Just a quick follow up to my earlier post examining some of the North American men in today’s U23 pursuit, this time for the women. I’ll just focus on three ladies: Sadie Bjornsen, Caitlin Patterson and Alysson Marshall.
Here’s the graph for Bjornsen:
Today’s race (blue) seems neither much better nor much worse than she’s performed against this group of skiers this season, although most of the data from earlier this season is comparing her to other North Americans.
Caitlin Patterson has somewhat less data for me to use, so things here are a little murkier: Read more
As before, with the U23 racers we typically have more data to work with, since these folks have generally been around longer and so are likely to have raced against each other more often. We’ll continue using the percent back difference plots, that examine how a particular skier has fared against each of their competitors in the past. So let’s get a sense for what some of the North American performances mean, starting with Alex Harvey:
Harvey was certainly a favorite to win coming into this race, given his strong results on the World Cup this season, and he didn’t disappoint. Today’s race is in blue, with the median from today’s race in red and a red trend line to give you a sense of how his performance against these particular skiers has changed (or stayed the same) over time. What this graph suggests is that his victory today wasn’t an unusually good or bad performance for him; it also suggests that the folks today that he’s raced against before he’s almost always beaten.
Moving on to Noah Hoffman: Read more
It shows the progression of WC points for the top 30 athletes or so throughout this season. Distance points are on the x axis and sprint point are on the y axis. This does not include the “general” points awarded for overall stage wins. My main complaint about the way they implemented this is that I can’t find a way to exert finer control over how the time steps proceed. This results in the appearance of WC points accruing slowly over several days in a lot of cases, when really the dots should remain perfectly still in a lot of those cases. (Or maybe I’ve just messed something up, I’ll keep working on it.) Men first and then the women. Other than requiring Flash, I’m not sure how these will react to various browsers; at the very least they may load somewhat slowly on occasion.
There are various menus and options you can play with, including the ability to select some specific people and having their dots leave a trail as they move.
My eyes are a bit bleary from grading all day, otherwise I’d try to watch some of the WJC/U23 racing action that appears to be available online. As it is, here’s what’s been cooking here over the last week:
- Wrapping up a weekend of racing on both WC circuits before they take this one off: men’s and women’s classic distance races in Otepää, the classic sprints the next day, and the men’s and women’s biathlon mass start races.
- With Marit Björgen continuing her winning ways, I took a look at how to more fairly compare her record with that of Bente Skari and Elena Välbe.
- Another recap of the WC sprint heats and some other various odds and ends that occurred to me.
- In honor of the racing action being focused in the nation of Estonia, I devoted a post to the question of whether Andrus Veerpalu really does race unusually well at big events like the Olympics and World Championships.
- A brief summary of the results for two American biathletes who’ve been skiing well lately.
- I showed a little graphical evidence of something that I’ve mentioned a few times: that the men’s heats in WC sprint races are more likely to become slower as the day goes on than the women’s heats.
- Finally, an attempt to put the races of several of the U23 racers efforts in some more helpful context.
I’ve been swamped with grading this week, so I feel a little bad about not posting anything about World Juniors specifically. I’ll try to put something together over the weekend. But these percent back difference plots are pretty easy to pump out for the U23s, since they tend to have a bit more racing behind them, and hence more data.
If you’ve forgotten what it is I’m plotting here, you can check here for an explanation. The basic idea is to compare someone’s performance (e.g. Musgrave’s) to their past performances against the particular people at that race. This provides for better, more stable, comparisons than simply plucking out a single skier to use as a benchmark.
First up, Andrew Musgrave:
This is mostly as I suspected. Musgrave’s race looks somewhat better than what he was averaging earlier this season. So, not terribly surprising that he had a good race, perhaps, but maybe more notable for how close it actually is to a typical race for him this season. It’ll be interesting to see if he can put a great one together at World Championships and maybe break into the top thirty.
Noah Hoffman commented after his race something to the effect of “Meh.” He should know best:
I’ve mentioned this from time to time in other posts, but one thing I’ve noticed from the sprint heat times that Jan at WorldOfXC.com has been providing me is that the women’s heats are more likely to increase in speed as the day progresses. I’ve said it enough that I should probably provide some evidence:
In blue I’ve shown the median value (and error bars representing the interquartile range, or middle 50%) for each heat. As you can see, the men’s heats tend to remain roughly flat, or slightly slower as the day wears on. The women, on the other hand, see more of a trend toward faster times as we progress through the quarters, semis and finals.
There could be fitness or physiological reasons for this, or it might be related to how competitive each gender’s qualification round happens to be. For instance, it may be that qualifying in the women’s sprints is enough easier (by this I mean that there is a tendency for the men’s field to be somewhat deeper) which allows many of the women to back off somewhat in qualification.
But another possibility I’d like to suggest is that as the men move into the semis and finals, they are less willing to hammer right out of the gate in an attempt to break their competitors. Instead, they’ll tend to play more of a waiting game.
Today’s U23 races in Estonia resulted in victories for some talented young skiers, Russia’s Evgeniy Belov and Finland’s Krista Lahteenmaki. In both cases their wins were fairly convincing, with Belov distancing himself from second place by 16 seconds and Lahteemaki crushed the women’s field by a full minute.
One of the difficulties in gauging performance in situations like these is that it’s hard to know how exceptional a race the winner had. Were these typical efforts for Belov and Lahteenmaki, or were they better (or worse) than usual? One way we can assess that is to look at plots of the differences in percent back between these athletes and the rest of the field, and see how today’s race compares to their past performances against these specific skiers:
The blue represents today’s race and the rest of the dots represent times when Belov has raced against some of these people in the past. Notice the short time scale here, as Belov hasn’t faced many of these racers very often prior to this season. In general, today’s race represents what looks like a typical effort for Belov against this field, compared to his other races this season.
How about Lahteenmaki’s race?