LECTURE 12

Contents of the Lecture

• Introduction.
• Important definitions.
• Two person game.
• Pure strategies.
• Mixed strategies.
• Algebraic method.
• Graphical method.

2
 Introduction
• Game refers to a situation of conflict and competition
in which two or more competitors (or participants)
are involved in the decision making process.

• The competitors are referred to as players.

• A player may be an individual, individuals, or an

organization.

• Example: pricing of products, where sale of any

product is determined not only by its price but also by
the price set by competitors for a similar product.
3
 Important Definitions
• Number of players: If a game involves only 2 players,
then it is two-person game. If number of players are
more, then it is n-person game.

• Sum of gains/losses: If in a game, the sum of gains to ne

player equals the sum of losses to another player, then it
is zero-sum game. Otherwise, it is non-zero sum game.

• Strategy: The strategy for a player is the list of all

possible actions that are likely to be adopted by him for
every payoff (outcome).

4
 Continued…
• The particular strategy that optimizes a player’s losses
or gains, without knowing the competitor’s strategies is
called optimal strategy.
• The expected outcomes when player use their optimal
strategy is called Value of the game.
• Pure strategy: A particular strategy that a player
chooses to play again and again regardless of other
player’s strategy.
• Mixed strategy: A set of strategies that a player
chooses on a particular move of the game with some
fixed probabilities.
5
 Two-Person Zero-Sum Game
• A game with only two players, say A and B, is called
a two-person zero-sum game, only if one player’s
gain is equal to the loss of other player, so that total
sum is zero.
• The payoffs in terms of gain or loss, when players
select their particular strategy, can be represented as
matrix called as Payoff Matrix.
• If player A has m strategies denoted by and player B
has n strategies namely . Let be the payoff values.

6
 Continued…
• It is assumed that player A is always the gainer
whereas player B is the looser. Therefore, represents
the gain of player A from player B, when A uses
strategy and B uses strategy . On the other hand, the
gains to player B and losses to player A can be
written as .
Player A’s Player B’s strategy
strategy …

…
…
. . . . .
: : : : :
7
 Assumptions of the Game

• Each player has available to him finite number of

possible strategies (courses of action).

• List of strategies of each player and the amount of

gain/loss on an individual’s choice of strategy is
known to each player in advance.

• One player attempts to maximize gains and the other

attempts to minimize losses.

• The payoff is fixed and determined in advance.

8
 Pure Strategies
• Maximin Principle: For player A, the minimum value
in each row (row minima) represents the least gain to
him corresponding to each strategy. He will then select
the strategy that gives largest gain. This is called
maximin principle.

• Minimax Principle: For Player B, the maximum value

in each column (column maxima) represents the
maximum loss to him corresponding to each strategy.
He will then select the strategy that gives smallest loss.
This is called minimax principle.
9
 Continued…
• Optimal Strategy: If the minimax value equals the
maximin value, then the game is said to have a saddle
(equilibrium) point and the corresponding strategies are
called optimal strategies.
• Value of a Game (V): This is the expected payoff at the
end of the game, when each player uses his optimal strategy
i.e. the amount of payoff at an equilibrium point.
• Fair game: When the maximin value equals minimax
value, and both equals 0.
• Strictly determinable game: When the maximin value
equals minimax value, and both equals V.
10
 Illustration 1
For the game with payoff matrix. Determine the optimal
strategies for players A & B. Also determine the value
of the game. Is this game fair or strictly determinable?

Player A Player B
B1 B2 B3
A1 -1 2 -2
A2 6 4 -6

11
 Solution
Player A Player B Row
B1 B2 B3 minimum

A1 -1 2 -2 -2
A2 6 4 -6 -6
Column 6 4 -2
maximum

Player A Player B Row

B1 B2 B3 minimum
Maximin
A1 -1 2 -2 -2
A2 6 4 -6 -6
Column 6 4 -2
maximum
Minimax
12
 Continued…
• As it can be observed in previous payoff matrix, the
optimal strategy for player A is A1 whereas the
optimal strategy for player B is B3.

• Since the maximin value equals minimax value i.e. -

2, we say saddle point exists and value of game (V) is
-2.

• Since the value of game is non-zero, by definition, it

is a strictly determinable game.

13
 Illustration 2
A company management and the labour union are negotiating
a new three year settlement. Each of these has four strategies:
I= hard and aggressive bargaining; II= reasoning and logical
approach; III= legalistic strategy; and IV= conciliatory
approach. The costs to the company are given for every pair of
strategy choice. What strategy will the two sides adopt? Also
determine V.

