Shakespeare’s Julius Caesar made March famous and NCAA basketball made it fun. In March, all of us college basketball gurus show the world how smart we are — or are not in most cases — by filling out our brackets for the Big Dance. Millions of us do it (even the President) and the results are a testament to just how lousy we as human beings are at forecasting anything.


Smart people like Warren Buffett know just how bad we are at forecasting, which is why Buffet was willing to offer $1 billion to anyone able to successfully pick the outcome of all 63 games.1What Warren knows that most of us bracketologists don’t is that you’ve got a 1 in 9,223,372,036,854,775,808 chance of picking a perfect bracket.2That’s 9 quintillion, 9 billion billion, or simply 9 followed by 18 digits. Of course, with odds like that nobody had a chance of winning. In fact, most were out of the running after the first day and nobody was left standing by the end of the second.3
For most, including me, the chance at a billion dollars went out the window when the Mercer Bears upset the Duke Blue Devils. What is fascinating is that pick distribution showed 97.6 percent of bracketeers thought the Blue Devils would shut down the Bears.4How could so many self-proclaimed experts have been so wrong?
To answer that question we need to turn to Philip Tetlock, Professor of Management and Psychology at Wharton.
Tetlock says one of the reasons people are so bad at forecasting is that our ability to prognosticate is limited by the complexity of the system we are trying to forecast; the more complex the system, the less accurate our forecasts.5Another reason is that forecast horizons vary. Take weather forecasts for example. They may be accurate a day or two out, but become unreliable as they go beyond that time.
The experts are not immune. Tetlock found the average expert was only slightly more accurate than a dart-throwing chimpanzee. A comparison of several Wall Street prognosticators supports his findings. The Wall Street professionals were off by an average of 21% when trying to guess what the value of the S&P 500 Index would be at the end of 2013.6Many expert prognosticators are probably better off with random guesses.
This is disconcerting since so many investors rely on Wall Street pundit predictions to make investment decisions. If you don’t believe me, just look at the popularity of investment shows such as Mad Money which makes predictions ad nauseam. The next time you hear a Wall Street pundit make a statement about which way the market is going, take it with a grain of salt. There are more Wall Street predictions out there than busted March Madness brackets this time of year and most are no better than a busted bracket in Buffett’s competition.
IRN R 14-174 (Exp 4/16)

1Ember, Sydney. $1 Billion for a Perfect NCAA Bracket, Courtesy of Warren Buffett. January 21, 2014.
2I assume that each team has an equal chance of winning. The formula is 2^63. If you know something about basketball you can increase your chances but not by much. Your exact chance picking a winner is incalculable. As Buffet said “Einstein himself could not figure out the odds.”
3Fields, Liz. Early Wipeout: Warren Buffett’s $1B NCAA Bracket Contest Has No Winner. March 21, 2014.
5Tetlock, Philip and Dan Gardner. Why Most Predictions Are So Bad. March 17, 2011.
6Ro, Sam. Wall Street: Here’s What the S&P 500 Will Do This Year. January 5, 2013