Tomorrow, March Madness begins in earnest with the first games in the round of 64. That means today, sports fans everywhere are feverishly filling out their brackets. Whether you know the teams inside and out or take a random-guessing approach, odds are your bracket will end up busted. Still, in every office pool, even if nobody has a perfect bracket, somebody has to win. So why not you? We asked Mark Soczek, a professor at Washington University’s Olin Business School and coordinator of the school’s bracket pool, for his tips on how to best your coworkers. (Spoiler alert: He likes SLU.)
Are there any statistical strategies you can share for filling out a bracket? One thing to keep in mind is that it’s not so much that you’re trying to get the highest possible score, but rather you’re trying to get a higher score than everyone else who enters into the contest. Someone who’s filling this out will want to think, “If Kentucky’s in the tournament, will there be 30 other people who are going to pick Kentucky? Because if I pick Kentucky too, then I won't stand out in that particular pool. Whereas if I pick maybe a No. 3 or No. 4 seed, I might be the only one to pick that team, and if they come through, then I’m going to win the contest.” So choosing these one-offs, or plausible-but-not-necessarily-the-favorite teams I think is a good strategy for a person who wants to win.
Any downsides to that strategy? The downside is that it could internalize the bad outcome if that team gets knocked off relatively early. In a contest where there’s a winner and there’s everyone else, probably from an academic perspective you’d characterize it as a higher-variant strategy, where there’s a low chance of you winning, but if your outcome happens, then you’re probably going to be up on the top.
What about special rules? Everyone has their own rules regarding points. In some contests you get bonus points for picking underdogs, but in some you don’t. Thoroughly understanding the rules that any particular contest has is going to be critically important. In a contest where there is no reward for picking underdogs, people will pick the favorites, because there’s no incentive not to. You have to stand out in a pool of a lot of people who pick the favorites.
Is there any research that outlines the best strategy for filling out a bracket? There’s some academic research, and I don’t know that you can characterize it as coming to any consensus. But a lot of people will look at power ratings and things of that nature as a basis for predicting brackets. If you’re really hardcore, you might get into simulating games and breaking down games by possession. So if you’re going to have a tip-off, what’s the percentage of each team winning the tip-off? Then you might play out each possession for an entire game, and then with computing power, simulate a game 10,000 times. One team wins 53 percent and the other team wins 47 percent, so you pick the team with 53 percent. There are definitely a lot of different theories.
Is it ever to my advantage for a pick to turn out wrong? The one thing to keep in mind is that there are an incredible number of possibilities of outcomes in a tournament, so what tends to happen is that when a long-shot wins, that breaks your brackets, but it does the same thing for most other people, too. You might have lost a Final Four team, but another person might have lost their final champion. From a relative-standing perspective, that actually might be to your benefit when you get a game wrong, if it hurts someone else more than it hurts you.
It seems like it’s extremely rare, even in large pools like ESPN that have thousands of brackets, to guess the outcome of every single game correctly. Last year, there were primarily favorites that made it to the Final Four, but 0.4 percent of their participants selected the Final Four correctly. That’s one out of every 250. Other years where there was a Butler or a George Mason or some real underdog who made it, they had two people out of 5 million get it right. So when these long shots play well, everyone’s bracket is busted. If you happened to be the one that picked George Mason, you’re on national TV, but if you lose on the first round, you get to anonymously lose.
How does filling out a bracket compare to the financial markets? You have winners in an NCAA contest, and then you have the people who are labeled as successful investors. Whether you pick the right teams or the right companies in which to invest, people love success. There’s a lot of evidence that money chases past results, and there’s a warning that past returns are not indicative of future performance. I think there’s a lot of truth to that. Successful picking could be luck or it could be skill, but it’s really hard to differentiate between the two, whether you’re in the financial arena or picking winners in an NCAA contest.
Do you think people in the financial world are especially into March Madness because of that comparison? Absolutely. In terms of participation and success in our contest, the finance professors do well. The finance arena and the gaming arena are very close.
Which teams are you picking? I’m a frequent fan of Michigan State, since I’m from Michigan. This year, I’m currently thinking Indiana, and I’m still debating New Mexico. Those are the two front-runners. I also like Saint Louis University’s chance to make it to the Final Four.
