5 has been played in a couple of games, and in both Black

recaptured with the e-pawn. I believe 8...cxd5 is better, as White has two
problems in this position. First of all, d4 needs preparation, and secondly,
the knight is misplaced on c3. Then:

a1) 9 Rd1 Ngf6 10 d4 Bd6 is equal, thanks to White’s knight on c3. A

possible line is 11 Nb5 Bb8 12 c4 a6 13 Nc3 Qc7 and Black picks up
the pawn on c4, although White gets some compensation.
a2) 9 Na4 Bd6 10 d4 Rc8 11 c4 dxc4 12 Qxb7 Ne7 is interesting
and probably equal.

b) 8 d3 Ne7 is comfortable for Black:

b1) 9 Bd1 0-0 10 Ne2 dxe4 11 dxe4 Ng6 12 Nf4 a5 13 Nxg6?! fxg6!? 14
Qg3 Qb6 15 Bf3?! Rf7 16 c3 Raf8 –/+ Gabrielsen-Šulskis, Barcelona 2007.

b2) 9 Re1 0-0 10 Bf1 Ng6 11 Bd2 (Veszpremi-Prohaszka, Balatonlelle

2009) 11...Nde5 =+.

b3) 9 Rd1 Bd4 10 Nb1 0-0 = Leko-Timman, Dortmund 1994.

b4) 9 Qg3 Ng6 10 Bg5 Qb8 11 Qxb8+ Rxb8 = A.Kozlov-Chigaev, Minsk

blitz 2015.

8...Bd4

In many of these lines the dark-squared bishop plays an important role.

9 d3

9 exd5 exd5 (9...cxd5?! 10 Nb5 Be5 11 f4 is annoying for Black) 10 Qg4

(10 Bg4 Ne7 11 Re1 and here, instead of 11...Bf6?! as in E.Berg-Dreev,
Internet blitz 2004, I suggest 11...0-0 =, when Black has nothing to worry
about) 10...Qf6 11 d3 h5 12 Qf4 Qxf4 13 Bxf4 0-0-0 14 Rfe1 Bf6 15 Bd2
Ne7 16
Nd1 Nf5 17 c3 (Harikrishna-Praggnanandhaa, Kolkata blitz 2018) 17...d4 =+.

9...Ne7 (D)
 White to play

The bishop on d4 is annoying for White.

10 Bd1

With this move, White makes room for Ne2. Some alternatives:

a) 10 Nd1 Qb8 11 Bf4 and here the position got unnecessarily messy after
11...e5?! in Chernyshov-Dreev, Russian Team Ch, St Petersburg 1999.
Safer and better is 11...Be5 12 Bxe5 Qxe5 13 Qxe5 Nxe5 =.

b) 10 Kh1 0-0 11 f4 was met by 11...f5 in Lane-Bjelobrk, Australian Ch,

Melbourne 2014, but Black need not weaken his kingside. After all, White
is not threatening to advance. After, say, 11...a5 12 f5?! exf5 13 exf5 Re8
Black is positionally better, and I don’t see how White’s attack breaks
through.

c) 10 Be3 Bxe3 11 fxe3 was tested in Walek-Splichal, Czech Team Ch

1994/5. In all these lines I think Black should appreciate the solidity of
his position. Here Black should go for 11...0-0 with no problems. For
instance, 12 d4 Rc8 13 Bd3 c5 =.
10...0-0 11 Ne2 Bb6 12 c3 e5

Black is starting to become ambitious, and intends ...f5. McShane-

Yuffa, Khanty-Mansiisk 2019 continued 13 exd5 cxd5 14 d4 and here I
believe 14...e4 is best. I don’t see why Black should be worse in this
complicated position. For instance, 15 f3 Nf6 16 fxe4 Nxe4 17 Qd3
Qd7 =.

27E: Hammer’s Recommendation

(1 e4 c6 2 Nc3 d5 3 Nf3 Bg4 4 h3 Bxf3 5 Qxf3 e6)

6 d4 Nf6 7 exd5

This move is more dangerous than it looks, and it has recently been my
choice with the white pieces. Also with Jon Ludvig Hammer recently
recommending it in his 1 e4 repertoire on Chess24, we need to take this
line seriously.

7...cxd5 (D)

Other recaptures give White an advantage.

 White to play

Black has freed c6 for the natural development of his queen’s knight.

8 Bd3 Nc6 9 Ne2

White aims to play c3 and achieve a more harmonious set-up for the
pieces. At this point, I believe Black should play to restrict White’s dark-
squared bishop.

9...h6

In the lines given by Hammer, Black often ended up in slightly passive

positions. One of the reasons was that Black allowed White to play Bg5,
and ended up playing ...Be7. With this move we are aiming for a more
active set- up, and to prevent Bg5 altogether. 9...Nb4 should be avoided
due to 10 Bb5+.

10 c3

Alternatively 10 0-0 Bd6 11 c3 transposes to the main line, while 10 Bf4

Bd6 should be of no concern for Black, as White will no longer have the
bishop- pair.

10...Bd6 (D)

Even though this position has only arisen a few times in practice, I have
taken it quite seriously as I suspect more players might go for this line.
 White to play

11 0-0

Or:

a) White could try the cunning 11 Bc2, when Black has to be careful. White
makes a semi-waiting move, keeping the option of attacking with g4.
11...e5 (I believe this direct move is the best; 11...0-0?! is strongly met by
12 g4, when Black may regret combining ...h6 and castling before White
has committed his king) 12 dxe5 Nxe5 13 Ba4+ (13 Qf5 0-0 14 0-0 is fine
for Black, as White has problems with his exposed queen) 13...Kf8 (at first
sight this looks wrong, but White’s pieces are not well placed, and this
gives Black time to coordinate) and after, for instance, 14 Qe3 g6 15 0-0
Kg7 with moves like ...Nc4 and ...Qc7, or even initiating a minority attack
with ...a6 and ...b5, Black’s activity should compensate for White’s bishop-
pair.

b) 11 Be3 is similar to the previous line. 11...e5 (11...0-0?! 12 g4 is still

dangerous for Black) 12 dxe5 Nxe5 13 Bb5+ (13 Qf5?! 0-0 turns out to
be awkward for White, especially as 14 Bc2?? g6 traps the queen)
13...Kf8 14 Qf5 a6 15 Ba4 g6 16 Qc2 Kg7 and like in many of these lines
with castling
by hand, Black’s activity provides enough compensation for the bishop-pair.

c) 11 Ng3 e5 12 dxe5 Nxe5 13 Bb5+ Kf8 14 Qd1 Qb6 15 Be2 g6 16 0-0 Kg7
17 Rb1 Rae8 18 Be3 Bc5 19 Bxc5 Qxc5 = Ghinda-Dorfman, Polanica
Zdroj 1978.

d) 11 g4 e5 12 Qg2 exd4 13 cxd4 Qa5+ 14 Bd2 Bb4 and I believe Black is

likely to be slightly better, as it’s hard to find time to bring the king to safety
and prove the correctness of g4. For instance, after 15 Nc3 0-0 16 g5?!
Rae8+ 17 Kf1 Ne4 18 Nxe4 dxe4 19 gxh6 g6 I think White’s aggression is
about to backfire.

e) 11 h4 e5 12 dxe5 Nxe5 13 Bb5+ Kf8 14 Qh3 Qb6 and ...Neg4 may

cause White to regret the weakening of the g4-square. Black is not
worse.

We now return to 11 0-0 (D):

Black to play
11...0-0

As soon as White castles, we can safely do the same.

12 Nf4

Or:

a) 12 Ng3 Re8 13 Nh5 transposes to the main line.

b) 12 Bf4 e5 13 dxe5 Nxe5 14 Bxe5 Bxe5 is equal. White can still go for
g4, but this time it’s riskier.

c) 12 g4 e5 13 Qg2 is of course very ambitious and may backfire. Black

has a few options, but I like 13...Bc7 14 g5 hxg5 15 Bxg5 Qd6, when the
pressure on the b8-h2 diagonal limits White. The position is sharp, but with
roughly equal chances.

12...Re8

Black is preparing ...e5.

13 Nh5 Nxh5 14 Qxh5 e5

Also 14...Qf6 can be considered; for instance, 15 Bd2 e5 16 dxe5 Nxe5

17 Bc2 Bc5 18 Rad1 Rad8 and with ...d4 coming, the position looks
equal.

15 dxe5 Nxe5

Both 15...Rxe5 and 15...Bxe5 are valid alternatives.

16 Bc2 Re6

The point of Black’s play is to ask White how he plans to develop

his queenside without being hit by ...Nc4.

17 Be3 Nc4

Two possibilities now are:

a) 18 Bd4 Be5 19 Rad1 Qd6 20 b3 Na3 21 Bd3 Bxd4 22 cxd4 a6 followed by

...Nb5 with an equal game.
b) 18 Rad1 Re5 19 Qf3 Nxe3 20 fxe3 Qe7 and we have a complex
middlegame with opposite-coloured bishops and major pieces, a
situation which is known to favour the attacker. Here White is slightly
better mobilized, but the weakening of the dark squares will give Black
good play in the long run.

Lesson 28: Steiner Variation (2 c4)

By: SJ

The only thing chess-players have in common is chess.

Lodewijk Prins

1 e4 c6 2 c4 (D)

This move is fourth in popularity (after 2 d4, 2 Nc3 and 2 Nf3) but more
remarkably the second-best scoring move (after 2 Ne2) at almost 58%.
White usually steers for positions similar to those we have seen in the
Panov lessons but with certain extra options.
 Black to play
2...d5

While 2...e5 is a relatively common alternative, it will usually lead to an

Old Indian, which currently isn’t too popular.

Anyway, this is the natural move for a Caro-Kann player – after all, this is
the advance we prepared with our modest-looking first move.

3 exd5

I don’t know why this is more than twice as popular as 3 cxd5 cxd5 4
exd5, which leads to the same position.

3...cxd5 4 cxd5 Nf6 5 Nc3 (D)

Black to play

White has an isolated d-pawn (for the moment it’s actually two), just like
in the Panov Attack. There are plenty of transpositional options but
mostly White will keep his d2-pawn back for a while and it may well end
up at d3
rather than d4.

5...Nxd5

An alternative is to construct a repertoire consistent with the Panov Attack

lines of Lessons 21 to 23 by delaying the recapture of the pawn with 5...g6.
However, this leads to some murky positions, which is why I prefer the
more straightforward text-move. If you do wish to investigate this option,
then note that after 6 Bc4 Bg7 7 Nf3 0-0 8 0-0 Black should avoid
8...Nbd7?!, as 9 d3 Nb6 10 Qb3 is very difficult for Black, with 10...Bg4 met
by 11 Ne5. Instead 8...Na6! (with ...Nc7 ideas) is the move to examine, and
indeed this complex tactical line looks very playable for Black if he is well-
prepared.

Now (after 5...Nxd5) we have to be rather careful to avoid ending up in a

Panov position outside our prepared repertoire. The main thing to
remember is to avoid an early ...Nc6 – unless you achieve something
concrete – and instead focus on rapid kingside development.

6 Nf3

In the Panov (Lesson 21 to be precise), we saw that this was generally too
slow to achieve an advantage. Here we shall see that by delaying d4
White is just in time to generate an initiative.

6...g6!?

Although rare, this move scores quite decently for Black. If you don’t like it,
we offer an almost untested idea in the 6...Nxc3 line in 28D.

Model Game 28

Illescas – Kamsky

Manila Interzonal 1990

1 e4

1 c4 c6 2 e4 is another important move-order.

1...c6 2 c4 d5 3 exd5

Ideas other than an ‘accelerated Panov’ are possible but White

shouldn’t expect any advantage from the opening with them:

a) 3 Nc3?! d4 4 Nce2 e5 5 d3 is probably already better for Black. I

would start with 5...Bb4+, forcing the exchange of light-squared bishops
and then normal development with ...Nf6 and ...0-0, but Black has a lot
of freedom.

b) 3 e5 dxc4 4 Bxc4 Qd4!? 5 Qe2 Bg4 6 f3 Bf5 7 e6 Bxe6 8 Bxe6 fxe6

9 Qxe6?! Na6 10 Ne2 Qh4+ =+ Short-D.Howell, London rapid 2013.

3...cxd5 4 cxd5

4 d4 immediately transposes to the Panov Attack (Lessons 21 to 24).

