Forecasting the Olympics


31 July 2012


Forecasting sporting events is a growing research area. The International Journal of Forecasting even had a special issue on sports forecasting a couple of years ago.

The London 2012 Olympics has attracted a few forecasters trying to predict medal counts, world records, etc. Here are some of the articles I’ve seen.

  1. Which Olympic records get shattered?, Nate Silver, New York Times.
  2. Statisticians predict the number of Olympic records that will fall at London 2012, Physics arXiv blog.
  3. London Olympics and a prediction for the 100m final, Markus Gesmann.
  4. Overcoming the doping legacy. Can London’s winners outperform the drugs of 1988?, Ray Stefani, California State University.
  5. Predicting the London Olympics Medal Count, Dan Graettinger, Discovery Corps.
  6. Modelling Olympic performance, PwC.
  7. The Olympics and Economics, José Ursúa and Kamakshya Trivedi from Goldman Sachs.
  8. Who will win the 2012 London Olympics?, Emily Williams, London Business School.
  9. Olympic Predictions, Daniel Johnson, Colorado College.
  10. FT consensus forecasts (combining the previous four predictions for medals), Martin Stabe.

I’m not sure how many of these people are going to do a follow-up forecast evaluation. I hope they do!  No-one should attempt to produce forecasts without also reviewing the results once the data are available to see how they did.

The FT consensus forecasts are being tracked against actual medal counts, although you will need to register (for free) to be able to see it.

There is also an interesting special issue of Significance, with a lot of statistical analysis of the Olympics.