Usain Bolt says he can run the hundred meters in 9.4 seconds in London. The triple Olympic champion has set himself the goal of making that time at the forthcoming Olympic Games. Back in the summer of 2009, Tilburg econometricians John Einmahl and Sander Smeets used extreme-value theory to calculate that the ultimate record for men was 9.51 seconds, but now the researchers have shifted that down to 9.36 seconds.     

The Jamaican Usain Bolt has long held the world record for the hundred-meter sprint, which now stands at 9.58 seconds. A few days after the 2009 prediction, Bolt improved his own world record by 0.11 seconds from 9.69 to 9.58. However, only about 25% of the change in the predicted ultimate record is due to that phenomenal performance. The new times of other good athletes have also contributed to the shift. In their article ‘Ultimate 100m world records through extreme-value theory’, published in Statistica Neerlandica in 2011, Smeets and Einmahl apply the extreme value theory to calculate how much more time can be shaved off the world record.

Male sprinters may run as fast as 9.36 seconds in the near future, the econometricians estimate. That is 0.22 seconds faster than Usain Bolt’s current world record. The 9.36 seconds is a pure statistical estimation. The true value may deviate, but with a confidence level of 95% the ultimate record is above 8.98 seconds.

The econometricians analyzed the personal best times of the world’s 1034 best male athletes between January 1991 and July 2012. Each athlete was only included in the analysis once. Times from before 1991 were not included because of the inadequate doping controls in place at that time. The times of the men ranged from 9.58 to 10.30 seconds.

Extreme values
Extreme value theory is a branch of statistics that answers questions about extreme events (which are, by definition, unusual) on the basis of information about less extreme events. The theory is widely used in finance and insurance to estimate risks of extreme damage caused by storms, earthquakes or breached flood defenses.

Sander Smeets graduated from Tilburg University in finance and actuarial sciences in 2008 and now works as an actuarial advisor at AZL in Heerlen. John Einmahl is Professor of Statistics at the Department of Econometrics and Operations Research and a research fellow at CentER. 

Note for editors
For more information, please contact Sander Smeets (+31 6 53 67 69 43, Sander.Smeets@azl.eu) or Margreet Punt (+31 13 466 3765 or +31 6 51399 804, margreet.punt@tilburguniversity.edu).