4...Nf6 (D)

White to play
5 Nc3
This is clearly the main move, but the two checks are important attempts
to keep the advanced d-pawn. Of the two, 5 Bb5+ is the more common
but 5...Nbd7 6 Nc3 a6 7 Qa4 Rb8 seems a straightforward path to an
interesting position. Black will usually sacrifice a pawn for the initiative
with ...e6. For the details, see 28A.

The slightly less popular 5 Qa4+ leads to more diverse positions. After
5...Nbd7 6 Nc3 g6, White’s most popular move is 7 Nf3. White’s 7th-move
alternatives range from the weird-looking 7 g4 to the popular 7 Bc4. For
the details of all these lines, see 28B.

5...Nxd5 6 Nf3

This flexible developing move is probably the hardest to meet. An assault

on f7 isn’t very dangerous as long as Black hasn’t played ...g6. The direct
attacking moves 6 Bc4 and 6 Qb3, plus some even rarer 6th moves, can
be found in 28C.

6...g6!? (D)

This is a relatively rare move, with 6...Nc6 much more common. However,
as we know from the Panov lessons, the fianchetto is the ideal set-up
against White’s isolated d-pawn. So the big question here is whether it’s
tactically sound. I believe it is, and Stockfish has no real objections. If you
don’t trust the analysis, the back-up line 6...Nxc3 is analysed in 28D.
 White to play
7 Qb3

This is probably White’s only serious try for an opening plus. Other
moves seem less critical:

a) 7 Bb5+ Bd7 8 Bc4 Nb6 9 Bb3 transposes to note ‘b’ to White’s 6th
move in 28C.

b) 7 Bc4 Nb6 (7...Nxc3?! 8 Qb3 +=) 8 Bb3 Bg7 and now 9 d4 takes us
back to Model Game 22, but the lightly tested 9 a4!? could be worth a
closer look. After 9...a5, the additional a-pawn moves shouldn’t change
the overall assessment too much but it’s possible that permanent access
to b4 could help Black control the important d5-square.

7...Nb6

White’s point, as we know from the Panov lessons, is that 7...Nxc3?! 8 Bc4
is better for White (8...e6?! 9 Qxc3 +/–).

8 Bb5+
8 d4 transposes to a rather toothless Panov line (see the note to White’s
8th move in 22B).

8...Bd7

8...N6d7?! looks very risky after 9 Ne5.

9 Ne5

The point; thanks to the delay of d4, White is just in time to play this
attacking move with some effect. This forces a weakening of the
dark squares. Other moves shouldn’t worry Black:

a) 9 a4 seems rather pointless as long as the b6-knight can safely go via

c8 to d6 if White’s a-pawn advances further.

b) 9 d4 looks too sedate. After 9...Bg7, Black is ready to parry any attack
on f7 with ...0-0, keeping his pawn-structure flexible.

9...e6 (D)

White to play
Has White achieved anything of substance in this position (compared to the
Panov lines from Lessons 21 to 23)? Black has had to play an early ...e6,
which hems in his light-squared bishop and weakens some dark squares
on the kingside. On the other hand we are ready to drive back White’s
advanced forces by threatening to exchange and get closer to an endgame
where White’s isolated queen’s pawn is a genuine weakness. I believe
Black is close to equality in an interesting position.

10 Ne4

At this point, 10 d4 is an important alternative. After 10...Bg7 11 Ne4 0-0 12

Bg5 Qc7, 13 Rc1 presents Black with some challenges. Instead I suggest
10...Bc6!?. At first it seems strange to move a bishop that is half-way
deployed a second time before completing development. However, the
bishop is well placed on c6 where it attacks g2, and Black is only a few
moves away from his ideal set-up:

a) 11 Be3 Bg7 12 0-0 0-0 gives White a slightly passive IQP position.

b) 11 Bxc6+ Nxc6 12 Nxc6 bxc6 13 0-0 Bg7 isn’t great for White, who is left
with a weak bishop and a d-pawn that is easier to attack than Black’s c-
pawn.

c) 11 Nxc6 Nxc6 12 0-0 Bg7 transposes to line ‘d’.

d) 11 0-0 Bg7 12 Nxc6 (12 Qa3 Bf8 13 Qb3 Bg7 is a reasonable but far
from forced repetition) 12...Nxc6 13 d5 Nxd5 14 Rd1 (14 Nxd5 Qxd5 15
Qxd5 exd5 =+) 14...0-0 15 Nxd5 exd5 16 Rxd5 Qf6 =.

10...Be7

This isn’t the bishop’s ideal diagonal, but that can be rectified once Black
has consolidated. Right now getting the king into safety has top priority.

11 d4 (D)
 Black to play
11...Bxb5?
!

It seems Black should try a different move here, and there is a choice:

a) 11...0-0!? is fine according to Stockfish. The tricky lines with Bh6 all work
out OK for Black, but to play like this with confidence would require some
computer preparation.

b) I like 11...Bc6!? because it forces some clarification of the position so

Black doesn’t have to worry about his king so much: 12 Be3 Qd5 13
Nxc6 Nxc6 14 Qxd5 Nxd5 = or 12 Nxc6 Nxc6 13 Be3 Rc8 14 Rc1 0-0
15 Bxc6 Rxc6 16 Rxc6 bxc6 17 0-0 Qd5 =.

12 Qxb5+ N8d7

After the pawn sacrifice 12...Nc6!? 13 Nxc6 bxc6 14 Qxc6+ Qd7 15 Qxd7+
Kxd7 it’s hard to say whether Black’s play on the queenside files is
sufficient for equality.

13 Bh6 a6 14 Qe2 Bb4+?!

After 14...Nxe5 15 dxe5 Qd4 16 Bg7 Rg8 17 Bf6, White has a moderate plus.

15 Kf1

Now White’s advantage is considerable.

15...Nd5??

After 15...Be7 16 h4, White has a dangerous initiative, but there doesn’t
seem to be anything forced.

16 Nxf7!

You don’t often see this kind of early tactical shot in games between
top GMs.

16...Kxf7 17 Ng5+

White wins back the piece immediately and is now winning trivially.

17...Ke8 18 Qxe6+ Qe7 19 Qxd5 Kd8 20 Ne6+ Kc8 21 Rc1+ Bc5 22 Bg5
Qf7 23 Qd6 Nb8 24 Rxc5+ Nc6 25 Be7 1-0

Theory Magnifiers

1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6 (5 Nc3 Nxd5 6 Nf3 Nxc3)

28A: 5 Bb5+

28B: 5 Qa4+

28C: (6 Bc4)

28D: 6...Nxc3

28B and 28C are quite important as the positions under discussion are
rather sharp. You should have a good look at the bolded moves and any
sidelines that look scary. 28D is an additional option for Black and thus
not required
reading unless you intend to use it, either as your main choice or as a
surprise weapon.

28A: 5 Bb5+

(1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6)

5 Bb5+ Nbd7

I believe this is objectively stronger than covering with the bishop.

6 Nc3 a6 (D)

This leaves White less choice than 6...g6, which is also very playable.

White to play
7 Qa4

This somewhat committal pin is clearly White’s most popular try. This
position could also arise via the move-order 4...a6 5 Nc3 Nf6 6 Qa4+ Nbd7
7 Bb5.
Other ideas:

a) Surprisingly, Bologan doesn’t mention 7 Ba4, though after 7...b5 8 Bc2

Bb7 it’s White who must be careful: 9 Nge2 (9 Nf3 Nxd5 10 0-0 g6 11 Re1
Bg7 12 d4 0-0 13 Bg5 N7f6 14 Ne4 Nb4 =+ Golubović-V.Georgiev,
Cannes 1996) 9...Nb6 10 Nf4 Nbxd5 11 Nfxd5 Bxd5 12 0-0 e6 13 d4 Be7
14 Nxd5 Qxd5 15 Re1 0-0 = Looshnikov-Iankowski, Internet 2006.

b) 7 Be2 b5 8 d4 b4 9 Na4 Bb7 10 Nf3 Bxd5 11 Bf4 e6 12 Rc1 Be7 13 0-0 0-

0 and then:

b1) 14 Bc7 Qc8 15 Bd6 and now 15...Qd8 16 Bc7 Qc8 17 Bg3 Qd8 was a
quick repetition in Varga-Blohberger, Dunajska Streda 2018. If Black
wants to play on in an equal position, 15...Qe8 is possible.

b2) After 14 b3, as in Ponomariov-Gelfand, Medias 2010, Bologan

recommends 14...Rc8! 15 Bxa6 Rxc1 with excellent chances for Black: 16
Qxc1 Bxf3 17 gxf3 Qa8 18 Be2 Nd5 =+ or 16 Bxc1 Nb6 17 Nxb6 Qxb6
=+.

c) 7 Bxd7+ Qxd7! (D) and now:

White to play
c1) After 8 Qb3 b5 9 Nf3 Bb7 (9...Qd6!?) 10 Ne5 Qd6 11 Nc4 bxc4!?
(11...Qc5, as in Formanek-Khenkin, Antwerp 1995, can lead to
wonderful complications but overall Black’s task seems the harder) 12
Qxb7 Rb8 13 Qa7 g6 looks fine for Black; e.g., 14 Qd4 Bg7 15 Qxc4 0-0
16 0-0 Rfc8 17 Qd3 Rd8 =.

c2) 8 Nf3 Nxd5 9 0-0 Nxc3 10 bxc3 e6 and now:

c21) 11 d4 Bd6 12 Ng5 and in Ashley-Kacheishvili, New York 1998 it would

have been interesting to see the natural 12...0-0!?, planning ...b5 and
...Bb7 with harmonious play. After the normal-looking moves 13 a4 b6 14
Re1 h6 15 Ne4 Be7 16 Rb1, Black may develop with 16...Bb7, as 17 Rxb6
Qc7 18 Rb1 Bxe4 19 Rxe4 Qxc3 seems good for Black.

c22) 11 Ne5 Qc7 12 d4 Bd6 13 Ba3 Bxa3 14 Qa4+ b5 15 Qxa3 Bb7 16 c4?!
(16 Rfe1 =) 16...f6 17 Nd3 bxc4 18 Nc5?! Bd5 –/+ Pranav-
S.Kasparov, Mumbai 2010.

c23) 11 a4 Bd6 12 Ba3 0-0 13 Qb3 Rb8 14 Qb6?! Bc7 15 Qc5 Re8 16 Ne5
Qd8 17 Rfe1?! f6 18 Nf3 Bd7 =+ Berkovich-Erenburg, Israeli Team
Ch 2011/12.

We now return to 7 Qa4 (D):

 Black to play
7...Rb8!

This is after all a somewhat marginal line and it would be nice to deal with
it in a forcing way that somewhat reduces the memorization required.

8 Bxd7+

8 Be2 b5 offers Black interesting play:

a) 9 Qd4?! g6 10 Bf3?! Bg7 11 Nge2 0-0 12 Qf4?! b4 13 Nd1 Bb7 –/+

Schelfhout-Euwe, Amsterdam 1933.

b) 9 Qh4 (Kiik – Mäki-Uuro, Salo 1991) 9...g5!? 10 Qxg5?! Rg8 11

Qh4 Rxg2 =+.

