Statistical Skier

Visualizing split times is trickier than it seems. You have data from an array of different situations that arguably aren’t terribly comparable. We all know that mass start and pursuit races play out very differently than interval start races. It may also be difficult to compare split times from a Tour de Ski prologue to a 50k. Add to all this the fact that each race will record their splits at different points in the race, even for races of the same length, and you’ve got quite a jumble of information. (Once again, I’m using split time data kindly supplied by Jan at WorldOfXC.com.)

Still, we can do some stuff with this:

This graph includes a wide range of race lengths and types, and as you can see I only have 1-2 races for some skiers. I’ve scaled the x axis to represent the proportion of the way through the race for each split time measurement so that races of different lengths can be compared. The y axis is a bit more complicated. I’ve calculated the percent behind the leader (for each split time) and then centered and scaled these values for each race for each skier. This helps make different types of races (mass start, interval start, etc.) more easily comparable. The important thing to remember here is that this means the y axis values are only relevant for each skier, so do not compare the values between skiers. (Noah Hoffman isn’t starting races faster than Kris Freeman.) Scaling helps us compare trends between athletes, not the absolute values themselves. The y axis values are speed relative to that athlete in that race.

Short races, particularly stage race prologues can lead to some strange split times that bounce around a lot early on. More generally, the interval split times can display weird fluctuations when the distance travelled between splits is quite short. And several skiers have too few races to discern any “trend”.

Still, some interesting trends can be seen. Kris Freeman has a slight “bathtub” shape to his races, where he starts fairly slowly, picks up the pace and then fades a bit towards the end. Hoffman and Elliot both tend to slow considerably during races (which, I might add, is fairly normal). The correct interpretation here is that both skiers have a strong tendency to slip further behind the leader as the race progresses. Lars Flora did this even more dramatically, although my understanding is that he had some health issues during his European stint.

A good cautionary lesson in jumping to conclusions about whether this is good or bad can be found in Kikkan Randall’s graph. While not a distance specialist, she had some strong distance races this year. She fades as well, but mostly in mass start and pursuit races. This is common across many skiers, as when the field loses contact with the leaders, the pace will often slow considerably as folks begin to race primarily against the people near them, rather than the leaders.

Liz Stephen, however, is an interesting case. There’s one obviously “bad” mass start race, but generally speaking her other races saw her gaining time on the leaders as the race wore on. Morgan Arritola displays a more ambiguous pattern, sometimes gaining time and sometimes losing time.

I’ll be back with similar graphs for the Canadians, and also for some notable Europeans.

[ad#AdSenseBanner]