I think everyone who has ever played, watched, coached, or in any other way been involved with basketball will tell you that it is in a team's best interest if its players are unselfish. But from a player's perspective, is that the case? Is there an incentive for each individual player to play unselfish basketball?
The coach may want each player to play for the team and focus on doing whatever is most likely to help the team win, but as I am hopefully about to show, a player's incentives make it possible, if not likely, that the player won't play unselfish, team-first basketball.
Before we dive in, I should clarify what I mean by "unselfish" and "selfish."
A player who is playing unselfishly is putting the team's interests before his own. This means that a player is using his specific abilities for the good of the team rather than his own good. Often in the context of basketball, when we describe a player as unselfish we are describing a player who doesn't score much and is always "doing the dirty work." This is not necessarily what I mean when I use the word unselfish. A scorer can be unselfish too. For example, a player who is a talented shot-creator and scorer would be acting in the team's best interests if he shoots more often than a player who is not a talented scorer. Similarly, a player who is a terrible shooter is acting in the team's best interests when he doesn't shoot as much relative to the better shooters on his team. Another example of playing unselfishly is a player who accepts the role of defensive stopper even though he is not a great defender because relative to his teammates and their abilities he is the best candidate for the job.
A player is playing selfishly when he is looking out for his own interests ahead of the team's. In the NBA it has been shown that scoring average is the biggest determinant of salary (The Wages of Wins, Stumbling On Wins). In general, the more you score, the more you get paid. If this is the case, and a player wants to maximize his salary, then the player has an incentive to score more even if it is at the expense of his team.
Now that we know what is meant by "unselfish" and "selfish," let's look at the tool that I will use to help prove my point.
Game theory provides us with a way of representing strategic situations, or games, called a matrix. A matrix like the one we are going to use can show us the different options from which players have to choose and what the payoffs are for each of those options depending on what another player chooses.
In our simple model, there are two players, each of whom has two options. They can either play selfishly or unselfishly. Then, depending on what they each choose, they receive certain payoffs. These payoffs reflect the incentives that a player has. A player has a greater incentive to choose a over b if a's payoff is greater than b's payoff. Obviously actual situations involve many more players and other factors, but I think you will see that the model makes sense.
The top right cell in the following matrix tells us that if Player 1 plays unselfishly and Player 2 plays selfishly, Player 1 would receive a payoff of -1 while Player 2 would receive a payoff of 3.
Basketball Matrix: Incentives from an Individual's Perspective
| Player 2 | |||
|---|---|---|---|
| Unselfishness | Selfishness | ||
| Player 1 | Unselfishness | 2, 2 | -1, 3 |
| Selfishness | 3, -1 | 0, 0 | |
This matrix tells us the following:
- When both players choose to play unselfishly they each receive a payoff of 2.
- When one player chooses to play unselfishly while the other player chooses to play selfishly (like in cells 2 and 3), the player who chose to play selfishly is rewarded (payoff of 3) while the player who played unselfishly is punished (payoff of -1).
- When both players choose to play selfishly they each receive a payoff of 0.
I think we can agree that the payoffs are reasonable. To make sense of these payoffs let's think about what we would do if we are Player 1. Being Player 1, if Player 2 plays unselfishly we can help our individual cause (stats, contracts, etc) more by playing selfishly than unselfishly (hence payoff of 3 > payoff of 2). If you know that your teammates are going to be playing unselfishly, and you want to maximize your personal payoffs, then you will play selfishly. Let's think about this is terms of basketball. For example, if you know that your teammates will all play their roles and not take more shots than they should, then there is an opportunity for you to pad your scoring stats and therefore increase your prestige and possibly salary.