8...Bxd7 (D)

Recapturing with the queen is somewhat more popular and also

seems perfectly fine.
 White to play

9 Qf4

Or 9 Qb3 b5, and now:

a) 10 Nf3 b4 is no problem for Black:

a1) Against 11 Nd1, Bologan recommends 11...Rb5! 12 Ne3 Nxd5 13

Nxd5 Be6 but it’s not clear to me why he considers the position equal.
Black has a sounder structure, a small lead in development and the two
bishops. I believe Stockfish’s evaluation of ‘-0.88’ is spot-on, give or take a
few centipawns.

a2) 11 Ne2 Bb5 12 Nf4 Qd6 13 d3 Rd8 14 0-0 Nxd5 15 Nxd5 Qxd5 16 Qxb4
e6 = Chiburdanidze-K.Lehmann, Biel 1991.

b) After 10 d4 b4 11 Bf4?!, several strong players have chosen 11...Rc8,

which seems good for Black, but perhaps 11...e5!? is even stronger: 12
dxe6 (12 Bxe5? bxc3 13 Bxb8 cxb2 14 Qxb2 Nxd5 and Black wins, as 15
a3 is met by 15...Bb5) 12...Bxe6 13 d5 bxc3 14 Qxb8 (or 14 Bxb8? Qa5
15 Kf1 Bxd5 –+) 14...Qxb8 15 Bxb8 cxb2 16 Rb1 Bxd5 17 Nf3?! Bb4+ 18
Kd1 0-0
–/+.
c) 10 d3 e6!? leads to interesting play:

c1) 11 dxe6 Bxe6 12 Qd1 Bb4 gives Black sufficient compensation; e.g., 13
Ne2 0-0 14 0-0 Re8 15 Be3 Qd7 16 Rc1 Nd5 with rough equality since
White must avoid 17 Qd2?! Rbd8 18 d4? Nxe3 19 fxe3 Bxa2 –+.

c2) 11 Bg5 exd5 12 Nf3 (12 Nxd5? Be6 –/+) 12...Be7 13 0-0 0-0 14 Nxd5
Nxd5 15 Qxd5 Bxg5 16 Qxg5 Rb6

=. We now return to 9 Qf4 (D):

Black to play
9...b5

The immediate 9...e6 seems equally playable (and not terribly

different). Some high-level examples:

a) 10 Nf3 exd5 11 0-0 Be6 is an idea worth noting (and possibly more
active than 11...Be7). Chandler-Speelman, Hastings 1986/7 was a quick
draw after 12 d4 Bd6 13 Qh4 Rc8 14 Ng5 0-0 15 Nxe6 fxe6 16 Qh3 ½-½.
b) 10 dxe6 Bxe6 11 Nge2 Bd6 12 Qh4 0-0 13 0-0 Re8 14 d3 Ng4 15
Qxd8 Rbxd8 16 Bf4 (Bologan-Astolfi, Nîmes 1991) and now 16...Bc5 is
the easy path to equality, as 17 Rad1?! Bxa2 is better for Black and 17
d4 Bxd4 is dead equal.

10 a3 e6 11 Nf3

Or:

a) 11 d3 b4 12 axb4 Rxb4 13 Qg3 exd5 14 Nf3 Be7 15 0-0 0-0 =.

b) 11 dxe6 Bxe6 12 Nge2 Bd6 13 Qh4 0-0 14 0-0 b4 15 axb4 Rxb4 16 d4

Bc4 17 Re1 (Hatzidakis-Milonakis, Khania 1997) and Black should
probably take the time for 17...h6, preventing Bg5, with equal chances.

11...exd5 12 0-0 Be7 (D)

Based on the similar position without the moves ...b5 and a3 (note to
Black’s 9th move), one may wish to investigate 12...Be6!?, which looks
rather equal after 13 d4 Bd6 14 Qh4 Rc8.

White to play
With a symmetrical pawn-structure and Black having the bishop-pair and
somewhat easier development, you might think that Black is better.
However, with fixed isolated d-pawns, White’s knight may well be better
than Black’s light-squared bishop. Overall, chances seem equal in this
position, which hasn’t been tested much at GM level unless you count
Romero-Seirawan, Wijk aan Zee 1992, which ended in a draw after 13 d3
0-0 14 Be3 b4.
Perhaps 13 d4 0-0 is a more natural continuation, from where you may
wing it or carefully prepare the relatively quiet lines after 14 Re1 or 14
Ne5.

28B: 5 Qa4+

(1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6)

5 Qa4+ Nbd7 (D)

White to play

This may look a bit clumsy but it actually goes well together with a
kingside fianchetto.

6 Nc3
6 Bc4 g6 7 Nf3 Bg7 8 d3 0-0 9 Nc3 is just one of several other ways to
reach the main-line position of the section.

6...g6 7 Nf3

This natural knight development is the most common. Several other

moves are worth a somewhat detailed look:

a) For 7 Qb3 Bg7 8 Nf3, see the note to White’s 8th move below.

b) 7 Nge2 Bg7 8 h4?! 0-0 9 Nf4 Re8 10 Be2 Nc5 11 Qd1 e5 is given by
Bologan. After 12 dxe6 Nxe6 13 Nxe6 Bxe6 Black seems better
following both 14 0-0 Qd4 and 14 d4 Nd5 15 Bg5 Qb6.

c) 7 Bc4 Bg7 8 d3 (8 Nf3 transposes to the main line of this section) 8...0-0
9 Nge2?! (for 9 Nf3, see the main line) 9...a6 10 Qa3 b6 11 0-0 Bb7 is very
comfortable for Black:

c1) After 12 d6?! e6 13 d4 b5 14 Bd3 e5 15 d5 Rc8 16 Ng3, as in Pile-

Svetushkin, Cap d’Agde rapid 2013, Black can try 16...h6!?, planning
...Ne8 with a clear plus.

c2) 12 Bg5 b5 13 Bb3 (Doncea-Kirszenberg, Lyons 2007) and again

13...h6 makes sense, this time in order to make White’s bishop and knight
fight over the f4-square. 14 Be3 Nb6 15 d6 exd6 leaves Black better.

c3) 12 Nf4 b5 13 Bb3 Nb6 14 Re1 Bxd5 (14...Re8!?) 15 Qxe7 Bxb3 16 Qxd8
Rfxd8 17 axb3 Nfd5 18 Nfxd5 Nxd5 gave Black just a cosmetic advantage
in Scherbin-Velička, Pardubice 2008.

d) 7 d4?! Bg7 and now:

d1) 8 Qb3 0-0 will mostly transpose but 9 Bg5!? Nb6 10 Bxf6 is an
independent idea. After 10...Bxf6 11 Nf3, the typical pawn sacrifice
11...e6 seems strong:

d11) 12 dxe6?! Bxe6 13 Qd1?! Rc8! 14 Be2 Nc4 leaves Black better: 15
Rb1 (15 b3? Qa5 –+) 15...Nxb2 –/+.
d12) 12 d6 Nd5 13 Ne4 Qa5+ 14 Ned2 Rd8 15 Bd3 Rxd6 =+
Tischbierek- Böhnisch, Leipzig 1979.

d2) 8 Be2 0-0 9 Bf3 Nb6 10 Qb3 Bf5 11 Nge2 Bd3 12 d6 exd6 13 0-0 Re8 14
Rd1 Bc4 15 Qc2 Qd7 16 Be3 Rac8 = Künzler-Mittelman, Berlin

1998. d3) 8 Nh3 0-0 9 Nf4 Nb6 10 Qb3 and now:

d31) 10...g5!? looks crazy but after 11 Nd3 Nfxd5 12 Bxg5 Bxd4 13 Be2
Nxc3 14 bxc3 Bg7 Black’s position seems safe enough and his centre
pawns could be the basis for future central dominance.

d32) 10...a5 11 a4 Qd6 12 Be2 Rd8 13 0-0?! (13 Bf3 Qb4 14 Qd1 Nfxd5 15
Nxd5 Nxd5 16 Bd2 Nxc3 17 bxc3 Qc4 18 Be2 Qc6 19 0-0 e5 =)
13...Nfxd5 14 Nfxd5 Nxd5 15 Bc4 Be6 =+ Khader-Ivanchuk, Riyadh rapid
2017.

e) The primitive-looking 7 h4?! shouldn’t be a problem:

e1) 7...h5 is a safe solution: 8 d4 Bg7 9 g3 0-0 10 Bg2 Nb6 11 Qb3 Bf5 12
Nf3 and rather than 12...Bd3?! 13 Ne5 +/– Lein-Tarasov, USSR 1967,
Black should play 12...Nfxd5 13 0-0 Nxc3 14 bxc3 Be6 =+.

e2) 7...Bg7!? is more challenging but seems good for Black:

e21) 8 Be2 h5 9 Nh3 0-0 10 Nf4 a6 11 d4 b5 12 Qb3 Nb6 13 0-0 Qd6 14 Bf3
(Miezis-Hartikainen, Finnish Team Ch 2010/11) 14...Bg4 =+.

e22) 8 h5 Nxh5 9 g4 Nf6 10 g5 (after 10 Be2?, the move-order actually

used in Tadić-Šćekić, simply unpinning with 10...Kf8! denies White any
target and leaves him in deep trouble given the weaknesses he has just
created to no effect) 10...Nh5 11 Be2 0-0! and from this position 12 Bxh5
gxh5 13 Rxh5 Qb6 14 Qe4?! Qg6?! (14...f5!) 15 Rh4?! Ne5 –/+ was played
in Tadić- Šćekić, Senta 2002.

f) 7 g4 (D) is a very crude attacking attempt that achieves little:

 Black to play

7...Bg7 8 g5 Nh5 9 Nge2 0-0 10 Ng3 e6 (10...Nb6!?) 11 Nxh5?! (11 Qh4 f6

12 d4 e5 13 Nxh5 gxh5 14 Ne4 exd4 with the point 15 Rg1? Ne5 –/+, but
15 Be2 is equal) 11...gxh5 12 Qh4 (after 12 d4 Nb6 {12...e5!?} 13 Qd1
exd5 14 Qxh5, as in Korchnoi-Salov, Amsterdam 1991, Black has the
noteworthy option 14...Re8+ 15 Be3 Nc4, based on the defensive idea 16
0-0-0?! Nxe3 17 fxe3 Rxe3 18 Bd3? Rh3 –+) 12...Ne5 and now:

f1) After 13 Bg2?, as in V.Gurevich-B.Rõtov, USSR Army Ch 1966, Black

has 13...Nd3+ 14 Ke2 exd5! 15 Kxd3 Re8 16 Qf4 Be5 17 Qf3 Bg4 18
Qxd5 Bf5+ 19 Ke2 Bxc3+, winning.

f2) 13 Be2 Ng6 14 Qxh5 exd5 =+.

g) 7 g3 Bg7 8 Bg2 0-0 (D) and now:

 White to play

g1) 9 d4 Nb6 10 Qb3 a5 11 a4 Bf5 12 Nge2 Bd3 13 d6 exd6 and after 14

Bxb7?! Rb8 15 Bf3 (G.West-R.Smith, Auckland 2004) Black’s best is
probably 15...Qc8; e.g., 16 0-0 Qf5 17 Bg2 Rfe8 18 Qd1 Bc2 19 Qd2 Nc4
20 Qf4 Qc8 =+.

g2) 9 Nge2 and now 9...e6!?, as advocated by Bologan, gives Black the
initiative. He offers the lines 10 dxe6? Nc5 –/+ (as 11 exf7+? Rxf7 12
Qc4 Nd3+ 13 Kf1 Qb6 14 Qxd3 Ng4 is outright losing for White) and 10
Nf4?! Nb6 11 Qb3 exd5 =+. Still, the cautious 10 0-0 should avoid any
real disadvantage.

We now return to 7 Nf3 (D):

 Black to play
7...Bg7 8
Bc4

After 8 Qb3 0-0 9 d4?! (for 9 Bc4?!, see the next note) 9...Nb6 White is
a move down compared to a well-known theoretical line and his d4-
pawn is vulnerable. Black is probably already better.

8...0-0 9 d3

9 Qb3?! is probably too slow, and should be met by 9...b5!:

a) 10 Qxb5?! Rb8 11 Qa4 Nb6 12 Qb3 Qd6 13 Nb5 Qd7 14 Nc3 Qg4 15
Bf1 Bb7 16 Ne5 Qc8 17 Nc6 Bxc6 18 dxc6 Qxc6 19 Qb5 Qa8 –/+.

b) 10 Nxb5 Rb8 11 Qa3 (11 Qd1 Qa5 12 a4 Nb6 13 Ba2 Nfxd5 14 0-0 a6
15 Nc3 Bb7 16 Re1 Nb4 17 Bb1 Nc4 =+) 11...Nb6 12 d3 Nfxd5 =+.

c) 10 Bxb5 Rb8 and now:

c1) 11 Nd4? Ne5 12 Qa4 (12 Nc6? Nxc6 13 dxc6 a6 –+ Boidman-

Bertrem, Belgian Team Ch 2006/7) 12...Nxd5 13 Nc6 Nxc6 14 Bxc6 Nb4
15 0-0 Ba6
16 Bb5 Bxb5 17 Nxb5 Nd3 –/+.

c2) 11 Qc4 Nb6 12 Qe2 Nfxd5 13 d4 Bg4 14 0-0 Nxc3 15 bxc3 Nd5 16
Bd2 Bxf3 17 gxf3 Qc7 and Black has compensation (and probably a bit
more) for his pawn.

