## The “Evolution” of the 3-Point Shot in The NBA

The purpose of this post 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.



### 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 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 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 here. In an article by John Schuhmann, 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. 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. (Article research and other statistics provided by: Hargun Singh Kohli)

### 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 ${s_{1}, s_{2}}$, such that we define

$\displaystyle \mathbf{S} = \left\{s_{1}, s_{2}\right\}. \ \ \ \ \ (1)$

We then let ${n_{i}}$ represent the number of teams using ${s_{i}}$, such that the total number of teams in the league is given by

$\displaystyle N = \sum_{i =1}^{k} n_{i}, \ \ \ \ \ (2)$

which implies that the proportion of teams using strategy ${s_{i}}$ is given by

$\displaystyle x_i = \frac{n_{i}}{N}. \ \ \ \ \ (3)$

The state of the population of teams is then represented by ${\mathbf{x} = (x_{1}, \ldots, x_{k})}$. It can be shown that the proportions of individuals using a certain strategy change in time according to the following dynamical system

$\displaystyle \dot{x}_{i} = x_{i}\left[\pi(s_{i}, \mathbf{x}) - \bar{\pi}(\mathbf{x})\right], \ \ \ \ \ (4)$

subject to

$\displaystyle \sum_{i =1}^{k} x_{i} = 1, \ \ \ \ \ (5)$

where we have defined the average payoff function as

$\displaystyle \bar{\pi}(\mathbf{x}) = \sum_{i=1}^{k} x_{i} \pi(s_{i}, \mathbf{x}). \ \ \ \ \ (6)$

Now, let ${x_{1}}$ represent the proportion of teams that predominantly shoot two-point shots, and let ${x_{2}}$ represent the proportion of teams that predominantly shoot three-point shots. Further, denoting the game action set to be ${A = \left\{T, Th\right\}}$, where ${T}$ represents a predominant two-point shot strategy, and ${Th}$ represents a predominant three-point shot strategy. As such, we assign the following payoffs:

$\displaystyle \pi(T,T) = \alpha, \quad \pi(T,Th) = \beta, \quad \pi(Th, T) = \gamma, \quad \pi(Th,Th) = \delta. \ \ \ \ \ (7)$

We therefore have that

$\displaystyle \pi(T,\mathbf{x}) = \alpha x_{1} + \beta x_{2}, \quad \pi(Th, \mathbf{x}) = \gamma x_{1} + \delta x_{2}. \ \ \ \ \ (8)$

From (6), we further have that

$\displaystyle \bar{\pi}(\mathbf{x}) = x_{1} \left( \alpha x_{1} + \beta x_{2}\right) + x_{2} \left(\gamma x_{1} + \delta x_{2}\right). \ \ \ \ \ (9)$

From Eq. (4) the dynamical system is then given by

$\boxed{\dot{x}_{1} = x_{1} \left\{ \left(\alpha x_{1} + \beta x_{2} \right) - x_{1} \left( \alpha x_{1} + \beta x_{2}\right) - x_{2} \left(\gamma x_{1} + \delta x_{2}\right) \right\}}$,

$\boxed{\dot{x}_{2} = x_{2} \left\{ \left( \gamma x_{1} + \delta x_{2}\right) -x_{1} \left( \alpha x_{1} + \beta x_{2}\right) - x_{2} \left(\gamma x_{1} + \delta x_{2}\right) \right\}}$,

subject to the constraint

$\displaystyle x_{1} + x_{2} = 1. \ \ \ \ \ (10)$

Indeed, because of the constraint (10), the dynamical system is actually one-dimensional, which we write in terms of ${x_{1}}$ as

$\displaystyle \boxed{\dot{x}_{1} = x_{1} \left(-1 + x_{1}\right) \left[\delta + \beta \left(-1 + x_{1}\right) - \delta x_{1} + \left(\gamma-\alpha\right)x_{1}\right]}. \ \ \ \ \ (11)$

From Eq. (11), 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. We note that for a dynamical system ${\mathbf{x}' = \mathbf{f(x)} \in \mathbf{R^{n}}}$ with flow ${\phi_{t}}$, if we define a ${C^{1}}$ function ${Z: \mathbf{R}^{n} \rightarrow \mathbf{R}}$ such that ${Z' = \alpha Z}$, where ${\alpha: \mathbf{R}^{n} \rightarrow \mathbf{R}}$, then, the subsets of ${\mathbf{R}^{n}}$ defined by ${Z > 0, Z = 0}$, and ${Z < 0}$ are invariant sets of the flow ${\phi_{t}}$. Applying this notion to Eq. (11), one immediately sees that ${x_1 > 0}$, ${x_1 = 0}$, and ${x_1 < 0}$ are invariant sets of the corresponding flow. Further, there also exists a symmetry such that ${x_{1} \rightarrow -x_{1}}$, which implies that without loss of generality, we can restrict our attention to ${x_{1} \geq 0}$.

