Skip to main content



Game Theory in the NBA – a Game of Chicken

Article referenced: http://arturoarturo.com/nba-lottery/

An ongoing issue in the NBA is tanking, or purposely losing games at the end of the season in hopes of securing a better draft pick. This issue exists as well in other sports, but for many reasons it is easier to study in the NBA. The article examines the incentives of different teams each season, depending on their place in the standings towards the end of the season. In the NBA, there are 15 teams in each conference and the top 8 teams in each conference. This means that 16/30 (more than half the teams) make the playoffs. The article assumes that teams at the top of the standings will always try to win a championship and “play hard” throughout the season, and that teams at the bottom of the standings will tank to try to finish lower in the standings to secure a better draft pick (draft pick is order is randomized, but skewed towards teams with worse records). What is most interesting is the teams towards the middle of the standings and this is what the article analyzes. Ultimately, the author concludes that statistics show that it is actually better for teams in the middle of the standings to tank and lose at the end of the season (if two teams are fighting for a playoff spot, it is best if they adopt opposite strategies, and the worst scenario is if the both attempt to win, see diagram below). If you are a team ranked between 7-10 in your conference, it is statistically better in the long run to attempt to lose and miss the playoffs and secure a better pick. Although finishing in the top 8 (somewhere between 6 and 8th place) and making the playoffs gives the appearance of success – statistical data shows that these teams have virtually no chance of winning the championship and the glory that goes along with it. Furthermore, teams in these middle spots will often try to trade for players to make a “playoff push” and overpay for these players, hurting them even more. The author sites many reasons why most teams abandon this “optimal strategy” and seem to play irrationally and try to make the playoffs every year. First, the owners of these teams, with ultimate control, are billionaires and don’t need to act rationally to attempt to maximize their value of the outcome of each season. They much prefer to have a slim chance at ultimate glory, a championship every year. Second, although in reality making the playoffs at the #8 seed doesn’t matter much, it changes the perception of fans and keeps them happy. There is much more analysis explaining why this irrational behavior continues to happen.

For me, this article highlights just how difficult a game of perfect information can be to analyze. We know exactly what the outcomes are for each strategy employed by NBA teams, yet it is still difficult to explain why teams are not tanking. Furthermore, it is evident that to fully comprehend this game we must examine it from many different points of view. How will the fans, players, front office, and owner benefit and react to each strategy adopted? What is most interesting is that in most scenarios, both teams fighting for a playoff spot will attempt to win, which is actually the worst possible strategy, when payoffs are based purely on chances of winning a championship. You would think that the NBA would need to put rules in place to prevent middle ranked teams from tanking, yet the teams actually do this on their own. When you factor in things like pride, and making your fan-base happy we get outcomes that at first seem counter intuitive.  If we can adapt the payoff matrix to account for these “intangibles” then the players decisions will actually be rational. Perhaps we don’t always need rules to prevent people from adopting a strategy that is optimal for themselves (NBA teams) but sub-optimal for others (the fans, NBA ratings). However, ultimately these teams are left with sub-optimal outcomes so the question we may want to ask is: should we be incentivizing these players who are already acting “in the spirit of the game” with suboptimal outcomes, particularly to stop the few teams that do not act this way? This can be applied to many social scenarios such as pollution, employment and more. Should we incentivize people who are working at extremely low salaries for not purposely being unemployed? Should we incentivize companies that are already particularly good at limiting pollution? These questions and more can perhaps be answered by examining this case study.

Comments

Leave a Reply

Blogging Calendar

September 2014
M T W T F S S
1234567
891011121314
15161718192021
22232425262728
2930  

Archives