Advertisement

SKIP ADVERTISEMENT

Choosing Wisely

What Your N.C.A.A. Brackets Teach About Investing and Bias

The main draw of the N.C.A.A. men’s basketball tournament begins Thursday, and the New York Times bracket is a fun way to test your predictions in a system that, like financial markets and most forms of sports betting, rewards you for taking an against-the-grain pick that proves accurate.

It also, along the way, teaches some important lessons about overcoming biases to make smart decisions in investing, and in life.

We were curious how the expectations of Times readers who entered a bracket differed from other types of projections about tournament outcomes. In particular, how do our readers’ forecasts compare with purely quantitative measures of the likelihood of tournament success calculated by FiveThirtyEight, Jeff Sagarin and Ken Pomeroy?

The good news is that all of these sources of tournament projections present broadly similar forecasts. There are no major disagreements between them on which teams are most likely to win it all (Kentucky. Definitely Kentucky.) and which have little chance of winning even a single game (sorry, Coastal Carolina).

But the divergences that do exist are a fascinating study in cognitive bias, in particularly the “familiarity heuristic,” or a tendency to overweight the value of that which is familiar to us. It is why you are more likely to reach for the brand of toilet paper you’ve bought in the past rather than do a careful analysis of which offers the best quality for the cost, or to stay at a hotel just because you recognize the brand name.

Consider some of the teams that Times readers overweight relative to the teams’ purely quantitative measures (the numbers are from Tuesday evening, and may have shifted by the time you read this). They tend to be one or more of the following: 1) based in the Northeast, where there is the heaviest concentration of Times readers; 2) academically elite or 3) teams that have had either recent high-profile success in the N.C.A.A. tournament or a long track record of excellence.

Image
Butler, with players like Gordon Hayward,  controlling a rebound, imprinted itself in the minds of many sports fans by going on an improbable run in the 2010 N.C.A.A. tournament, losing to Duke in the final.Credit...Jonathan Daniel/Getty Images

Here are a few of them:

Butler (recent N.C.A.A. success)

Percentage of NYT readers picking to win Round of 64 game: 58 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 48.3 percent

Michigan State (long record of N.C.A.A. success)

Percentage of NYT readers picking to win Round of 64 game: 76 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 64.3 percent

Davidson (academically elite, recent N.C.A.A. success)

Percentage of NYT readers picking to win Round of 64 game: 49 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 43 percent

St. John’s (in Northeast)

Percentage of NYT readers picking to win Round of 64 game: 50 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 44.3 percent

Harvard (academically elite)

Percentage of NYT readers picking to win Round of 64 game: 19 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 16 percent

U.C.L.A. (history of N.C.A.A. success)

Percentage of NYT readers picking to win Round of 64 game 44 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 38 percent

Meanwhile, the teams whose likelihood of success appears to be undervalued by Times readers relative to quantitative projections are the opposite of that: colleges from outside the Northeast without a track record of success in the tournament that are also not very academically selective. The list weights heavily toward non-flagship state universities.

Here are three examples:

Wichita State

Percentage of NYT readers picking to win Round of 64 game: 56 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 68.5 percent

Oklahoma State

Percentage of NYT readers picking to win Round of 64 game: 48 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 56 percent

Georgia State

Percentage of NYT readers picking to win Round of 64 game: 15 percent

Average of FiveThirtyEight, Sagarin and Pomeroy: 21.7 percent

It’s easy to see how the familiarity heuristic is at work here. Rather than forecasting basketball success with methodical analysis using all the statistical information at our disposal, entrants to our bracket are letting their judgments be influenced by the colleges they are familiar with because they or a friend went there, the college is near where they live, or they recall that team having tournament success in the past.

The good news is that this opens up a market opportunity for people who think about things more analytically, and thus make predictions that are undervalued by the marketplace.

This has a close analogue in investing. It is easy to invest in flashy companies that are widely known and whose products you use. But often the highest-return investments are “value stocks,” companies that are more obscure and less popular. In this metaphor, Butler, Harvard and U.C.L.A. are the Google, Facebook and Apple of the N.C.A.A. tournament. Wichita State and Oklahoma State are the obscure industrial companies that aren’t talked about on the financial pages very often but offer high potential returns.

There’s a big difference between playing an N.C.A.A. bracket game and investing, of course. The former is for amusement, the latter to build wealth. So if people want to be a little bit irrational with their tournament picks, there’s nothing wrong with that. (I am guilty of this myself; my bracket predicts stronger-than-consensus performances by Georgetown, whose games I attend sometimes, and Wofford, a college located in my hometown.)

That’s all fun and games. But when saving for retirement, the consequences of letting familiarity, rather than clearheaded analysis, guide your decisions can be quite a bit more costly.

The Upshot provides news, analysis and graphics about politics, policy and everyday life. Follow us on Facebook and Twitter. Sign up for our weekly newsletter.

A version of this article appears in print on  , Section A, Page 3 of the New York edition with the headline: What Bracket Picks Tell Us About Investor Bias. Order Reprints | Today’s Paper | Subscribe

Advertisement

SKIP ADVERTISEMENT