Occassional thoughts about orienteering
Wednesday, November 28, 2012
Another look at WOC qualifying results 2003-2012One of the things I can do with my spreadsheet of US WOC qualifying results is compare those results with what you might expect given a simple measure of each orienteer's performance. For the US runners, I've looked at each individual race and ordered the entrants based on their OUSA rankings. It makes more senses with an example.
The US entered three women in the middle distance qualifying race in 2003: Pavlina Brautigam, Erin Olafsen and Karen Williams. If you look at the OUSA rankings for 2003, you'll see that Pavlina was the best ranked of that group, followed by Erin. I put the runners in a rank order:
You can then look at how they placed in the WOC qualifying races.
Pavlina placed 22nd
Erin placed 18th
Karen placed 28th
There's not a perfect relationship between the rank order and the places. You wouldn't expect it to work out that way always, but you might expect that over the longer term the better ranked orienteers would tend to have better WOC races.
I looked at all 10 years of WOC data for the US (2003-2012) and created graphs the graphs below. The first set of graphs illustrate the places for the best ranked US runner , the second set of graphs the next best ranked US runner, and the last set of the graphs the 3rd best US runner.
The graphs show you that generally, the better US runners have better WOC qualifying races. But, there is a lot of overlap. But the overlap suggests that you shouldn't be too surprised if the 2nd or 3rd ranked US runner beats the 1st ranked.
If you are trying to forecast US WOC results in qualifying races, you might start by assuming that all three runners will have an average qualifying place. But then you might give some "extra credit" to the best ranked of the three and forecast that they'll finish a little better than the average. You might make a similar adjustment to the worst ranked of the three.
That is more-or-less what the SPROUT method for forecasting WOC results does.
I'm guessing no one has read this far...but if you have, here's a couple of other little graphics based on my US and Canada WOC spreadsheet. These word clouds show the names of each WOC runner from 2003-2012 with the size of the text related to the number of qualifying races they ran.
Back to okansas.blogspot.com. posted by Michael | 5:52 PM
Sunday, November 25, 2012
Looking at US and Canada WOC qualifying results (2003-2012)I came up with a really simple way to try to forecast the U.S. WOC results last summer and called it SPROUT. I've been planning to improve the system a bit, mostly by adding 2012 results and by more carefully collecting some historic data. This weekend, I spent some time putting together a spreadsheet with all of the U.S. WOC qualifying race results from 2003-2012.* In response to a tweet, I also added Canadian results.
I have some ideas about how to refine SPROUT, but before I do anything with that, I thought it would be interesting to share some of the basic information.
The first graph shows shows Canada and the U.S. qualifying race places across 10 years of WOC qualifying races.
The second graph shows the same information, but I've added some straight trend lines.
One of the things that you notice looking at both of the graphs is that the difference between Canada and the U.S. is pretty small. Another thing you'll notice is that the women have done better than the men. Neither of those things is a surprise if you've been following North Americans at the World Champs.
I also took a look at how the nations have done by discipline. In the graph below you can see results for the long (lq), middle (mq) and sprint (sq) qualifying races.
Now it is time to get away from the computer and head out into the forest.
* I used to sources for WOC results - the World of O' WOC team overviews and Maprunner's WOC results database.
Back to okansas.blogspot.com. posted by Michael | 11:09 AM
Thursday, November 22, 2012
Adding some more training hours to the graph of training hours
I updated my graph of the relationship between total training hours and orienteering training hours. I added 7 of the North American orienteers and their hours. I also changed the line from a straight line to a "loess" smoothing. The North Americans that I've added include five who live in Europe.
Back to okansas.blogspot.com. posted by Michael | 10:05 AM
Saturday, November 17, 2012
Looking at some training hours for elite orienteersThe Norwegian O' Federation web page listed some training data for a bunch of elite orienteers and elite junior orienteers. Most, but not all, are Norwegians. I spent a few minutes looking at the numbers for total physical training and orienteering technique training.
Here's a simple graphic showing the relationship between total hours of physical training and total hours of orienteering technique.
One of the juniors was sick for almost half of the year and had 103 hours of physical training. The people with the most physical training hours are ski orienteers (and I left out one of the ski orienteers who didn't have much data due to a PC crash).
Two people are extreme outliers in terms of hours of orienteering technique training - Fredric Tranchand and Johan Runneson.
Back to okansas.blogspot.com. posted by Michael | 10:31 AM