# Economists Do It With Models

## Warning: “graphic” content…

#### March 24th, 2011 · 7 CommentsBehavioral Econ · Decision Making · Sports

My parents are from Ohio, and my best friend is from Kansas (damn army brats), so March Madness is a pretty lively time around these parts. I, of course, have a bet with him regarding which team will make it further in the tournament (ex ante, this was more or less a fair bet since both teams are #1 seeds), but this whole bracket thing just usually frustrates me. I initially took the strategy of choosing whatever school I knew more people from, but given that I am of the fairly nerdy persuasion, this was probably not the best idea. (I may have realized this when I had Princeton and BU in the finals. Amusingly enough, another economist actually went through and filled out a bracket based on economics department ratings and reached the same outcome.) The opposite strategy wasn’t really appealing either, so I needed to research another option:

(Note: That book doesn’t actually have anything to do with the topic at hand, but it is the one that I was reading on the plane to Cleveland. Fun fact: it’s co-authored by one of the guys who founded Honest Tea.)

Turns out that economists and mathematicians have something to say on the matter. From the Nudge blog:

Always pick the higher seed. Never pick an upset.

The behavioral phenomena is called probability matching, explains Matthew Hutson.

Let’s say you’re drawing balls from a large box that contains 25 red balls and 75 blue balls. Many people tasked with predicting draws will predict red 25 percent of the time and blue 75 percent of the time. They match their guesses to the probabilities of the outcomes, in this case yielding a 62.5 percent success rate. But if they just guessed blue on each draw, they’d be right, on average, 75 percent of the time.

Similarly, people know there will be a certain number of upsets in each NCAA tournament, and therefore betting on a Cinderella-free tourney seems silly. The only logical thing to do, they conclude, is to figure out when those upsets will take place. The drawback, of course, is that by shooting for perfection, they end up handicapping themselves. In McCrea and Hirt’s research using real NCAA data, for example, people would have been much better off just sticking to the seedings.

One thing is for sure. You won’t have much fun filling out your bracket with this strategy.

Well it seems so obvious when put that way. Technically, this assumes that there is no information other than seed number available to help compare teams, which isn’t really the case, but for the casual participant who isn’t going to go hunt down the additional information, the no-upset strategy is the best bet. Before you start planning a boring bracket for next year, however, consider the following event that just happened the other day in Florida. From my friend’s Facebook wall:

Sean: Tonight’s Fantasy 5 lottery numbers were 14-15-16-17-18. That is all.

Sean: If a drawing has one winner, the ticket might be worth around \$250,000. An average drawing probably has three to five winners who each get around \$40,000 to \$80,000. Tonight there were 47 winners who each got \$4,581 (before taxes).

Danielle: 47 people played that random sequence? I wouldn’t think so many would. Weird.

Sean: Through the power of the Internets, I’ve discovered the Orlando Sentinel has a Florida Lottery blogger. (Really.) And he’s already written about it (and shares your surprise, D). http://blogs.orlandosentinel.com/news_lo…

Me: I am only a little surprised. I thought it was well known that you should never pick 1-2-3-4-5-6 in the lotto- even though it obviously has the same odds as every other combination, so many people pick it that it’s a bad strategy from a payout perspective. I am now curious as to why some people felt the need to start at 14.

Sean: Many more people are starting at 14 these days. Just ask the Palins.

My guess is that the no-upset strategy suffers from the same problem- because it’s the easiest and even the best in a way, you’re probably not the only one going with that strategy. Therefore, in games where where the winners share a fixed pot, it’s worth weighing the increased likelihood of winning against the lower amount likely won.

Now back to the most important part of this- go Buckeyes.

If sports aren’t your thing, here are some other nerdy brackets to consider instead.

### 7 responses so far ↓

• 1 LJC // Mar 24, 2011 at 4:05 pm

The higher-seed strategy seems as though it would beat most other brackets, but the point isn’t to beat most other brackets, the point is to beat all the other brackets. And the “pick the higher seed” strategy will rarely win, right? It may beat a randomly selected bracket that includes some upsets (that is, it may be better than if you had tried to pick some upsets), but the problem is you are up against a whole host of other brackets, not just one other bracket. So, chances are, someone is going to pick a few correct upsets, and beat this higher-seed bracket. Going back to the red/blue marbles in a box example, your “always blue” strategy will beat most of the “sometimes blue and sometimes red” strategies. But by the same token, if enough people enter the contest, there will likely be some people that guess red correctly a handful of times, and get up to 80% correct. (Consider that only once have all 4 #1 seeds reached the final four [2008], but almost certainly in every pool there is one person who gets the final four correct.)

• 2 Scott // Mar 24, 2011 at 4:57 pm

Nalebuff also co-authored a good introduction to Game Theory called The Art of Strategy with Avinash Dixit. I thought it was really good.

• 3 Matt H. // Mar 25, 2011 at 1:39 pm

To me the whole alure is that one year I’ll pick a very high percentage of the games right, win my office/local bracket pool, and walk away with some cash. Someone who picks chalk the whole way might win a pool every once in awhile, but they can’t have any claim to insight or intuition when it comes to the game. That is the real thrill of filling out a bracket for a basketball fan.
Of course, picking winners involves a great deal of luck, but like all famous forecasters one can just pretend they “knew it all along”.

• 4 econgirl // Mar 25, 2011 at 3:05 pm

You picked Butler over Pitt didn’t you. =P

• 5 John // Mar 27, 2011 at 6:00 pm

I had no idea econgirl was an Ohio State fan. That happens to be the institution which bestowed my beloved Econ degree upon me. O-H!

• 6 econgirl // Mar 27, 2011 at 6:06 pm

I-O! My cousin goes to Ohio State, and she’s awesome, so by extrapolation…

• 7 John // Mar 28, 2011 at 1:01 am

Judging by the awesomeness of this site, your own was never in question. If you haven’t already, visit your cousin for a home football game. You will not be disappointed.