### 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:

$\displaystyle P_{1}: x_{1}^{*} = 0, \quad P_{2}: x_{1}^{*} = 1, \quad P_{3}: x_{1}^{*} = \frac{\beta - \delta}{-\alpha + \beta - \delta + \gamma}. \ \ \ \ \ (12)$

Note that, ${P_{3}}$ actually contains ${P_{1}}$ and ${P_{2}}$ as special cases. Namely, when ${\beta = \delta}$, ${P_{3} = 0 = P_{1}}$, and when ${\alpha = \gamma}$, ${P_{3} = 1 = P_{2}}$. We will therefore just analyze, the stability of ${P_{3}}$. ${P_{3} = 0}$ represents a state of the population where all teams predominantly shoot three-point shots. Similarly, ${P_{3} = 1}$ represents a state of the population where all teams predominantly shoot two-point shots, We additionally restrict

$\displaystyle 0 \leq P_{3} \leq 1 \Rightarrow 0 \leq \frac{\beta - \delta}{-\alpha + \beta - \delta + \gamma} \leq 1, \ \ \ \ \ (13)$

which implies the following conditions on the payoffs:

$\displaystyle \left[\delta < \beta \cap \gamma \leq \alpha \right] \cup \left[\delta = \beta \cap \left(\gamma < \alpha \cup \gamma > \alpha \right) \right] \cup \left[\delta > \beta \cap \gamma \leq \alpha \right]. \ \ \ \ \ (14)$

With respect to a stability analysis of ${P_{3}}$, we note the following. The point ${P_{3}}$ is a: • Local sink if: ${\{\delta < \beta\} \cap \{\gamma > \alpha\}}$, • Source if: ${\{\delta > \beta\} \cap \{\gamma < \alpha\}}$, • Saddle: if: ${\{\delta = \beta \} \cap (\gamma < \alpha -\beta + \delta \cup \gamma > \alpha - \beta + \delta)}$, or ${(\{\delta < \beta\} \cup \{\delta > \beta\}) \cap \gamma = \frac{\alpha \delta - \alpha \beta}{\delta - \beta}}$.

What this last calculation shows is that the condition $\delta = \beta$ which always corresponds to the point $x_{1}^* = 0$, which corresponds to a dominant 3-point strategy always exists as a saddle point! That is, there will NEVER be a league that dominantly adopts a three-point strategy, at best, some teams will go towards a 3-point strategy, and others will not irrespective of what the analytics people say. This also shows that a team's basketball strategy really should depend on its respective payoffs, and not current "trends". This behaviour is displayed in the following plot.



Further, the system exhibits some bifurcations as well. In the neigbourhood of ${P_{3} = 0}$, the linearized system takes the form

$\displaystyle x_{1}' = \beta - \delta. \ \ \ \ \ (15)$

Therefore, ${P_{3} = 0}$ destabilizes the system at ${\beta = \delta}$. Similarly, ${P_{3} = 1}$ destabilizes the system at ${\gamma = \alpha}$. Therefore, bifurcations of the system occur on the lines ${\gamma = \alpha}$ and ${\beta = \delta}$ 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: If ${\mathbf{x}^{*}}$ is an asymptotically stable fixed point, then the symmetric strategy pair ${[\sigma^{*}, \sigma^{*}]}$, with ${\sigma^{*} = \mathbf{x}^*}$ is a Nash equilibrium. We will primarily make use of the monotonicity principle, which says let ${\phi_{t}}$ be a flow on ${\mathbb{R}^{n}}$ with ${S}$ an invariant set. Let ${Z: S \rightarrow \mathbb{R}}$ be a ${C^{1}}$ function whose range is the interval ${(a,b)}$, where ${a \in \mathbb{R} \cup \{-\infty\}, b \in \mathbb{R} \cup \{\infty\}}$, and ${a < b}$. If ${Z}$ is decreasing on orbits in ${S}$, then for all ${\mathbf{x} \in S}$,

$\boxed{\omega(\mathbf{x}) \subseteq \left\{\mathbf{s} \in \partial S | \lim_{\mathbf{y} \rightarrow \mathbf{s}} Z(\mathbf{y}) \neq \mathbf{b}\right\}}$,

$\boxed{ \alpha(\mathbf{x}) \subseteq \left\{\mathbf{s} \in \partial S | \lim_{\mathbf{y} \rightarrow \mathbf{s}} Z(\mathbf{y}) \neq \mathbf{a}\right\}}$.