Union Company strategies

strategy
I II III IV

I 20 15 12 35
II 25 14 8 10
III 40 2 10 5
IV -5 4 11 0 14
 Solution

Union Company strategies Row

strategy I II III IV minimum

I 20 15 12 35 12
II 25 14 8 10 8
III 40 2 10 5 2
IV -5 4 11 0 -5
Column 40 15 12 35
maxima

15
 Mixed Strategies
• In certain cases there exists no saddle point i.e.
minimax is not equal maximin. In such cases, players
use mixture of strategies to find value of the game.

Algebraic Arithmetic

Methods Matrix

Simplex Graphical

16
 Dominance Rules
• For player A, if each element in a row, say , is less than
or equal to the corresponding element in another row,
say , in the payoff matrix. Then the row is s.t.b.
dominated by row , and therefore, row can be deleted
from the payoff matrix.

• For player B, if each element in a column, say , is

greater than or equal to the corresponding element in
another column, say , in the payoff matrix. Then the
column is s.t.b. dominated by column , and therefore,
column can be deleted from the payoff matrix.

17
 Algebraic Method
Player A Player B Probability
B1 B2
A1
A2
Probability

𝑎 22 − 𝑎 21 𝑎 22 −𝑎 12
𝑝 1= 𝑞1 =
𝑎 11 +𝑎 22 −( 𝑎12 +𝑎 21) 𝑎11 + 𝑎 22 −(𝑎 12+ 𝑎 21)

𝑝 2=1 − 𝑝 1 𝑞 2=1 − 𝑞1

18
 Illustration 1
A company is currently involved in negotiations with its
union on the upcoming wage contract. Positive sign
represents wage increase while negative sign represents
wage reduction. What are the optimal strategies for the
company as well as the union? What is the game value?

C1 C2 C3 C4
U1 .25 .20 .14 .30
U2 .27 .16 .12 .14
U3 .35 .08 .15 .19
U4 - .02 .08 .13 .00

19
 Solution
• We make use of dominance rule to convert the matrix
into 2X2.
• We observe that strategy U4 is dominated by U3.
therefore, we delete it.

C1 C2 C3 C4

U1 .25 .20 .14 .30

U2 .27 .16 .12 .14

U3 .35 .08 .15 .19

U4 - .02 .08 .13 .00

20
 Continued…
• Strategy C1 is dominated by C2. Therefore, we delete
C1.

• Also, strategy C4 is dominated by C3. Therefore, we

delete C4.

C1 C2 C3 C4

U1 .25 .20 .14 .30

U2 .27 .16 .12 .14

U3 .35 .08 .15 .19

21
 Continued…
• Again, U2 is dominated by U1. Therefore, we delete U2.

C2 C3

U1 .20 .14

U2 .16 .12

U3 .08 .15

22
 Continued…
C2 C3 Prob.

U1 .20 .14 .07/.13

= .538
U3 .08 .15 .06/.13
= .461
Prob. .01/.13 .12/.13
= .076 = .923

• Optimal strategy for Company = (0, .076, .923, 0).

• Optimal strategy for Union = (.538, 0, .461, 0).
• Value of the game (V) = Rs. 14360

23
 Illustration 2
Solve the game whose payoff matrix is given below:

Player A Player B
B1 B2 B3 B4
A1 3 2 4 0
A2 3 4 2 4
A3 4 2 4 0
A4 0 4 0 8

24
 Solution
• Using the dominance rule, we can observe that
strategy A1 is dominated by Strategy A3. Therefore,
we delete A1.

Player A Player B

B1 B2 B3 B4

A1 3 2 4 0

A2 3 4 2 4

A3 4 2 4 0

A4 0 4 0 8

25
 Continued…
• Again, the strategy B1 is dominated by strategy B3.
Therefore, we delete B1.

Player A Player B

B1 B2 B3 B4

A2 3 4 2 4

A3 4 2 4 0

A4 0 4 0 8

26
 Continued…
• If you will observe, there is no way the dominance rule
can break the order of matrix.
• In such case, we compare the row/column dominance
with the average of remaining rows/columns.
• In this case, the average of B3 & B4 is (3,2,4) dominates
strategy B2. Therefore, we delete B2.