We now return to 9 d3 (D):

Black to play

This position can arise from a number of move-orders and is quite

well explored.

9...Nb6

This seems at least as good as the more common 9...a6. Quite

surprisingly, the lightly-tested 9...b5!? also seems playable. White is a
useful move up on the 9 Qb3?! lines but Black’s play on the open lines
seems sufficient for dynamic equality.

10 Qb3 Bg4 11 Ne5

After 11 Ng5 Rc8 12 Be3 h6 13 Nge4 Nxe4 14 Nxe4 (Jobava-Wang Yue,
Taiyuan 2006 and other games) Black’s best option is probably the
thematic pawn sacrifice 14...e6!, when Black’s compensation is illustrated
by these sample variations: 15 dxe6 (15 0-0 exd5 16 Bxb6 Qxb6 17 Bxd5
Qxb3 18
axb3 Be2 19 Rfe1 Bxd3 20 Rxa7 Rfd8 21 Bxb7 Rc2 =; 15 Bxb6 Qxb6 16
Qxb6 axb6 17 dxe6 Bxe6 18 Bxe6 fxe6 19 Nc3 Bxc3+ 20 bxc3 Rxc3 21 Kd2
Ra3 =) 15...Nxc4 16 exf7+ Rxf7 17 dxc4 Be6 18 Rc1 Qa5+ 19 Bd2 Qd8
20 Be3 (20 0-0? Bf5 –/+) 20...Qa5+ =.

11...Qc7 (D)

White to play

Black has completed development and White hasn’t got his king to safety
yet. While there are pitfalls for both sides, Black’s position is arguably the
easier to play:

a) 12 Nxg4 Nxg4 13 h3 Qe5+ 14 Ne2 Nf6 15 Bf4 Qf5 16 g4 Qd7 17 Nc3 h5!
and the chances seem equal. After 18 f3 Rad8 19 Bb5 Qc8, 20 Rc1? is
an error. In G.Jones-Izoria, European Ch, Warsaw 2005 Black then went
for 20...Nfxd5!, which gave him a big plus but required a lot of
calculation. He
had several other moves that retain an advantage, including the
prosaic 20...Qc5.

b) After 12 f4 Qc5 13 h3 Bc8 14 Ke2 Black must be very careful.

14...Nh5?!, as in A.Zhigalko-Hovhannisyan, Erevan 2016, leads to trouble
for Black after the precise 15 Re1; e.g., 15...Nxc4 16 dxc4 g5 17 Ne4 Qd4
18 Qd3 Qxd3+ 19 Nxd3 gxf4 20 Nxf4 Nxf4+ 21 Bxf4 Bxb2 22 Rab1 with a
tricky ending ahead for Black. It’s unlikely an opponent would find that over
the board, but you never know when you are going to run into a well-
prepared player! I therefore suggest 14...Qd6 15 Re1 Rd8, which seems
fairly balanced. A computer line runs 16 Nb5 Qb8 17 d6 Nxc4 18 dxe7 Re8
19 dxc4 a6 =. 14...Qc7!? 15 Re1 Nfd7 is also viable, as White can’t quite
make a tactical breakthrough on f7 work, but the play is even more
‘computery’ in that case.

28C: 6 Bc4 and Rare 6th Moves

(1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6)

5 Nc3 Nxd5 6 Bc4

This direct move is White’s main alternative to the natural

knight development. Other moves are harmless:

a) After 6 Qb3 Nb6 7 Nf3, 7...Be6! seems like a good way to get the
pieces out; e.g., 8 Bb5+ Nc6 9 Qc2 g6 10 0-0 Bg7 11 d4 0-0 12 Bxc6
bxc6 13 Be3 Qc7 14 Rfe1 Rfd8 15 Ne5?! (Garcia Jimenez-Narciso
Dublan, Catalonian Team Ch, Barcelona 2016) and now 15...c5 leaves
Black better.

b) 6 Bb5+ Bd7 7 Bc4 Nb6 8 Bb3 g6 (8...e6 9 Nf3 Bc6 10 0-0 Be7 11 d4 +=)
9 Nf3 Bg7 10 d4 0-0 11 0-0 Nc6 12 h3 Na5 13 Bc2 Bc6 14 Ne5 Rc8 =
Nijboer-Miles, Dieren 1988.

We now return to 6 Bc4 (D):

 Black to play
6...Nb6

The knight is surprisingly happy here. As you may remember from Lesson
22, it doesn’t necessarily have to return to d5 later, as there could well be a
future on c4 as well. 6...Nxc3 has the point that 7 Qb3 e6 is rather safe for
Black given that he hasn’t yet weakened his kingside with ...g6, but 7 Qf3
has scored well for White, who is developing rapidly.

7 Bb3 Nc6

Adventurous readers may wish to take a look at 7...g6!? (when 8 d4

transposes to Lesson 22), since 8 Qf3, attempting to force the weakening
8...e6 (which isn’t too bad for Black in any case) can also be answered
with 8...Bf5!?, when White will have to get involved in some very
committal and risky play (such as playing g4 and/or grabbing the b7-
pawn) if he is to hope to justify his early queen deployment.

8 Nf3 Bf5 (D)

Or:
a) Obviously not 8...Bg4?? 9 Bxf7+ Kxf7 10 Ng5+ +–.

b) 8...g6 is probably playable but in Hamdouchi-Demuth, French Team

Ch 2012, White had a plus after 9 a4 a5?! 10 d4 Bg7 11 d5 Nb4 12 0-0
0-0 13 Re1.

White to play

Black has solved all opening challenges except for one: how to play for a
win against an opponent looking for a safe draw?

9 d4

White should avoid 9 0-0?! Bd3 10 Re1 e6, when Black is at least
more comfortable.

9...e6 10 0-0 Be7 11 Be3

11 d5 exd5 12 Nxd5 Nxd5 is very drawish:

a) 13 Qxd5 Qxd5 14 Bxd5 0-0 15 Bg5 Bxg5 16 Nxg5 h6 17 Ne4

(P.Morley has played 17 Nf3 in several correspondence games but has
only succeeded
in showing that this is even more drawish) 17...Rfd8 18 Bxc6 bxc6 19
Rfe1 Bxe4 20 Rxe4 Rd2 = Seirawan-Shankland, USA Ch, Saint Louis
2011.

b) 13 Bxd5 0-0 14 Be3 Bf6 15 Rc1 Nb4 16 Bxb7 Rb8 17 Qxd8 Bxd8 18
Bc6 Nxc6 19 Rxc6 Rxb2 20 Bxa7 ½-½ Spassky-Portisch, Candidates (7),
Mexico City 1980.

11...0-0 12 Ne5

Or:

a) 12 h3 Bf6 13 Qe2 Na5 14 Bc2 Bxc2 15 Qxc2 Rc8 =+ Miezis-

Engqvist, Stockholm 2012/13.

b) 12 Re1 a6 13 Rc1 Rc8 14 h3 Na5 15 Bc2 Bxc2 16 Qxc2 Nd5 =.

c) 12 a3 Bf6 13 h3 Rc8 14 Qe2 Rc7 15 Rfd1 Rd7 16 Qd2 Qb8 =

Topalov- Morović, Las Palmas 1993.

d) It’s not too late for the equalizing 12 d5, but after 12...exd5 13 Nxd5
Nxd5 14 Qxd5 Qxd5 15 Bxd5 Nb4, White must be a little careful in view of
16 Bxb7? Rab8 17 Nd4 Bd3 –/+ J.Zyla-Evseev, Koszalin 1999. Instead 16
Bb3 is equal; e.g., 16...a5 17 a3 Nc2 18 Bxc2 Bxc2 19 Rac1 =.

12...Na5!? (D)
 White to play

Instead of simplifying, Black fights for the c-file and the c4-square.
Chances are roughly equal:

a) 13 Bc2 Bxc2 14 Qxc2 Rc8 15 Qe2 Nac4 16 Nxc4 (16 Rac1 Nxe3 17
fxe3 Bd6 =) 16...Nxc4 17 Rad1 Qb6 18 Bc1 Bf6 19 b3 Nd6 =+ Miezis-
Cruz Lledo, Charleroi 2018.

b) 13 g4 Bg6 14 f4 looks very loosening but after 14...Nxb3 15 Qxb3 Bd6

(15...Bf6!?), the consistent 16 f5! exf5 17 gxf5 Bh5 18 Ne4 should boil
down to equality. Instead Meshcheriakova-Demina, Russian Women’s Ch,
Voronezh 2009 continued 16 Rf3?! Rc8 17 Bf2?! Bxe5 18 dxe5 Nc4 19
Qxb7?, when Black could have gained a huge plus with 19...Rc7! 20 Qa6
Nxb2.

28D: 6...Nxc3

(1 e4 c6 2 c4 d5 3 cxd5 cxd5 4 exd5 Nf6 5 Nc3 Nxd5)

6 Nf3 Nxc3
This leads to play somewhat similar to the Grünfeld Defence with
black piece-play against White’s pawn-centre.

7 bxc3

7 dxc3 Qxd1+ 8 Kxd1 is attractive for White only if he considers himself a

much better endgame player than Black. Most normal moves should
equalize, including 8...g6 and 8...Nc6 intending ...Bf5.

7...g6 (D)

White to play

Can White’s c- and d-pawns neutralize Black’s dark-squared bishop, or will

they be targets? As a matter of fact, the position is quite well explored. So
far White has been scoring well but it seems Black can equalize.

8 d4

White has also tried to delay the central advance:

a) 8 h4 Bg7 9 h5 looks primitive but is a serious try. 9...Nc6 10 Rb1 (10 Bc4
Qc7 11 Kf1 was played in Smagin-Garcia Palermo, Bundesliga 1990/1,
when 11...Bf5 12 d4 0-0-0 would have been unclear) 10...Qc7 11 Ba3?! Bf5
12 Rb5 a6 13 Rc5 (A.Sokolov-Karpov, Candidates (11), Linares 1987)
13...Rd8 14 d4?! b6 –/+.

b) 8 Bc4 Bg7 and now:

b1) 9 Ng5 0-0 10 Qf3 Bf6 11 Ne4 Nc6 12 Nxf6+ exf6 13 Be2?! Re8 14 0-0
Be6 15 d4 Bd5 was agreed drawn in Miljanić-Dragašević, Montenegrin
Team Ch, Ulcinj 2014 but if anyone is better, it must be Black.

b2) 9 0-0 0-0 10 Re1 Nc6 11 h3 b6 12 a4 Bb7 13 Ba3 Bf6 14 Ba2 Re8 15 d4
e6 (not 15...Rc8?! 16 d5! followed by Ne5) is OK for Black, Reeh-
Dautov, Bundesliga 2003/4.

c) 8 Bb5+ Bd7 9 a4 Bg7 10 0-0 0-0 and then:

c1) 11 d4 Nc6 12 Re1 Rc8 13 Ba3 Re8 14 Ng5 h6 15 Ne4 (Popilski-

Macieja, Dallas 2017) and now 15...a6 16 Bf1 Na5 is a safe path to
equality.

c2) 11 Re1 a6 12 Bc4 Nc6 13 Ba3 Bf6 (13...Re8 is playable, but White
can reply 14 Bxf7+, which leads to a forced draw) 14 d4 Rc8 15 Nd2
(K.Szabo- Wojtaszek, French Team Ch 2013) 15...Na5 16 Bd5 Bf5 is
roughly level.