Consider the function

$\displaystyle Z_{1} = \log \left(-1 + x_{1}\right). \ \ \ \ \ (16)$

Then, we have that

$\displaystyle \dot{Z}_{1}= x_{1} \left[\delta + \beta \left(-1 + x_{1}\right) - \delta x_{1} + x_{1} \left(\gamma - \alpha\right)\right]. \ \ \ \ \ (17)$

For the invariant set ${S_1 = \{0 < x_{1} < 1\}}$, we have that ${\partial S_{1} = \{x_{1} = 0\} \cup \{x_{1} = 1\}}$. One can then immediately see that in ${S_{1}}$,

$\displaystyle \dot{Z}_{1} < 0 \Leftrightarrow \left\{\beta > \delta\right\} \cap \left\{\alpha \geq \gamma\right\}. \ \ \ \ \ (18)$

Therefore, by the monotonicity principle,

$\displaystyle \omega(\mathbf{x}) \subseteq \left\{\mathbf{x}: x_{1} = 1 \right\}. \ \ \ \ \ (19)$

Note that the conditions ${\beta > \delta}$ and ${\alpha \geq \gamma}$ correspond to ${P_{3}}$ above. In particular, for ${\alpha = \gamma}$, ${P_{3} = 1}$, 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

$\displaystyle Z_{2} = \log \left(x_{1}\right). \ \ \ \ \ (20)$

We can therefore see that

$\displaystyle \dot{Z}_{2} = \left[-1 + x_{1}\right] \left[\delta + \beta\left(-1+x_{1}\right) - \delta x_{1} + \left(-\alpha + \gamma\right) x_{1}\right]. \ \ \ \ \ (21)$

Clearly, ${\dot{Z}_{2} < 0}$ in ${S_{1}}$ if for example ${\beta = \delta}$ and ${\alpha < \gamma}$. Then, by the monotonicity principle, we obtain that

$\displaystyle \omega(\mathbf{x}) \subseteq \left\{\mathbf{x}: x_{1} = 0 \right\}. \ \ \ \ \ (22)$

Note that the conditions ${\beta = \delta}$ and ${\alpha < \gamma}$ correspond to ${P_{3}}$ above. In particular, for ${\beta = \delta}$, ${P_{3} = 0}$, 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 ${\beta > \delta}$ and ${\alpha = \gamma}$, the strategy ${[1,1]}$ is a Nash equilibrium. On the other hand, for the specific case where ${\beta = \delta}$ and ${\alpha < \gamma}$, the strategy ${[0,0]}$ is a Nash equilibrium. 5. Discussion In the previous section which describes global results, we first concluded that for the case where ${\beta > \delta}$ and ${\alpha = \gamma}$, the strategy ${[1,1]}$ is a Nash equilibrium. The relevance of this is as follows. The condition on the payoffs thus requires that

$\displaystyle \pi(T,T) = \pi(Th,T), \quad \pi(T,Th) > \pi(Th,Th). \ \ \ \ \ (23)$

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 (23) 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 ${\beta = \delta}$ and ${\alpha < \gamma}$, the strategy ${[0,0]}$ is a Nash equilibrium. The relevance of this is as follows. The condition on the payoffs thus requires that

$\displaystyle \pi(T,Th) = \pi(Th,Th), \quad \pi(T,T) < \pi(Th,T). \ \ \ \ \ (24)$

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 (24) 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 ${x_{1} = 1}$ is globally stable under the conditions in (23). 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 (24) were shown to imply that the point ${x_{1} = 0}$ 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 ${\pi(T,Th) - \pi(Th,Th) < 0}$, while it is unstable if ${\pi(T,Th) - \pi(Th,Th) \geq 0}$. In addition, a predominant two-point strategy was found to be locally stable when ${\pi(Th,T) - \pi(T,T) < 0}$, and unstable when ${\pi(Th,T) - \pi(T,T) \geq 0}$. There is also they key point of which one of these strategies has the highest probability of being executed. We know that

$\displaystyle \pi(\sigma,\mathbf{x}) = \sum_{s \in \mathbf{S}} \sum_{s' \in \mathbf{S}} p(s) x(s') \pi(s,s'). \ \ \ \ \ (25)$

That is, the payoff to a team using strategy ${\sigma}$ in a league with profile ${\mathbf{x}}$ is proportional to the probability of this team using strategy ${s \in \mathbf{S}}$. 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 ambiguous statement. In actuality, one needs to analyze what offensive strategy is optimal which is constrained by a particular set of payoffs.

## Article on Three-Point Shooting in the Modern-Day NBA

Continuing the debate of the value of three-point shooting in today’s NBA, my article analyzing this issue from a mathematical perspective has now been published on the arXiv, check it out!