Player A Player B

B2 B3 B4

A2 4 2 4

A3 2 4 0

A4 4 0 8
27
 Continued…
• Again, there is no clear dominance relationship.
• Now, the average of A3 & A4 is equal to strategy A2.
Therefore, A2 can be deleted.

Player A Player B

B3 B4

A2 2 4

A3 4 0

A4 0 8

28
 Continued…
• Finally, we have received a 2X2 matrix.
• Next we attach probabilities corresponding to the
strategies.

Player A Player B Prob.

B3 B4

A3 4 0

A4 0 8

Prob.

29
 Continued…
• The given payoff matrix has no saddle point.
Therefore, in order to evaluate the probability values,
we solve the following:

• Similarly, we can get the value for .

30
 Continued…
Player A Player B Prob.

B3 B4

A3 4 0 2/3

A4 0 8

Prob.

• Optimal strategy for A = (0, 0, 2/3, 1/3)

• Optimal strategy for B = (0, 0, 2/3, 1/3)
• Value of game (V) = 4*(2/3) + 0*(1/3) = 8/3

31
 Graphical Method
• This method is useful for the game where the payoff
matrix is of size 2Xn or mX2 i.e. the game with
mixed strategies that has only two non-dominated
pure strategies for one of the players in the two-
person zero-sum game.

32
 Illustration 1

Use the graphical method to solve the following game

and find the value of the game.

Player A Player B
B1 B2 B3 B4
A1 2 2 3 -2
A2 4 3 2 6

33
 Solution
• The game has no saddle point (Check!!).
• If the probability of player A’s playing A1 and A2 is
denoted by and . Then the expected payoff (gain) to
player A will be:

B’s Pure Strategies A’s Expected Payoff

B1
B2
B3
B4

34
 Continued…

• While plotting the graph, we suppose on the X-axis

and the expected payoff values on the Y-axis.

• Since the probability lies between 0 & 1, so the X-

axis also varies within the same.

• For example, considering the strategy B1 by player B,

the expected payoff of player A is 2 when A plays A1
with prob. 1 and 4 when plays A2 with prob. 0.

• Similarly, we can plot the strategies B2, B3, and B4.

35
Continued…

36
 Continued…
• It is assumed that player B will always play his
strategies in such a way so as to minimize his loss.

• Thus the payoffs (gain) will be represented by the

lower boundary.

• Since player A will choose his best possible strategies

in order to realize max expected gain. It can be found
at point P which represents the intersection of
strategy B3 & B4.

37
 Continued…

P (Maximin Point)

V=22/9

=4/9

38
 Continued…
Player A Player B Prob.
B3 B4
A1 3 -2
A2 2 6
Prob.

Solving using the methods described in the algebraic

method. We can easily get the probability values.
Optimal strategy for A = (4/9, 5/9)
Optimal strategy for B = (0, 0, 8/9, 1/9)
Value of game (V) = 22/9
39
 Illustration 2
Obtain the optimal strategies for both persons and the
value of the game for two-person zero-sum game whose
payoff matrix is as follows:

Player A Player B
B1 B2
A1 1 -3
A2 3 5
A3 -1 6
A4 4 1
A5 2 2
A6 -5 0

40
 Solution
• The game does not have a saddle point (check!!).
• If the probability of player B’s strategies B1 & B2 be
denoted by and . Then the payoff matrix to player B
will be:

A’s Pure Strategies B’s Expected Payoff

A1
A2
A3 -
A4
A5
A6

41
 Continued…
• While plotting the graph, we suppose on the X-axis
and the expected payoff values on the Y-axis.

• Since the probability lies between 0 & 1, so the X-

axis also varies within the same.

• For example, considering the strategy A1 by player A,

the expected payoff of player B is 1 when B plays B1
with prob. 1 and -3 when plays B2 with prob. 0.

• Similarly, we can plot the strategies A2 to A6.

42
Continued…

43
 Continued…
• It is assumed that player A will always play his
strategies in such a way so as to maximize his
gains.

• Thus the payoffs (loss) to B will be represented by

the upper boundary.

• Since player B will choose his best possible

strategies in order to realize min expected loss. It
can be found at point P which represents the
intersection of strategy A4 & A2.
44
Continued…

45
 Continued…
Player A Player B Prob.
B1 B2
A2 3 5
A4 4 1
Prob.

Solving using the methods described in the algebraic

method. We can easily get the probability values.
Optimal strategy for A = (0, 3/5, 0, 2/5, 0, 0)
Optimal strategy for B = (4/5, 1/5)
Value of game (V) = 17/5
46