Nate Silver, the statistical genius behind the New York Times columns and website that predicts election results with spooky accuracy, recently crashed and burned looking at the NFL. He predicted that this year, the Seahawks and the Patriots would be heading to New Orleans. Clearly, we know that’s not the case now. While cold hard statistics work so well for various predictions, past performance is often an indicator of present success. Why then do sports predictors get it wrong so often?
The websites that sell weekly sports picks, for those who like to gamble on the outcome of games shout about their accuracy. In reality, when subjected to a cold, hard analysis of these picks, it shows an accuracy of less than 50%. That means that you would be better off flipping a coin in many cases than turning to so called experts for the result of a game. The world of sports has more high profile analysts than ever before, claiming what they are sure will be the correct result are going to be. Truth is, predictions are no better than random guess. So why do we listen? In many cases, it’s a numbers game. Team X have beaten team Y, 7 times out of 10 in recent years. Surely the odds have to be slanted towards team X, right? The vagaries of sport seem to confound the experts year in year out but we still listen. We still make our own predictions because it helps the narrative we weave around the event, even when we consistently get proven incorrect. Now, knowing that, care to make a wager?