8...Bg7 9 Bd3

Or:

a) 9 Bc4 transposes to note ‘a1’ to Black’s 7th move in 21B.

b) 9 Be2 0-0 10 0-0 Nc6 11 Be3 (11 Bf4 b6 12 Qd2, as in Ermenkov-Yu

Ruiyuan, Albena 2015, is best met by 12...e5 =) 11...b6 12 Rc1 Na5
(12...e6!? with ...Ne7-f5 ideas) 13 Qd2 Bg4 14 Bh6 (14 c4!?) 14...Bxf3
15 Bxf3 Nc4 = Nijboer-Groszpeter, Zalaegerszeg 1991.

9...0-0 10 0-0 Nc6 11 Re1

White has also tried 11 Be4 Na5:

a) 12 Bf4 Be6 13 Re1 Rc8 14 Qd3 Bc4 15 Qe3 Re8 16 Ng5 Qd7 was fine
for Black in Lautier-Sargissian, Paris rapid 2009.

b) 12 Re1 doesn’t look dangerous either but could be a way for White to
move his pieces slowly over to the kingside while Black is busy
occupying all the juicy squares on the queenside.

We now return to 11 Re1 (D):

Black to play

If you head for it, chances seem good that you will reach this position as
Black. I don’t really see why White should be better, as Black has easy
development, no obvious weaknesses and White’s pawn-centre can easily
be attacked:

11...e5!

This move has so far been tested only in two games out of more than 150
reaching this position. It seems to equalize immediately (as some of the
more explored alternatives like 11...Bg4 and 11...Qa5 probably also do
with some care); e.g., 12 Nxe5 Nxe5 13 dxe5 Qa5 14 Bf4 with two
practical tests:
a) 14...Qxc3 15 Be4 Be6 16 Bxb7 Rab8 17 Qf3 Qxf3 18 Bxf3 Rb4 19 Bg5
Rb5 20 Bf6 Bxf6 21 exf6 (Hamdouchi-C.Aryan, Doha 2015) 21...Rf5
seems equal; e.g., 22 Bg4 Rxf6 23 Bxe6 Rxe6 24 Rxe6 fxe6 =.

b) 14...Be6 15 Be4 Rac8 (15...Qxc3 transposes to line ‘a’ above) 16

Bxb7 (Veček-A.Novak, corr. 2009) 16...Rc4 17 Qf3 Qc7 18 Ba6 Rxc3 is
equal.

Lesson 29: King’s Indian Attack

By: TRH

All you need in this life is ignorance and confidence, and then success is sure.

Mark Twain

The King’s Indian Attack is one of rather few 1 e4 openings where White
is striving for a certain formation, rather than fighting Black move by
move. If nothing particular happens, White is heading for a set-up like
this:

White’s pawn/piece formation could be referred to as either the ‘King’s

Indian Castle’ or the ‘King’s Indian Hut’, depending on how impressive you
find it. It’s fairly flexible and White may follow up with play in the centre or
on the queenside, but most frequently he expands on the kingside with
moves like Nh4, f4, h3, g4 and f5. Alternatively White may play e5 and
follow up with h4-h5 and a transfer of pieces to the kingside. However,
none of that is very likely to happen against the plan I’m going to
recommend for Black, which is a very ambitious set-up with a big pawn-
centre.

Model Game 29

Mamedyarov –

Grishchuk Huaian blitz

1 e4
2017

The King’s Indian Attack can be played with many different move-orders.
The actual first six first moves of this game were 1 g3 d5 2 Bg2 e5 3 d3
Nf6 4 Nf3 Bd6 5 Nbd2 0-0 6 e4 c6, and only now have we entered Caro-
Kann territory.

1...c6 2 d3

This seemingly modest move is part of a slow attacking set-up.

2...d5 (D)
 White to play

3 Nd2

Or:

a) 3 Be2 may also be met by 3...e5 and a set-up with ...Bd6 similar to
what we see in the game.

b) In 29A we shall have a closer look at 3 Qe2, which is specially designed

to prevent our recommended ...e5 set-up.

c) Also 3 Nf3 stops Black from playing ...e5. Players who tried this line
include J.van Foreest, Ivanchuk and Andreikin. Against this move Black
should go for the endgame: 3...dxe4 4 dxe4 Qxd1+ 5 Kxd1 and now I
believe 5...Nf6 is the most precise reply, as it immediately forces White to
defend his e-pawn:

c1) 6 e5?! is no threat as both 6...Ng4 and 6...Nd5 look better for Black.

c2) 6 Bd3 Nbd7 7 h3 Nc5 gives Black no reason to worry, Nehyba-

T.Vojta, Tatranske Zruby 2016.
c3) 6 Nfd2 Ng4 7 Ke1 e5 can’t be bad for Black, as White has lost the
right to castle. 8 Nb3?! was played in Piskur-P.Neuman, Aschach 2010
(8 Nc3 is better and should be equal). I believe Black should use the fact
that White’s king is stranded in the centre and go 8...f5, when he is not
worse, at least.

c4) 6 Nbd2 g6 7 c3 Bg7 8 Ne1 was played in J.van Foreest-Martirosyan,

Internet rapid 2019. Here I suggest 8...0-0 followed by ...b6 and ...c5,
when I believe Black is doing fine.

c5) 6 Nc3 (in general, with a black pawn on c6, the white knight is not well
placed on c3) 6...Bg4 7 h3 (7 Be2 Nbd7 8 Be3 e5 9 h3 Bxf3 10 Bxf3 Bc5
was played in A.Haast-Housieaux, Belgian Team Ch 2017/18; if anyone is
better, it’s Black) 7...Bxf3+ 8 gxf3 Nbd7 9 Be3 e6 10 Rg1 g6 is equal but
strategically complex, Rakesh-Černoušek, Marianske Lazne 2019.

3...e5

The most ambitious approach.

4 Ngf3 Bd6

There is nothing much wrong with 4...Nd7 (though 5 d4!? is a little trickier
to handle then) but as we shall see (in particular in 29C), the bishop
development allows Black some extra options with his knight.

5 g3

This is by far the most popular move, but White has the
interesting alternative 5 d4, which is analysed in 29B.

5...Nf6 6 Bg2 0-0 7 0-0 Re8 (D)

 White to play

With this useful move we are waiting to see which set-up White prefers.

8 c3

White has a number of reasonable moves here. In 29C we shall have a

closer look at White’s main alternatives, 8 Re1 and 8 b3.

8...Nbd7

Q: What makes it now clear that this is the right square for the knight?

From here the knight can reach c5 to target the d3-square, as well as
being transferred to f8 and g6 as in the game.

9 Qc2 h6

In nine games from this position, Black has won six with no draws. With
all the pieces on the board, we are guaranteed a fight.

10 Re1 Nf8 11 b3
Or 11 Nf1 Ng6 12 Ne3 (12 Bd2 Bd7 13 Rad1 Qc7 14 Ne3 Rad8 leaves
Black at least equal, Adams-Dreev, Moscow blitz 2007) 12...Be6 13 Nf5
Bf8 14 h3 Qd7 15 g4 Rad8 and with all his pieces mobilized and the
stronger centre, Black was better in Miciano-J.Christiansen, World Junior
Ch, Manavgat 2018.

11...Ng6 12 Bb2 Bg4 13 h3 Be6 14 Rad1 Qc7 15 Nf1?! Rad8 (D)

White to play

The position is more comfortable for Black, thanks to his greater space.

16 Ne3 Qc8 17 Kh2 b5

Black is intending to gain more space with ...a5.

18 a4 bxa4 19 bxa4 Qb7 20 Nf5??

After this relatively elementary blunder (it was after all a blitz game),
Black is winning. In any case, it must be said that Black’s opening was
already a success.
20...dxe4 21 dxe4 Bb3 22 Qc1 Bxd1 23 Rxd1 Bc5

White has no compensation. He now tried to get counterplay by sacrificing

a couple of pawns, but to no avail.

24 Re1 Bxf2 25 Rf1 Bc5 26 Nd2 Re6 27 Qc2 Qd7 28 Nb3 Bf8 29 Bc1 Qd3
30 Qb2 Nxe4 31 Rf3 Qd7 32 Qc2 Nf6 33 Rf2 Qd1 34 Qa2 Rb8 35 Rb2
Rxb3 36 Rxb3 Qxc1 37 Rb8 Qxc3 38 Qf2 e4 39 Qxa7 e3 0-1

Theory Magnifiers

1 e4 c6 2 d3 d5 (3 Nd2 e5 4 Ngf3 Bd6 5 g3 Nf6 6 Bg2 0-0 7 0-0 Re8 8 b3)

29A: 3 Qe2

29B: 5 d4

29C: (8 b3)

King’s Indian Attack players typically aren’t the best-prepared opponents

but frequently have a lot of experience in the resulting typical positions. We
have tried to utilize this by choosing a set-up where the early central clash
gives the better-prepared player a head start. So my recommendation is to
spend some time on these magnifiers as well as on the model game. If you
have to prioritize, I would leave 29A for last as it’s very rare.

29A: 3 Qe2

(1 e4 c6 2 d3 d5)

3 Qe2 (D)
 Black to play

This position could also arise from the move-order 2 Qe2 d5 3 d3. White’s
queen looks a bit awkward but makes it harder for Black to play ...e5 and
in certain lines his queen’s knight may find a better square than d2 (or just
wait while White develops his kingside).

3...dxe4

This seems consistent. 3...g6, hoping to prove that the queen isn’t ideally
placed on e2, is objectively fine, but the resulting positions will not have
much of a Caro-Kann feel. It’s also almost certainly possible to make
3...e5 work, but after 4 exd5, Black should avoid 4...Qxd5? 5 f4! and play
some gambit idea instead.

4 dxe4 e5

Now we have a position which is more like an Open Game (arising from 1
e4 e5) than a Caro-Kann and where White’s queen looks slightly misplaced
on e2. This is to some degree balanced by the fact that ...c6 isn’t an ideal
move either (it takes a square from Black’s knight) but if nothing else, the
pawn on e5 is protected in some lines because of the queen check on a5.
5 Nf3 Nf6

The rarer 5...Qc7 also seems fully playable; e.g., 6 Nbd2 Nf6 7 g3 (White
achieves little by 7 Nc4 Nbd7, transposing to the note to White’s 7th
move) 7...Be7 8 Bg2 0-0 9 0-0 Re8 leaves Black very solid, Mora-Hort,
Tel-Aviv
Olympiad 1964.

6 Nbd2 Nbd7 (D)

White to play

The position is relatively symmetrical but White is committed to a

kingside fianchetto and Black may use this to unbalance the position.

7 g3

Getting the light-squared bishop into play immediately looks natural.

Bauer apparently likes 7 Nc4, but after 7...Qc7 White is struggling to
equalize:

a) 8 Bd2?! a5! 9 Bc3 Bb4! 10 0-0-0? 0-0 11 Qe1 Ng4 (11...b5! 12 Ncxe5
Nc5! leaves White in deep trouble) 12 Bd2?! b5 13 Ne3 (C.Bauer-
Svane, Batumi Olympiad 2018) 13...Ngf6! and White may not even
survive much
longer.

b) 8 Bg5 Be7 (Black has ...b5 ideas) 9 0-0-0? 0-0 10 g4 Re8 11 Rg1 Nc5
12 Bxf6 Bxf6 13 Nd6 Rd8 –/+ C.Bauer-Borne, Lyons rapid 2019.

c) 8 g3 seems the only move to ensure White a playable position: 8...Bc5 9

Bg2 b5 10 Ne3 0-0 (10...Nxe4?! 11 Nf5 0-0 12 Qxe4 Nf6 13 Nh6+! is at
least OK for White) 11 Nf5 Re8 12 0-0 a5 and Black has the easier position
to play, but it should be possible for White to hang on.

7...Bc5 8 Bg2

8 Bh3 could be worth looking into but the position must be fine for Black.

8...0-0 9 0-0 (D)

Black to play

9...Re8

9...a5 may be just as good:

a) 10 a4 b6 11 Nb3 Ba6 12 c4 Be7 13 Nh4 g6 14 Bh6 Re8 =+
Maciejewski- Meduna, Katowice 1993.

b) 10 Nh4 Re8 11 Nf5 b6 12 Nb3?! Ba6 13 c4 Bf8 14 a4 (Hera-F.Vass,

Slovakian Team Ch 2005/6) and now the natural 14...Qc7 gives Black
some advantage.