Being Player 1, if Player 2 plays selfishly our best option in terms of our individual cause (stats, contracts, etc) is also playing selfishly (payoff of 0 > payoff of -1). If you know that your teammates are going to be stepping outside of their roles and doing what is in their best interest, then it is also in your best interest to step outside of your role and do what is in your best interest. For example, if everyone on your team is trying to pad their scoring stats by taking a lot of shots, then by playing unselfishly (rebounding, passing) you are only allowing them to overshadow you and gain all the attention and eventually the contract money (hence the -1 payoff). Although it doesn't really benefit your cause (payoff of 0), playing selfishly when your teammates are playing selfishly will at least ensure that you don't lose attention and prestige.
So, no matter what your counterpart chooses, it is in your best interest to play selfishly. If he chooses to play selfishly then you should also choose to play selfishly (0 > -1) and if he chooses to play unselfishly then you should choose to play selfishly (3 > 2). In game theory, playing selfishly would be called a strictly dominating strategy. And as you can see, playing selfishly is a strictly dominating strategy for both players. So the likely outcome of this game would be both players playing selfishly and receiving payoffs of 0. Clearly this is not the optimal situation. If they had both played unselfishly they would have both received payoffs of 2.
Now by this point you might be a little skeptical. Shouldn't you receive a higher payoff for paying unselfishly no matter what your teammate does? Wouldn't coaches and scouts notice that you are playing unselfish, team-first basketball despite your teammate playing selfishly and therefore value you higher, leading to a higher salary? You would think so, wouldn't you?
And you would be right...about the first part. Most coaches and scouts in the NBA are good enough to appropriately value players who play unselfishly. Unfortunately it's not the coaches or scouts that really matter when it comes to correctly assigning dollar figures to players. It's the General Managers...but not completely. The GM's might be managing the team, but they are influenced to varying degrees by the owners. The owners are the ones that sign the checks after all. And what are owners interested in? Well, clearly it varies somewhat by owner, but most are interested in two things: the bottom line and winning.
How do owners improve their bottom line? Well, I can't give you numbers to prove it, but it seems to me that putting butts in seats is probably the best way to improve your bottom line. And how do you put butts in seats? By winning. More than anything else, including having a "star" or scoring a lot of points, winning puts butts in seats. So basically an owner's goal should be winning and winning because winning is the best way to improve the bottom line and to win. And if a owner's goals are winning and winning wouldn't he want a team full of unselfish, team-first players?
You would think so. But apparently many owners don't seem to quite get this. They, like most people, value scorers more highly than they should (remember that scoring average is the biggest determinant of salary). This is understandable because scorers are the ones who seem to be having the most direct influence on the outcome of the game, but with each passing day the advanced stats movement is further and further disproving this idea. And this is where the problem of players having an incentive to play selfishly originates.
Players, whether consciously or not, know that they have an incentive to play selfishly, so over time they are likely to respond to those incentives and play selfishly. If everyone (fans, owners, GMs, coaches, scouts, players) correctly valued players (i.e. paid less attention to scoring and more attention to everything else) then players would have an incentive to play unselfish, team-first basketball. Eventually players would catch on and be more likely to play unselfish, team-first basketball. This would improve the quality of play across the league which would benefit everyone.
Hopefully what I have shown so far is that the scoring-centric basketball culture that has developed in the NBA is hurting the league and everyone who is associated with it. Players have incentives to play selfishly and in doing so are harming their teams.
So what is the solution?
Well like any cultural problem, there is no quick fix. But I do think that we can make slow progress by continuing to advance the advanced stats movement. The more truth about the NBA game that stat geeks can uncover and the better they can communicate that to the public, the more we can begin to change the perception that scoring trumps all.
So whether the only stats you've ever heard of are the ones in the box score or you invented your own version of adjusted plus-minus, if you want better basketball (and who doesn't?!), you'll give the NBA stat geeks of the world a little more respect.
(Note: I would be remiss if I didn't mention David J. Berri, Martin B. Schmidt, and Stacey L. Brook's The Wages of Wins which discussed many of these same ideas and originally got me thinking about them.)
(Note 12/8/10: I just read Chapter 2 of Stumbling on Wins and I realize that there is quite a bit of overlap between what I discuss here and what is written there.)
No comments:
Post a Comment