See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/279310708

On Optimal Offensive Strategies in Basketball

Article · June 2015

Source: arXiv

CITATIONS READS
0 65

1 author:

Ikjyot Singh Kohli

York University
26 PUBLICATIONS   30 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Ikjyot Singh Kohli on 18 September 2015.

The user has requested enhancement of the downloaded file.

 ON OPTIMAL OFFENSIVE STRATEGIES IN BASKETBALL

IKJYOT SINGH KOHLI

arXiv:1506.06687v2 [math.DS] 23 Jun 2015

Abstract. The purpose of this paper is to determine whether basketball teams who
choose to employ an offensive strategy that involves predominantly shooting three point
shots is stable and optimal. We employ a game-theoretical approach using techniques from
dynamical systems theory to show that taking more three point shots to a point where
an offensive strategy is dependent on predominantly shooting threes is not necessarily
optimal, and depends on a combination of payoff constraints, where one can establish
conditions via the global stability of equilibrium points in addition to Nash equilibria
where a predominant two-point offensive strategy would be optimal as well. We perform
a detailed fixed-points analysis to establish the local stability of a given offensive strategy.
We finally prove the existence of Nash equilibria via global stability techniques via the
monotonicity principle. We believe that this work demonstrates that the concept that
teams should attempt more three-point shots because a three-point shot is worth more
than a two-point shot is therefore, a highly ambiguous statement.

Contents
1. Introduction 1
2. The Dynamical Equations 2
3. Fixed-Points Analysis 4
4. Global Stability and The Existence of Nash Equilibria 4
5. Discussion 6
References 8

1. Introduction
We are currently living in the age of analytics in professional sports, with a strong
trend of their use developing in professional basketball. Indeed, perhaps, one of the most
discussed results to come out of the analytics era thus far is the claim that teams should
shoot as many three-point shots as possible, largely because, three-point shots are worth
more than two-point shots, and this somehow is indicative of a very efficient offense. These
ideas were mentioned for example by Alex Rucker [1] who said “When you ask coaches
what’s better between a 28 percent three-point shot and a 42 percent midrange shot, they’ll
say the 42 percent shot. And that’s objectively false. It’s wrong. If LeBron James just
jacked a three on every single possession, that’d be an exceptionally good offense. That’s a
Date: June 20, 2015.
1
2 IKJYOT SINGH KOHLI

conversation we’ve had with our coaching staff, and let’s just say they don’t support that
approach.” It was also claimed in the same article that “The analytics team is unanimous,
and rather emphatic, that every team should shoot more 3s including the Raptors and
even the Rockets, who are on pace to break the NBA record for most 3-point attempts in
a season.” These assertions were repeated in [2]. In an article by John Schuhmann [3],
it was claimed that “It’s simple math. A made three is worth 1.5 times a made two. So
you don’t have to be a great 3-point shooter to make those shots worth a lot more than a
jumper from inside the arc. In fact, if you’re not shooting a layup, you might as well be
beyond the 3-point line. Last season, the league made 39.4 percent of shots between the
restricted area and the arc, for a value of 0.79 points per shot. It made 36.0 percent of
threes, for a value of 1.08 points per shot.”
The purpose of this paper is to determine whether basketball teams who choose to employ
an offensive strategy that involves predominantly shooting three point shots is stable and
optimal. Although this problem to the best of the author’s knowledge has not been studied
before in the literature, several studies that provide an in-depth quantitative analysis of
various aspects of basketball games using statistical and game theoretical methods have
been established in [4], [5], [6], [7], [8], [9], [10], [11], [12] and references therein.
We will employ a game-theoretical approach using techniques from dynamical systems
theory to show that taking more three point shots to a point where an offensive strategy
is dependent on predominantly shooting threes is not necessarily optimal, and depends
on a combination of payoff constraints, where one can establish conditions via the global
stability of equilibrium points in addition to Nash equilibria where a predominant two-point
offensive strategy would be optimal as well.