10 a4 a5 11 Nb3

This is better than 11 b3?! b6 12 Bb2 Ba6 13 Nc4 Qc7 14 Bh3? b5 –

/+ M.Marković-M.Ivanov, Bački Petrovac 2018.

11...Bf8 12 Rd1 Qc7 13 Nfd2?! b6?!

13...b5! is a surprising computer suggestion, based on the b3-

knight’s vulnerability and lack of flight-squares. The threat is simply
14...Nb6.

14 Nc4

This is probably best. 14 Qe1?! was played in Azmaiparashvili-

Kotronias, Greek Team Ch, Kallithea 2003, when apart from the ‘human’
14...Ba6 15 Bf1 b5 =+, there is once again 14...b5! threatening ...Nb6.

14...Ba6 15 Be3 Rad8

Black’s position is still easier to play; e.g.:

a) The tactics following 16 Bf1 Nxe4 are slightly in Black’s favour: 17 Bxb6
Nxb6 18 Rxd8 Nxg3 19 hxg3 Qxd8 20 Nbxa5 Qc7 21 Qe4 Nd5 =+.

b) 16 f3 Nd5! 17 Bf2 Nc5 18 Bf1 Nxb3 19 cxb3 Nb4 =+.

29B: 5 d4

(1 e4 c6 2 d3 d5)

3 Nd2 e5 4 Ngf3 Bd6 5 d4

White opens the centre in an attempt to prove that the bishop is misplaced on
d6.

5...exd4 (D)

White to play
6 Bd3

6 exd5 is hardly critical, although it was played in a blitz game by Carlsen.

6...Nf6 7 dxc6 (7 Nc4?! Bb4+ 8 Bd2 Bxd2+ 9 Qxd2 cxd5 10 Nce5 0-0 11
0-
0-0 Nc6 12 Nxc6? bxc6 –/+ Carlsen-So, Internet blitz 2017, although White
won) 7...Nxc6 8 Be2 0-0 9 0-0 resembles a line of the Tarrasch French
where it’s important for White to control d4 to prevent Black’s d-pawn from
advancing. Here it has already reached d4, which means Black is doing
fine. The position after 9...Bc7 10 Nb3 Bb6 has been reached in a number
of grandmaster games, two with Tari playing Black. The position is
interesting, but probably equal.

6...Ne7 7 0-0

7 Nxd4 0-0 8 0-0 Nd7 9 exd5 Nxd5 is notable as it’s not every day you
see eight pieces on the d-file. This equal position can be reached by
various
move-orders.

7...0-0 8 e5

8 Nxd4 transposes to the note to White’s 7th move.

8...Bc7 9 Nb3

The Greek Gift doesn’t work: 9 Bxh7+?? Kxh7 10 Ng5+ Kg8 11 Qh5
and now the simple defence 11...Bf5 is winning for Black.

9...h6 10 Re1 Nd7 11 Nbxd4 Nc5 12 Bf4 Nxd3 13 Qxd3 c5

Black should be preferred here, though in Catt-Vlasveld, corr. 2009

White held on for the draw.

29C: 8th-Move Alternatives

(1 e4 c6 2 d3 d5 3 Nd2 e5 4 Ngf3 Bd6)

5 g3 Nf6 6 Bg2 0-0 7 0-0 Re8 8 b3

8 Re1 is the most popular move, to which our reply, 8...Bc7!?, has only
been played in a few games. Once again, I recommend a useful semi-
waiting move that allows us to react logically to White’s next move. Then:

a) 9 b3 d4 reaches positions similar to those after 8 b3, the difference

being White’s rook on e1 and Black’s bishop on c7. I believe this is to
Black’s advantage, as the rook is misplaced when it comes to playing f4.

b) 9 c3 dxe4 10 Nxe4 (or 10 dxe4 a5 with the idea of ...Na6-c5, when

Black is not worse) 10...Nxe4 11 dxe4 Qxd1 12 Rxd1 Nd7 =.

8...d4 (D)
 White to play

The position resembles a Classical King’s Indian where White has ‘wasted’
a move on b3, while Black has arguably done the same by placing the
pawn on c6, though this move may nevertheless prove useful in some
cases. Another important difference is Black’s bishop on d6, where it
normally isn’t in a Classical KID. Now:

a) 9 Kh1 has the idea of Ng1 and f4, but I don’t believe such a slow plan
should be dangerous. 9...b5!? 10 Ng1 Qc7 11 Nb1?! (against 11 Ne2 I
suggest 11...a5 12 a4 bxa4 13 Rxa4 c5 intending ...Bd7 followed by
...Nc6,
...Nb4 and ...a4) 11...a5 leaves Black better, Francsics-A.Vajda, Eger 2003.

b) After 9 Nc4?! Bc7 10 a4, as in Malmdin-E.Hermansson, Skellefteå

2000, Black can transpose to line ‘e’ below with 10...a5.

c) 9 Nh4 is a typical King’s Indian move. White prepares f4 or Nf5.

The prophylactic 9...Bc7 is strong:

c1) 10 f4? exf4 11 gxf4 Nxe4 leaves Black a sound pawn up.

c2) After 10 Nf5 the knight can be ignored for the time being: 10...a5 11 a4
(11 Nf3 Bxf5 12 exf5 Qd7) 11...Na6 and White’s problem is that 12
f4? doesn’t work because of 12...Bxf5 13 exf5 exf4 14 gxf4 and now,
e.g., 14...Nb4 –/+.

d) 9 Ne1 is another logical attempt. The idea is to play f4. I like 9...Bg4
with the idea 10 f3 Bh5 11 g4 Bg6 and thanks to White’s misplaced knight
on e1, Black has every reason to be happy. If White ever plays f4, Black
will capture and place his knight on e5.

e) After 9 a4 a5 10 Nc4?! (the knight is misplaced here, as there is no

real pressure on e5, while the knight becomes a target) 10...Bc7 11
Qd2 b5 12 axb5?! cxb5 13 Na3 Bd7 14 Nh4 g6 (14...Na6!?) 15 Bb2?!
Nh5 –/+ Black went on to win in De Toledo-B.Johnson, corr. 2014.

Lesson 30: Rare Second Moves

By: SJ

I suggest a new strategy, R2: let the Wookiee win.

C-3PO

1 e4 c6

In this lesson you will find some information on White’s alternatives to 2

c4, 2 Nc3, 2 Nf3 and 2 d3. They don’t have much in common, so just a
brief overview before we go straight to the specifics:

2 Bc4?! is very dubious but covered in Model Game 30.

2 Ne2 is quite a respectable move. Thanks to our recommendation
of 3...c5 against the Advance Variation it will mostly transpose to
13A after 2...d5 3 e5 c5 4 d4. For a few independent lines, see
30A.
For 2 b3, see 30B
For 2 f4 and a few other very rare moves, see 30C.

Model Game 30
 S. Williams – G. Jones

British League (4NCL) 2011/12

1 e4 c6 2 Bc4?!
(D)

Black to play
The bishop is vulnerable
here.

2...d5 3 Bb3?!

This gambit is known as the Hillbilly Attack. Objectively it’s no doubt worse
than first exchanging pawns (when Black will be simply a little better), but
it’s an attempt to make sense of the early bishop development.

3...dxe4

The first step in refuting an unsound gambit is accepting it.

4 Qh5
Before you conclude that White is a rank beginner hoping to win by
attacking f7, you should be aware that the World Champion has played this
too – in blitz. Another high-level blitz game was Firouzja-Cheparinov, St
Petersburg blitz 2018 where after 4 Nc3 Nf6 5 Nge2 Na6 6 0-0 Nc5 7 Ng3
Black could have played 7...a5 –/+.

4...g6

Why not fianchetto when it wins a tempo?

5 Qh4 Bg7!

The second step of the refutation very frequently is returning the pawn for a
lead in development. This natural move is very flexible. It’s not yet clear
where the knight belongs, but 5...Nf6 should also be fine; e.g., 6 Nc3 Bf5 7
f3 exf3 8 Nxf3 Nbd7 9 d4 (9 h3!? may be close to equal) 9...e6 10 0-0 h6
=+ E.Berg-Hillarp Persson, Swedish Ch, Sunne 2015.

6 Nc3

Or:

a) After 6 Qxe4, Black wins more time on the queen with 6...Nf6. After 7
Qe2 0-0 8 Nf3 Black should go for 8...e5! as 9 Nxe5? costs a piece due to
the e-file pin after 9...Nd5!.

b) 6 f3 may be the best practical try but after 6...exf3 (6...a5!?) 7 Nxf3 Nh6
8 0-0 Nf5 9 Qf2 0-0 White’s compensation for the pawn is invisible.

6...a5! (D)
 White to play

This advance goes very well with the kingside fianchetto.

7 Nxe4?

7 Qxe4 Nf6 is still very good for Black. Probably White’s best tries are 7
a4 and 7 a3.

7...a4 8 Bc4 b5 9 Be2 Nh6!

This is the consistent follow-up of Black’s play so far.

10 g4?!

In case you were wondering, White is a respected grandmaster. He is

rumoured to play below his usual strength on some Sunday mornings,
but surprisingly it appears that this game was played on a Saturday.

10...a3! 11 Rb1?

This loses but even the relatively best 11 Nc3 b4 12 Nd1 is hopeless after
12...Qd5 13 Bf3 Qe6+ 14 Ne2 Nxg4.

11...Qd5!

This is completely winning.

12 Nc3

Or 12 d3 Qxa2.

12...Qxh1 13 Kf1 Be5!

The queen is safe and Black is a rook up.

14 Bf3 Qxh2 15 Nh3 Nxg4!

Of all Black’s winning moves, this is the clearest.

16 Bxg4 Bf6! 0-1

Theory Magnifiers

1 e4 c6

30A: 2 Ne2

30B: 2 b3

30C: (2 f4)

Honestly, White’s rare 2nd moves have little in common except being
basically harmless. Therefore I don’t recommend spending much of your
study time on these moves unless you have a good reason to believe that
your next opponent may go for one of them.

30A: 2 Ne2
(1 e4 c6)

2 Ne2

This move actually isn’t bad. It scores best of all White’s second moves
(including 2 d4) and was played by both Kasparov and Short as
teenagers.

2...d5 (D)

White to play
3 e5

Now the relationship with the Advance Variation is obvious.

3...c5

In contrast to the 3...Bf5 4 Ng3 (or even 4 Nd4!?) lines, which can be
quite complicated, this is a work saver, mostly transposing to the
Arkell/Khenkin Variation of Lesson 13.

If instead you prefer the experimental 3...Qc7 against the Advance, this is
obviously the move for you here too, as 4 d4 immediately transposes to
18A. That leaves only the untested 4 f4, when Black has a lot of options
but I like 4...h5!? more than in similar positions where it just helps control
g4 and indirectly f5. Here it’s hard to see where the e2-knight is going, if
not to g3 where it’s vulnerable to the ...h4 attack.

4 b4?!

A similar idea to the French Wing Gambit. Otherwise:

a) 4 d4 transposes directly to a not too critical line of the Advance

Variation discussed in 13A.

b) 4 f4 Nc6 5 Nbc3 Bg4 6 h3 Bxe2 7 Bxe2 e6 8 0-0 g6 9 Bf3 h5 10 Ne2

Nh6 11 c4 dxc4 12 Qa4 Qd7 =+ Varga-Burmakin, Eger 1993.

c) 4 c3 Nc6 and now:

c1) 5 d4 is another transposition to 13A.

c2) After 5 f4!?, I don’t particularly like 5...h5?!, as in Reinaldo Castineira-

Rodriguez Lopez, Mondariz 2002. 5...Nh6 has some of the same ideas
and looks more to the point.

4...Nc6!?

A good practical choice. 4...cxb4 is no doubt fine objectively, but White will
have compensation like in the Sicilian and French Wing Gambits after 5 a3
or 5 d4.

