Weighing in on everything from avocados to Zimbabwe

Weighing in on everything from avocados to Zimbabwe

Racing into the New Year


posted by Leila Z. on ,

1 comment

For New Year's 2014, I decided to race Seattle's Resolution Run to test the results of having adopted a modicum of seriousness about running. A few track workouts, a few hill sessions... nevermind that I was in the midst of what would be a nasty 3-week cold (thanks, New York City subways!) and had just flown in from the heartland 11 hours earlier. A resolution is a resolution, dammit! And I had already paid the registration fee.

The race course was a bit different than originally marked on the website, so after the first unexpected turn I decided to give myself over to the course and not worry about how much more to go. It helped, too, that GC was running alongside and cheering (a little discouraging, too, since he was also wearing a substantial backpack and not looking remotely winded -- but mostly nice to see a friendly face). I felt pretty good given the cold and jet lag, but not so good in absolute terms. I was nipped by the young son of a friend of GC's about a quarter mile from the end -- we've both gotten faster in the last year, but I fear my days of being competitive with a 9-year-old are fast drawing to a close.

My main goal in the race was to improve upon my 28:55 PR in the 5K. Imagine, then, my surprise when I crossed the finish line with a watch time of 27:00!! Maybe those track workouts really do something, after all?? I had a thrilling ten second walk to the end of the corral, where GC informed me that the course was shorter than 5K. SHORT!?! We confirmed with the race director that there had been some confusion on where to put out the cones, and the course was indeed less than the advertised 5K.

I was disappointed, of course (I was sick and I should have just slept in). But when I regained my composure, GC suggested a way to estimate what my race time would have been: the magic of statistics, of course! I downloaded 2013 and 2014 results from the internet, and wrote a simple script to extract the times of the 108 people who had run both years. The following plot shows the times and the results of a simple linear regression: 


The least-squares linear regression line is shown in blue, the line of identity in red, and a Lowess smoother in black. As expected, 2014 times (on the short course) are faster than times from 2013 (when the course, by all accounts, was the correct distance). I show in green my stated RR14 time, which corresponds to an average 2013 regulation time of 29:16. That's not a PR for me, but still respectable given the circumstances. Plus, statistics carried the day, and that's victory enough for me. :)

1 comment

Leave a Reply