2. The Dynamical Equations

For our model, we consider two types of NBA teams. The first type are teams that
employ two point shots as the predominant part of their offensive strategy, while the other
type consists of teams that employ three-point shots as the predominant part of their
offensive strategy. There are therefore two predominant strategies, which we will denote
as s1 , s2 , such that we define
(1) S = {s1 , s2 } .
We then let ni represent the number of teams using si , such that the total number of
teams in the league is given by
k
X
(2) N= ni ,
i=1
which implies that the proportion of teams using strategy si is given by
ni
(3) xi = .
N
The state of the population of teams is then represented by x = (x1 , . . . , xk ). It can
be shown [13] that the proportions of individuals using a certain strategy change in time
 ON OPTIMAL OFFENSIVE STRATEGIES IN BASKETBALL 3

according to the following dynamical system

(4) ẋi = xi [π(si , x) − π̄(x)] ,
subject to
k
X
(5) xi = 1,
i=1
where we have defined the average payoff function as
k
X
(6) π̄(x) = xi π(si , x).
i=1
Now, let x1 represent the proportion of teams that predominantly shoot two-point shots,
and let x2 represent the proportion of teams that predominantly shoot three-point shots.
Further, denoting the game action set to be A = {T, T h}, where T represents a predominant
two-point shot strategy, and T h represents a predominant three-point shot strategy. As
such, we assign the following payoffs:
(7) π(T, T ) = α, π(T, T h) = β, π(T h, T ) = γ, π(T h, T h) = δ.
We therefore have that
(8) π(T, x) = αx1 + βx2 , π(T h, x) = γx1 + δx2 .
From (6), we further have that
(9) π̄(x) = x1 (αx1 + βx2 ) + x2 (γx1 + δx2 ) .
From Eq. (4) the dynamical system is then given by
(10) ẋ1 = x1 {(αx1 + βx2 ) − x1 (αx1 + βx2 ) − x2 (γx1 + δx2 )} ,
(11) ẋ2 = x2 {(γx1 + δx2 ) − x1 (αx1 + βx2 ) − x2 (γx1 + δx2 )} ,
subject to the constraint
(12) x1 + x2 = 1.
Indeed, because of the constraint (12), the dynamical system is actually one-dimensional,
which we write in terms of x1 as
(13) ẋ1 = x1 (−1 + x1 ) [δ + β (−1 + x1 ) − δx1 + (γ − α) x1 ] .
From Eq. (13), we immediately notice some things of importance. First, we are able to
deduce just from the form of the equation what the invariant sets are. Following [14], we
note that for a dynamical system x′ = f (x) ∈ Rn with flow φt , if we define a C 1 function
Z : Rn → R such that Z ′ = αZ, where α : Rn → R, then, the subsets of Rn defined
by Z > 0, Z = 0, and Z < 0 are invariant sets of the flow φt . Applying this notion to
Eq. (13), one immediately sees that x1 > 0, x1 = 0, and x1 < 0 are invariant sets of the
corresponding flow.
Further, there also exists a symmetry such that x1 → −x1 , which implies that without
loss of generality, we can restrict our attention to x1 ≥ 0.
4 IKJYOT SINGH KOHLI

3. Fixed-Points Analysis
With the dynamical system in hand, we are now in a position to perform a fixed-points
analysis. There are precisely three fixed points, which are invariant manifolds and are given
by:
β−δ
(14) P1 : x∗1 = 0, P2 : x∗1 = 1, P3 : x∗1 = .
−α + β − δ + γ
Note that, P3 actually contains P1 and P2 as special cases. Namely, when β = δ, P3 =
0 = P1 , and when α = γ, P3 = 1 = P2 . We will therefore just analyze, the stability
of P3 . P3 = 0 represents a state of the population where all teams predominantly shoot
three-point shots. Similarly, P3 = 1 represents a state of the population where all teams
predominantly shoot two-point shots, We additionally restrict
β−δ
(15) 0 ≤ P3 ≤ 1 ⇒ 0 ≤ ≤ 1,
−α + β − δ + γ
which implies the following conditions on the payoffs:
(16) [δ < β ∩ γ ≤ α] ∪ [δ = β ∩ (γ < α ∪ γ > α)] ∪ [δ > β ∩ γ ≤ α] .
With respect to a stability analysis of P3 , we note the following.
The point P3 is a:
• Local sink if: {δ < β} ∩ {γ > α},
• Source if: {δ > β} ∩ {γ < α},
• Saddle: if: {δ = β} ∩ (γ < α − β + δ ∪ γ > α − β + δ), or ({δ < β} ∪ {δ > β}) ∩ γ =
αδ−αβ
δ−β .
Further, the system exhibits some bifurcations as well. In the neigbourhood of P3 = 0,
the linearized system takes the form
(17) x′1 = β − δ.
Therefore, P3 = 0 destabilizes the system at β = δ. Similarly, P3 = 1 destabilizes the
system at γ = α. Therefore, bifurcations of the system occur on the lines γ = α and β = δ
in the four-dimensional parameter space.