## Data Analytics and The 1995-1996 Chicago Bulls

It is without question that the greatest team in NBA history was the 1995-1996 Chicago Bulls. They went 72-10 that year and went on to win the NBA Championship against a top-notch Seattle Supersonics team.

Phil Jackson’s system and first-class coaching were the major reasons why the Bulls were so good, but I wanted to analyze their reason for winning using data science methodologies.

The results that I found were very interesting. First, I mined through each individual game’s data to obtain patterns in the Bulls wins and losses, and this is what I found:

One sees that the Bulls were a defensive nightmare, and if you look at these results in detail, it makes sense that the Sonics were really the only team that ever posed a threat to them. This shows that to beat the Bulls, the opposing team would have to simultaneously:

1.  Ensure Ron Harper had a FG% less than 44.95% in a game,
2. Ensure Dennis Rodman would have less than 17 total rebounds in a game,
3. Ensure Luc Longley had less than 2 blocks in a game,
4. Ensure Michael Jordan had a FG% less than 46.55% in a game.

If any one of these conditions were not met, the Bulls would win!

This analysis on some level also dispels the notion espoused by several sports analysts like Skip Bayless of ESPN who continually claim that the Bulls’ sole reason for success was Michael Jordan. Ron Harper’s contributions although of paramount importance are rarely mentioned nowadays.

This analysis also shows that the key to the success of the Bulls was not necessarily the number of points that Jordan scored, but the incredible efficiency with which he scored them.

A boosting algorithm also allows us to deduce the most important characteristics in the Bulls’ quality of play and whether they would win or lose a game.  The results are as follows:

We see that a key feature of the Bulls’ quality of play depends on how efficient Ron Harper in terms of his FG%.

It is quite interesting that this analysis shows that winning a championship is not about one player, sure, every team needs great players, but the Bulls were a great team, consisting of many great components working together.

## Breaking Down the Knicks’ Season

Like many of my fellow Knicks fans, I am in an absolute state of shock and disappointment as the Knicks are currently 5-29 to start the new year! Many analysts from the standard outlets, ESPN, Yahoo! sports, etc… have given their share of reasons why the Knicks are playing the way they are. Being a mathematical physicist and data scientist, I decided to see if one could deduce any useful information from how the Knicks have been playing to see what is the true reason why they are losing all of these games. Here is what I found. Based on the data available at Basketball-Reference.com,  I designed an algorithm in R to go through each game, and fit regression trees (Here is a link to more on regression trees if you are unfamiliar with the concept) and found the following:

1. The number of points the Knicks score per game:

From this regression tree, we see that if the Knicks for example make less than 33.5 FG’s in a game, and have a 3-Point shooting percentage of less than 0.309, they will be expected to score no more than 79 points in a game. On the other hand, if they make more than 38.5 FG’s in a game, and also attempt more than 19 free throws in a game, they can be expected to score more than 111 points in a game.

2. The number of points the Knicks’ opponents score per game:

From this regression tree, note that first “Tm” denotes how many points the Knicks score in a game. We see that for example, if the Knicks have less than 28 defensive rebounds in a game, also score less than 98 points in a game, and have fewer than 4-5 blocks in a game, their opponents will slightly outscore them, and win the game. In fact, if the Knicks manage to get less than 28-29 defensive rebounds per game, and score less than 98 points in a game, they will be expected to lose every game they play! Now, let’s say, the Knicks do manage to get more than 28 defensive rebounds in a game, if they still only manage to score less than 89 points in a game, they are still almost guaranteed to lose as well.

Although, many analysts have probably pointed these things out, the conclusion one draws from these regression tree analyses, is that the Knicks have a significant problem with defensive rebounding, as that seems to be the number one factor in them not winning games. Further, they also have a significant problem with how many points they score per game, which is a direct result of this Knicks team still not running their offense correctly.

Would Tyson Chandler have made a difference? As the above analyses show, no single factor determines whether the Knicks win games or not. It is reasonable to assume that if Tyson Chandler was on the team, then, the Knicks would get more than 28-29 defensive rebounds in a game. But, according to the above analyses, and the right of the previous regression team, if they still as a team would attempt more than 78-79 field goals, they would still be expected to lose every game. The question then remains would Tyson Chandler’s presence increase the Knicks’ offensive efficiency? In principle, according to his career FG% stats, I would say yes. According to Basketball-Reference.com, Tyson Chandler had a FG% of 0.638 while in New York, and for his career has a FG% of 0.588, which is quite high for NBA standards. It is quite reasonable to assume therefore, that the Knicks would have considerably less FGA’s (certainly less than 78-79) in a game, and their opponents would be held to around 91.0 points per game. One would conclude that from a statistical perspective, trading away Tyson Chandler was perhaps a mistake and had an overall negative impact on the team’s performance both defensively and offensively.