14 July 2010

Sport as Academic Laboratory

My latest column for Bridges is out, and in it I discuss how sport can serve as a useful laboratory for examining questions related to decision making, ethics, politics, prediction and more. I even discuss the outcome of the recent World Cup prediction competition that I ran here over the past month. Here is an excerpt from my essay:
Sports provide a valuable context for evaluating expertise, and not just among athletes but among those who purport to understand the dynamics of sporting events. For instance, ESPN, the US-based sports media enterprise, hosted a competition for predictions of the outcomes of the 2010 World Cup. Of the more than 1,000,000 entries submitted, only 10 percent would have improved on naïve predictions based on the transfer market-value of each team, i.e., assuming that the higher valued team would win each game. In fact, the "expert" predictions offered by the financial services firms Goldman Sachs, JP Morgan, and UBS fell only at the 61st, 67th, and 35th percentiles in the ESPN competition, respectively, all behind a naïve forecast based on FIFA World Rankings, which scored at the 70th percentile. What might this say about these firms' ability to predict market outcomes?
Have a look, I welcome your comments. You can also hear the column as an mp3.

And as usual, please peruse the entire issue.