4. Global Stability and The Existence of Nash Equilibria

With the preceding fixed-points analysis completed, we are now interested in determining
global stability conditions. The main motivation is to determine the existence of any Nash
equilibria that occur for this game via the following theorem [13]: If x∗ is an asymptotically
stable fixed point, then the symmetric strategy pair [σ ∗ , σ ∗ ], with σ ∗ = x∗ is a Nash
equilibrium.
We will primarily make use of the monotonicity principle, which says [14] let φt be a
flow on Rn with S an invariant set. Let Z : S → R be a C 1 function whose range is the
 ON OPTIMAL OFFENSIVE STRATEGIES IN BASKETBALL 5

interval (a, b), where a ∈ R ∪ {−∞}, b ∈ R ∪ {∞}, and a < b. If Z is decreasing on orbits
in S, then for all x ∈ S,
 
ω(x) ⊆ s ∈ ∂S| lim Z(y) 6= b ,
y→s

 
α(x) ⊆ s ∈ ∂S| lim Z(y) 6= a .
y→s

Consider the function

(18) Z1 = log (−1 + x1 ) .

Then, we have that

(19) Ż1 = x1 [δ + β (−1 + x1 ) − δx1 + x1 (γ − α)] .

For the invariant set S1 = {0 < x1 < 1}, we have that ∂S1 = {x1 = 0} ∪ {x1 = 1}. One
can then immediately see that in S1 ,

(20) Ż1 < 0 ⇔ {β > δ} ∩ {α ≥ γ} .

Therefore, by the monotonicity principle,

(21) ω(x) ⊆ {x : x1 = 1} .

Note that the conditions β > δ and α ≥ γ correspond to P3 above. In particular, for α = γ,
P3 = 1, which implies that x∗1 = 1 is globally stable. Therefore, under these conditions,
the symmetric strategy [1, 1] is a Nash equilibrium.
Now, consider the function

(22) Z2 = log (x1 ) .

We can therefore see that

(23) Ż2 = [−1 + x1 ] [δ + β (−1 + x1 ) − δx1 + (−α + γ) x1 ] .

Clearly, Ż2 < 0 in S1 if for example β = δ and α < γ. Then, by the monotonicity principle,
we obtain that

(24) ω(x) ⊆ {x : x1 = 0} .

Note that the conditions β = δ and α < γ correspond to P3 above. In particular, for β = δ,
P3 = 0, which implies that x∗1 = 0 is globally stable. Therefore, under these conditions,
the symmetric strategy [0, 0] is a Nash equilibrium.
In summary, we have just shown that for the specific case where β > δ and α = γ, the
strategy [1, 1] is a Nash equilibrium. On the other hand, for the specific case where β = δ
and α < γ, the strategy [0, 0] is a Nash equilibrium.
6 IKJYOT SINGH KOHLI