5 bxc5 Nxe5 6 d4

For 6 Bb2 Nc6 7 d4, see line ‘c’ below.

6...Nc6

Black probably has somewhat the better chances as ...e5 is an

important resource:
a) After 7 c4 Nf6 8 Bb2? (8 Nbc3 is superior), as in A.Tomas-
Vranković, Split 2012, Black has 8...e5 9 cxd5 Nxd5 10 Nbc3 Be6 –/+.

b) 7 Nbc3 Nf6 8 Bg5 e6 gave Black a modest plus in Sv.Johnsen-

T.Eriksen, Norwegian Team Ch 2001.

c) Against 7 Bb2, as in Arribas Lopez-J.Liang, Internet blitz 2020, Black

should probably develop normally with something like 7...Nf6 8 Nf4 g6
9 Bb5 Bg7 10 0-0 0-0 =+.

30B: 2 b3

(1 e4 c6)

2 b3

This is sometimes called the Euwe Attack after the world champion who
first played it.

2...d5 (D)

The standard Caro-Kann move. A case could also be made for 2...e5!?,
taking the play into ‘Open Games’ territory where a queenside fianchetto
doesn’t seem to fit well.
 White to play
3 Bb2

Only this pawn sacrifice has any bite. It’s sometimes called the Réti
Gambit, but that designation should probably be reserved for the French
version of the gambit (1 e4 e6 2 b3 d5 3 Bb2). White obviously has a few
moves which preserve material equality but none of them are of any
theoretical interest.
However, 3 exd5 cxd5 4 Bb2 Nc6 5 g3?! e5 6 Bg2, as in Euwe-Réti,
Amsterdam (3) 1920, is of some historical interest as the game was played
a couple of months before Réti introduced his gambit against the French.
Black got an advantage from 6...Be6 but 6...h5 and 6...Nf6 are probably
even stronger.

3...dxe4 4 Nc3

Stockfish considers 4 Ne2 inaccurate because of 4...e5!? 5 Bxe5 and now

the surprisingly strong 5...h5!?. The lines are fascinating but complicated
and considering how rarely you will face this line, it doesn’t seem worth the
effort to study them in detail. Instead I recommend 4...Bf5 5 Ng3 e6 with a
likely transposition to the main line after 6 Nc3 Nf6.
4...Nf6

Objectively, Black is already better.

5 Nge2

5 Qe2 Bf5 6 h3 (6 0-0-0 e6) 6...h5 7 g3 e6 8 Bg2 Bb4 9 0-0-0 (Tu Hoang
Thong-B.Cheng, Melbourne 2013) 9...Qe7 10 Nxe4 Bxe4 11 Bxe4 Ba3
12 Nf3 a5 =+.

5...Bf5

5...e5 6 Ng3 Be6 7 Ncxe4 Nxe4 8 Nxe4 Nd7 (8...f5 looks very attractive) 9
Be2?! (as this could be strongly met by 9...f5!, White may prefer 9 h4 or
even the outlandish gambit 9 f4!? exf4 10 Qh5, though a sensible response
should leave Black comfortable) 9...Be7?! 10 0-0 0-0 11 Re1 Re8 12 Bf1
Qc7 =
M.Popchev-Banov, Plovdiv 2014.

6 Ng3 e6 7 Qe2 Nbd7 (D)

White to play
White will get his pawn back but his set-up looks a bit artificial, so it
seems Black should be at least equal:

a) 8 Ncxe4 Bg6 9 h4 (9 0-0-0 leads to similar play) 9...h6 10 0-0-0 a5 11 a4

Qc7 12 h5 Bh7 13 Qf3 Nd5 14 Nc3 Nb4 =+ De Filippis-Rogetzer, corr.
2011.

b) 8 Ngxe4 Be7 and here:

b1) 9 Nxf6+ Bxf6 10 0-0-0 and now rather than 10...Qa5 11 Kb1 (11
g4!?) 11...0-0-0 =, as in Alexandria-Heemskerk, Women’s Candidates,
Subotica 1967, 10...h5! 11 d4 a5 12 a3 b5 may give Black an edge.

b2) 9 0-0-0 a5 10 f3 a4 11 Nxa4?! 0-0 (11...b5!? could be even stronger,

as 12 Nac3? Nd5 leaves White in deep trouble) 12 g4? (12 Nxf6+
minimizes White’s disadvantage) 12...Bxe4 13 fxe4 b5 14 Nc3 and
instead of 14...b4? 15 Na4 Nb6, as in Reprintsev-Dreev, Internet blitz
2003, Stockfish’s suggestion 14...Nb6, planning ...b4 or ...Ba3
depending on circumstances, gives Black the advantage.

30C: Really Rare 2nd Moves

(1 e4 c6)

2 f4

Even rarer moves include:

a) 2 c3 d5 3 exd5 cxd5 4 d4 transposes to a harmless Exchange

Variation found in note ‘b’ to White’s 4th move in 19A.

b) 2 Be2 can be a little tricky (in a similar way to the related 2 f4). After
2...d5, 3 d3 transposes to note ‘a’ to White’s 3rd move in Model Game
29, but 3 e5 (D) is more challenging:
 Black to play

b1) 3...Bf5 is a viable move, but doesn’t really fit with our repertoire
against the Advance Variation after 4 Nf3 e6 5 d4.

b2) 3...Qc7 can be considered if that’s how you normally meet the
Advance Variation. Then 4 d4 is similar to lines in 18A, while 4 f4 doesn’t
look terrifying; both 4...c5 and 4...Nh6 should be fine.

b3) 3...c5 is our main recommendation:

b31) For 4 f4, see the note to White’s 4th move below.

b32) Against 4 c3, I suggest 4...d4, which seems untested and interesting.

b33) 4 Nf3 doesn’t instantly strike one as very logical, but after 4...Nc6 5 0-
0 Bg4 White has the sneaky idea of 6 c4!?, when Black must respond with
some care. 6...e6?! 7 cxd5 exd5 (or 7...Qxd5 8 Qa4!) 8 Nc3 may prove
uncomfortable for Black, so he should probably play 6...d4!?. This looks
like a pawn blunder in view of the simple tactic 7 Nxd4, but Black then has
numerous good options including 7...Bxe2 (7...Nxe5 and 7...Qxd4 are
playable too) 8 Nxe2 e6, when White will hardly be able to cling to his
extra
pawn without paying a major positional cost; e.g., 9 f4 Qd7 intending ...0-0-
0, and bringing the knight to f5. 7 d3 has several good answers, including
7...e6, when taking on d4 is even less appealing for White, and 8 Bf4
invites Black to expand on the kingside; e.g., 8...Nge7 (8...h6!?) 9 h3 Bf5
10 Bh2 h5 11 Nbd2 Ng6 12 Re1 Be7 13 Bf1 h4 with at least equal
chances.

2...d5 (D)

Naturally we stay true to the Caro-Kann spirit.

White to play
3 e5

3 Nc3?! dxe4 4 Nxe4 Nf6 5 Nf2 is reminiscent of a reversed Leningrad

Dutch, and an interesting idea if you don’t mind fighting for equality with
White. After 5...Qc7 6 g3 g6 7 Nf3 Bg7 8 Bg2 0-0 9 0-0 Ng4 10 h3?! Nxf2
11 Rxf2 c5 12 d3 Nc6 13 c3 b6 Black’s position was a little more
harmonious in Perez Candelario-Illescas, Spanish Ch, Seville 2004.

3...c5!
Given our recommendation against the Advance Variation, this move
is obvious when it doesn’t even involve a pawn sacrifice.

4 Nf3

Or 4 Be2 Nc6 5 c3 (5 Nf3 transposes to note ‘b’ after Black’s 4th move)
and now Black’s best is probably 5...Bf5 6 Nf3 e6 7 0-0 Nh6 8 d3 Be7
leading to a position also known with Black to move from a Sicilian move-
order.

Q: That sounds ominous. The Sicilian is a sharp, counter-attacking

opening where every tempo is critical. Have we been tricked?

No, we haven’t. What you say is true for lines like the Dragon and
Najdorf but not for the slow and closed lines where White just gets his
pieces out. Black has his share of the centre and no problems with his
development.
Chances are equal and after 9 h3 f6, the charge 10 g4?! is probably too
optimistic: 10...Bg6 11 exf6 gxf6 12 f5 (12 Nh4 f5 13 g5 Nf7 =+) 12...Nxf5
13 gxf5 Bxf5 =+ 14 Nh4? Bxh3 –/+ An.Rodriguez-Burgos, São Paulo
2004.

4...Nc6 (D)
White to play
Despite having spent two moves getting his pawn to c5, Black has a
very comfortable position. Some examples of play:

a) 5 c3 Bg4 6 h3 Bxf3 7 Qxf3 e6 8 d3 f6 9 exf6 Nxf6 10 Na3 Bd6 11 Be2 0-0

=+ Beshukov-T.Ivanov, Novgorod 1999.

b) After 5 Be2 Bg4 (5...Bf5 looks fine too) 6 0-0 e6 7 d3 Nh6 8 Be3 Nf5 9
Bf2 Be7 10 Nbd2 h5 11 c3, as in Fedoseev-Naer, Moscow 2016, Black
has 11...g5!? (11...h4 is more measured) 12 Nxg5 Bxg5 13 fxg5 Qxg5
with a kingside initiative.

c) 5 Nc3 Bg4 (5...Nh6!?) 6 Bb5 e6 7 Bxc6+ bxc6 8 d3 Nh6 9 0-0 Nf5 10 Qe1
h5, as in Danielsen-Skytte, Copenhagen 1999, is hard to evaluate but
Black shouldn’t be worse.
Index of Variations

1 e4 c6

2 d4
2 Bc4?!; 2 c3; 2 f4; 2 Be2; 2 b3

2 Ne2 d5 3 e5:

a) 3...Qc7: 4 f4; 4 d4

b) 3...c5: 4 d4; 4 f4; 4 c3; 4 b4?!

2 d3 d5 3 Nd2 (3 Be2; 3 Nf3; 3 Qe2) 3...e5 4 Ngf3 Bd6 5 g3 (5 d4) 5...Nf6

6 Bg2 0-0 7 0-0 Re8: 8 c3; 8 Re1; 8 b3

2 c4 d5 3 exd5 (3 Nc3?!; 3 e5) 3...cxd5 4 cxd5 (4 d4 – 2 d4 d5 3 exd5

cxd5 4 c4) 4...Nf6 5 Nc3 (5 Bb5+; 5 Qa4+) 5...Nxd5 (5...g6) 6 Nf3 (6 Qb3;
6 Bb5+;
6 Bc4) 6...g6 (6...Nxc3) 7 Qb3 (7 Bb5+; 7 Bc4) 7...Nb6: 8 d4; 8 Bb5+

2 Nf3 d5 3 exd5 (3 Nc3 – 2 Nc3 d5 3 Nf3) 3...cxd5 4 Ne5 (4 d4) 4...Nc6 5 d4

e6 6 Bb5

2 Nc3 d5:

a) 3 d4 – 2 d4 d5 3 Nc3

b) 3 Qf3

c) 3 Nf3:

c1)

3...Bg4:
c11) 4 Be2

c12) 4 d4 e6 5 h3 (5 Bd3) 5...Bxf3 6 Qxf3 – 4 h3 Bxf3 5 Qxf3 e6 6 d4

c13) 4 h3 Bxf3 5 Qxf3 (5 gxf3) 5...e6 6 d4 (6 g4; 6 Qg3; 6 a3; 6 g3; 6 Be2;
6 d3) 6...Nf6: 7 Be3; 7 g3; 7 e5; 7 Bg5; 7 Bd3; 7 exd5

c2) 3...dxe4 4 Nxe4

Nf6: c21) 5 Ng3

c22) 5 d3

c23) 5 Nxf6+ exf6: 6 d4 – 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6+ exf6

6 Nf3; 6 Be2; 6 g3; 6 Bc4

c24) 5 Qe2 Na6!? 6 d4 (6 d3; 6 Nxf6+) 6...Qd5 7 Nc3 (7 Nxf6+

gxf6) 7...Qa5: 8 Qe5?!; 8 Ne5

2...d5

Now:

A: 3 exd5

B: 3 e5

C: 3 Nc3

Or:

3 Nd2 dxe4 4 Nxe4 – 3 Nc3 dxe4 4 Nxe4

3 Nf3?

3 c4?!
3 Bd3

3 f3 e6 4 Nc3 (4 e5?!; 4 Be3?!) 4...Bb4: 5 e5; 5 Qd3; 5 Bd2; 5 Ne2; 5 Bf4; 5

a3

A)

3 exd5 cxd5

4 c4
4 Nc3; 4 c3; 4 Bf4; 4 Nf3

4 Bd3 Nc6 5 c3 Qc7:

a) 6 Qb3?

b) 6 Nf3

c) 6 Bg5

d) 6 Nd2

e) 6 Na3

f) 6 h3 g6: 7 Qc2; 7 Ne2!?; 7 Nf3

g) 6 Ne2 Bg4 7 0-0 (7 Bf4; 7 f3) 7...Nf6: 8 Bf4; 8 Qc2; 8 h3; 8 f3; 8 Qe1

4...Nf6

4...dxc4 5 Bxc4 (5 Nf3; 5 Nc3) 5...Qc7 6 Bb3 (6 Qe2; 6 Be2; 6 Qd3; 6 Qb3)
6...Bg4: 7 Ne2?!; 7 Nf3; 7 f3

5 Nc3
5 c5?!; 5 Bg5

5 cxd5 Nxd5: 6 Nc3 g6 – 5 Nc3 g6 6 cxd5 Nxd5; 6 Bc4

5 Nf3 g6 6 cxd5 (6 Nc3 – 5 Nc3 g6 6 Nf3; 6 c5) 6...Bg7: 7 Nc3; 7 Bb5+

5...g6

6 Qb3

6 Bf4; 6 c5; 6 Bg5

6 Nf3 Bg7: 7 cxd5; 7 Bg5; 7 Be2; 7 Qb3

6 cxd5 Nxd5:

a) 7 Qb3 Nb6: 8 a4; 8 Nf3; 8 Bb5+; 8 d5

b) 7 Bc4 Nb6 8 Bb3 Bg7 9 Nf3 (9 Nge2; 9 Be3; 9 d5) 9...0-0: 10 Be3; 10 0-0;
10 h3; 10 d5

6...Bg7

7 cxd5 0-0

8 Be2

8 Nf3; 8 g3; 8 Nge2?!

8...a5!

9 a4?!; 9 Bf3; 9 Nf3!

B)
3 e5 c5
3...Qc7:

a) 4 Nd2

b) 4 c4

c) 4 Ne2

d) 4 Nf3

e) 4 Nc3 Bf5 5 h4 (5 f4; 5 Be3; 5 g4) 5...h5: 6 Nce2; 6 Be2; 6 Nge2; 6 Bd3

f) 4 c3 c5 5 Be2 (5 Bd3; 5 Nf3; 5 Be3) 5...Nc6

g) 4 Bd3 c5: 5 Nf3; 5 c3

4 dxc5
4 c4

4 Ne2

4 c3 Nc6 5 Be2 (5 Ne2; 5 dxc5; 5 Be3; 5 Nf3): 5...Qc7; 5...Bf5

4 Nf3 Bg4 5 dxc5 (5 Bb5+ Nc6 6 dxc5; 5 c4) 5...Nc6: 6 c3; 6 Bb5

4...e6

5 a3

5 b4?!; 5 Bd3; 5 Qg4

5 Be3 Nd7: 6 f4; 6 c4; 6 Nf3; 6 Bb5

5 Nf3 Bxc5 6 Bd3 (6 a3 – 5 a3 Bxc5 6 Nf3) 6...Ne7 7 0-0 Ng6 8 Nbd2 (8 a3;
8 Nc3; 8 Re1; 8 Bg5; 8 Qe2) 8...Nc6 9 Nb3 (9 Re1?) 9...Bb6 10 Re1
(10 Qe2) 10...0-0: 11 a4?!; 11 Bxg6; 11 c3; 11 Be3; 11 Bg5

5...Bxc5

6 b4

6 Bd3; 6 Qg4

6 Nf3 Ne7: 7 b4; 7 Bd3

6...Be7

7 Nf3

7 Bb2; 7 Bd3; 7 c4; 7 Qg4

7...a5

8 b5 Nd7

9 Bb2; 9 Be2; 9 Bd3; 9 c4

C)

3 Nc3 dxe4

4 Nxe4
4 Bc4?!; 4 f3?!

Now:

C1: 4...Bf5
C2: 4...Nf6

C1) 4...Bf5

5 Ng3
5 Nc3; 5 Bd3; 5 Qf3; 5 Nc5

5...Bg6

6 h4
6 f4; 6 Bc4

6 N1e2 e6: 7 Nf4 – 6 Nh3 e6 7 Nf4; 7 h4 h6 8 Nf4 – 6 h4 h6 7 Nh3 e6 8 Nf4

6 Nf3 e6: 7 h4 h6 – 6 h4 h6 7 Nf3 e6; 7 Bf4!?; 7 Bc4; 7 Bd3; 7 Ne5

6 Nh3 e6 7 Nf4 Bd6 8 h4 (8 Nfh5; 8 Ngh5; 8 Bc4; 8 Nxg6; 8 c3) 8...Qc7

9 Nxg6 (9 h5?!) 9...hxg6 10 Ne4 (10 Qf3) 10...Nf6!? (10...Bf4) 11 Nxd6+
Qxd6: 12 g3; 12 Be2; 12 c3

6...h6

7 Nf3
7 Bc4; 7 c3; 7 Bd3!?; 7 f4

7 N1e2 e6 8 Nf4 – 7 Nh3 e6 8 Nf4

7 h5 Bh7 8 Nf3 (8 f4; 8 Bd3) 8...e6 – 7 Nf3 e6 8 h5 Bh7

7 Nh3 e6 8 Nf4 Bh7 9 Bc4 (9 Be3 Nf6; 9 c3 Nf6) 9...Nf6 10 0-0 (10 Qe2?!;
10 Nfh5; 10 Ngh5; 10 c3) 10...Bd6:

a) 11 Bxe6?

b) 11 Qe2?!

c) 11 Nxe6?! fxe6 12 Bxe6 Na6!? 13 Re1 (13 Qf3?; 13 Nh5?):

13...Kf8?!; 13...Nc7!

d) 11 Ngh5

e) 11 Re1

7...e6

8 Ne5
8 Bc4

8 Bd3 Bxd3 9 Qxd3 Nf6 10 Bd2 (10 Bf4) 10...Qc7 11 0-0-0 (11 c4)
11...Nbd7: 12 Qe2; 12 c4; 12 Ne4; 12 Rhe1; 12 Kb1

8 h5 Bh7 9 Bd3 (9 Bc4) 9...Bxd3 10 Qxd3 Nf6 11 Bd2 (11 Be3; 11 Ne5; 11
Ne4!?; 11 Bf4) 11...Be7 12 0-0-0 (12 Ne4) 12...0-0: 13 Nf1?!; 13 c4?!; 13
Ne4; 13 Rhe1; 13 Ne5; 13 Qe2; 13 Kb1

8...Bh7

9 Bd3

9 Qf3; 9 c3; 9 Bf4; 9 Be3; 9 Qh5; 9 Bc4

9...Bxd3
10 Qxd3 Nd7

11 f4

11 Ng6??; 11 Nxd7; 11 Qe2

11 Bf4 Nxe5 12 Bxe5 (12 dxe5) 12...Qa5+ 13 c3 (13 Qd2?!) 13...Nf6: 14

Bxf6; 14 b4; 14 Qe2; 14 Ne4; 14 0-0

11...Ngf6

11...Ne7!?

11...Be7 12 Nh5 (12 Bd2) 12...Ngf6 13 Nxg7+ Kf8 14 Nxe6+ (14 Qg6?)
14...fxe6: 15 Ng6+; 15 Bd2!

12 Bd2 Be7

13 0-0-0

13 Qf3

13...0-0

14 Qf3

14 Ne4; 14 Ne2; 14 Rhe1; 14 Kb1; 14 Qe2!?

14...Qc7

15 c4

15 Kb1; 15 Rhe1
15...a5

15...Rfe8!?

16 Kb1

16 Ne2: 16...h5; 16...c5

16...Rfd8!?

16...Rad8?!

C2) 4...Nf6

5 Nxf6+

5 Nc3; 5 Ng5; 5 Qd3; 5 Ng3

5...exf6

6 c3

6 Bd3?!; 6 h3; 6 g3; 6 Be2; 6 Ne2; 6 Bf4; 6 Be3

6 Nf3 Bd6 7 Be2 (7 h3; 7 Qe2+?; 7 Be3; 7 Bc4; 7 Bd3) 7...0-0 8 0-0 Re8 9
h3 (9 Be3; 9 Re1) 9...Nd7: 10 Bd3; 10 c4

6 Bc4 Bd6 (6...Na6) 7 Qe2+ (7 Qh5; 7 Ne2) 7...Be7 8 Nf3 (8 Be3) 8...0-0 9
0-0 (9 Be3) 9...Re8 10 Re1 Nd7: 11 Bd2; 11 c3; 11 Bd3

6...Bd6

6...Bf5!?
7 Bd3 0-0

8 Qc2

8 Qf3

8 Ne2 Re8: 9 Qc2 – 8 Qc2 Re8+ 9 Ne2; 9 0-0; 9 Be3

8...Re8+

9 Ne2

9 Be3 h5: 10 Ne2 – 9 Ne2 h5 10 Be3; 10 0-0-0; 10 Nf3

9...h5

10 Be3

10 Bd2

10 h3

10 0-0 h4 (10...f5; 10...Nd7) 11 h3 (11 Re1; 11 Nf4; 11 Bf4) 11...Nd7 12 Bd2

(12 c4; 12 Kh1; 12 Be3; 12 Bf4!?) 12...Nf8: 13 Rad1; 13 c4; 13 Rfe1; 13
Rae1

10...Nd7

10...Na6

11 0-0-0

11 Ng3

11...Nf8
11...b5!?

12 Kb1

12 c4; 12 Ng3?!; 12 h3

12...Be6

13 Nf4?; 13 Ng3?!; 13 c4; 13 h3

Copyright Information

First published in the UK by Gambit Publications Ltd 2021

First Kindle edition published by Gambit Publications Ltd in 2021
Copyright © Sverre Johnsen and Torbjørn Ringdal Hansen 2021
The right of Sverre Johnsen and Torbjørn Ringdal Hansen to be identified as
the authors of this work has been asserted in accordance with the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in
or introduced into a retrieval system, or transmitted, in any form or by any
means (electronic, mechanical, photocopying, recording or otherwise),
without prior permission of the publisher. In particular, no part of this
publication or images thereof may be transmitted via the Internet or
uploaded to a website without the publisher’s permission. Any person who
does any unauthorized act in relation to this publication may be liable to
criminal prosecution and civil claims for damage.
ISBN-13: 978-1-911465-70-6
ISBN-10: 1-911465-70-8
(Printed edition: ISBN-13: 978-1-911465-67-
6; ISBN-10: 1-911465-67-8).
Gambit Publications Ltd, 27 Queens Pine, Bracknell, Berks, RG12
0TL, England.
E-mail: info@gambitbooks.com
Website (regularly updated): www.gambitbooks.com
Edited by Graham Burgess
Kindle edition prepared by Graham
Burgess Cover illustration by Wolff Morrow
Gambit Publications Ltd
Directors: Dr John Nunn GM, Murray Chandler GM and Graham
Burgess FM
German Editor: Petra Nunn WFM
About the Authors

Sverre Johnsen is a chess analyst, researcher, organizer, trainer and

writer from Norway. He is co-author of Win with the London System and
Win with the Stonewall Dutch, two of the best-selling openings books of
recent years. Grandmaster Torbjørn Ringdal Hansen is also from
Norway. He is one of the founders of the chess retail business Sjakkhuset
and works full-time as a chess trainer. He was the first coach of Magnus
Carlsen (in 1999) and has worked with three other players who went on to
become grandmasters.