Statistical Skier

If you’ve ever wondered why I haven’t attempted more statistical machinations to predict race results, today’s races are an example of why I haven’t.  Actually, I’ve spent a fair bit of time trying to build some models that will predict performance, but what I’ve learned is that skiers are too variable for reasonably simple models to make interesting predictions.  For instance, no one wil care much if I claim to be able to predict a skier’s finishing place in a WC to within 15 places; a prediction of 15th just means I’d be certain they’re going to finish in the top thirty.

Anyhow, with today’s mass start pursuit race, the points are once again unusually low, so I’ve adjusted them for a somewhat clearer picture of how today’s race compares to each athlete’s past.  Once again, I’m using the method outlined here and here.  So keep in mind that the actual race points I’m plotting here are generally inflated compared to what you’ll find in the official results.  But all of the points I’ve used have been adjusted, so the comparisons within each athlete are reasonably fair.

[ad#AdSenseBanner]