5. Discussion
In the previous section which describes global results, we first concluded that for the
case where β > δ and α = γ, the strategy [1, 1] is a Nash equilibrium. The relevance of
this is as follows. The condition on the payoffs thus requires that
(25) π(T, T ) = π(T h, T ), π(T, T h) > π(T h, T h).
That is, given the strategy adopted by the other team, neither team could increase their
payoff by adopting another strategy if and only if the condition in (25) is satisfied. Given
these conditions, if one team has a predominant two-point strategy, it would be the other
team’s best response to also use a predominant two-point strategy.
We also concluded that for the case where β = δ and α < γ, the strategy [0, 0] is a Nash
equilibrium. The relevance of this is as follows. The condition on the payoffs thus requires
that
(26) π(T, T h) = π(T h, T h), π(T, T ) < π(T h, T ).
That is, given the strategy adopted by the other team, neither team could increase their
payoff by adopting another strategy if and only if the condition in (26) is satisfied. Given
these conditions, if one team has a predominant three-point strategy, it would be the other
team’s best response to also use a predominant three-point strategy.
Further, we also showed that x1 = 1 is globally stable under the conditions in (25).
That is, if these conditions hold, every team in the NBA will eventually adopt an offensive
strategy predominantly consisting of two-point shots. The conditions in (26) were shown
to imply that the point x1 = 0 is globally stable. This means that if these conditions now
hold, every team in the NBA will eventually adopt an offensive strategy predominantly
consisting of three-point shots.
We also provided through a careful stability analysis of the fixed points criteria for
the local stability of strategies. For example, we showed that a predominant three-point
strategy is locally stable if π(T, T h) − π(T h, T h) < 0, while it is unstable if π(T, T h) −
π(T h, T h) ≥ 0. In addition, a predominant two-point strategy was found to be locally
stable when π(T h, T ) − π(T, T ) < 0, and unstable when π(T h, T ) − π(T, T ) ≥ 0.
There is also they key point of which one of these strategies has the highest probability
of being executed. From [13], we know that
XX
(27) π(σ, x) = p(s)x(s′ )π(s, s′ ).
s∈S s′ ∈S

That is, the payoff to a team using strategy σ in a league with profile x is proportional to
the probability of this team using strategy s ∈ S. We therefore see that a team’s optimal
strategy would be that for which they could maximize their payoff, that is, for which p(s)
is a maximum, while keeping in mind the strategy of the other team, hence, the existence
of Nash equilibria.
Hopefully, this work also shows that the concept that teams should attempt more three-
point shots because a three-point shot is worth more than a two-point shot is a highly
 ON OPTIMAL OFFENSIVE STRATEGIES IN BASKETBALL 7

ambiguous statement. In actuality, one needs to analyze what offensive strategy is optimal
which is constrained by a particular set of payoffs.
 8 IKJYOT SINGH KOHLI

References
[1] A. Fromal, “Why aren’t NBA teams shooting more threes?,” Bleacher Report, 2013.
[2] A. Fromal, “Why NBA teams will keep shooting more 3-pointers in 2014-15,” Bleacher Report, 2014.
[3] J. Schuhmann, “These days, more than ever, the best teams shoot the three,” NBA.com, 2014.
[4] M. Perse, M. Kristan, S. Kovacic, G. Vuckovic, and J. Pers, “A trajectory-based analysis of coordinated
team activity in a basketball game,” Computer Vision and Image Understanding, vol. 113, no. 5,
pp. 612–621, 2009.
[5] Y. Zhang, C. Xu, Y. Rui, J. Wang, and H. Lu, “Semantic event extraction from basketball games using
multi-modal analysis,” pp. 2190–2193, 2007.
[6] P. Kvam and J. Sokol, “A logistic regression/markov chain model for ncaa basketball,” Naval Research
Logistics, vol. 53, no. 8, pp. 788–803, 2006.
[7] P. Larkey, J. Kadane, R. Austin, and S. Zamir, “Skill in games,” Management Science, vol. 43, no. 5,
pp. 596–609, 1997.
[8] M. Popescu, P. Gader, and J. Keller, “Fuzzy spatial pattern processing using linguistic hidden markov
models,” IEEE Transactions on Fuzzy Systems, vol. 14, no. 1, pp. 81–92, 2006.
[9] P. De Melo, V. Almeida, and A. Loureiro, “Can complex network metrics predict the behavior of nba
teams?,” pp. 695–703, 2008.
[10] B. Clair and D. Letscher, “Optimal strategies for sports betting pools,” Operations Research, vol. 55,
no. 6, pp. 1163–1177, 2007.
[11] R. Dodge, Schelling’s Game Theory: How to Make Decisions. 2012.
[12] X.-J. Zhou, Z. Gao, and Q.-L. Zhang, “Analysis approach of winning factors in competitive basketball,”
pp. 1141–1144, 2008.
[13] J. N. Webb, Game Theory: Decisions, Interaction and Evolution. Springer, first ed., 2007.
[14] J. Wainwright and G. Ellis, Dynamical Systems in Cosmology. Cambridge University Press, first ed.,
1997.

View publication stats