Hawks, Doves, & Owls: Evolutionary Game Theory, Network Analysis, & the Men's NCAA Basketball Tourna
Updated: Jun 14, 2019
by Jess Behrens
© 2005-2018 Jess Behrens, All Rights Reserved
I'm going to take this rather large endeavor one small chunk at a time! Most people would begin by defining the elements of their analysis and work sequentially toward their conclusions. In the interest of attempting to make a complex, math heavy topic more readable, I'm going to do the opposite: begin with the results, gained after years of work, and head backwards into the nitty gritty details & evidence.
I'll start by saying this: based on my analysis, I don't believe that the NCAA Tournament, or any team sporting event, is what our eyes would lead us to believe. I believe that games and tournaments are not unique; that given a sport's consistent season and rule structure, nearly identical results keeps happening over and over. The team names, and the players, change, sure, but the role that a given team plays in a given year is more less identical to a team for a previous year. If you take one thing from reading my blog on this topic, this is it. Furthermore, if it's happening here, it's very likely occurring in other sports and competitive games as well.
Of course, this belief is heretical in a society that cartoonishly over values the personality, attitude, & contributions of a given star player or individual and gives them too much credit when their team wins and too much blame when that same team loses. I'm not saying that the individual isn't important in team sports. I'm saying we give the individual too much credit. Note: this ideologically driven belief in the supremacy of the individual's contribution in sport draws from a similarly absurd belief in the economic supremacy of the individual - but this isn't a political or economic blog, so that will wait for another day.
Anyone who has ever played a team sport competitively knows that the team is defined by its weakest link, and that the entire point of the team is to compensate for the player that is currently under performing, whether that is due to a random injury or because he/she just broke up with his/her girlfriend/boyfriend/significant other. Yes, this means I'm one of those annoying fans who will constantly remind the reader that even Michael Jordan, the greatest individual talent to ever play basketball, didn't win a championship until Scottie Pippen was a Bull. Sorry if that offends you, but it is true.
My belief also necessitates an acceptance that, perhaps, events not occurring on the present court or field of play on that given day will determine the winner and loser or the current contest; that teams in a league cluster into populations based on their experience over that season and that these populations can be identified. Even further, that the exact breakdown of those teams within the overall league population, the league or tournament structure, will affect, or even determine, which team wins the championship.
I searched for a way to define those populations for a long time. I even tried to invent a method for defining them. And then I stumbled back into the ecological literature (my degree is in ecology) and rediscovered Evolutionary Game Theory - which I will describe in my next post!