Tuesday, April 22, 2008

Trail Mix recap

The trail racing season has officially kicked off with the Trail Mix 25K and 50K ultra. It was a muddier affair than last year (my only reference point, as that was my first Trail Mix.) Shortened this year due to icy conditions on the ski hill, the trail was without its toughest segment. Despite the mud, it would prove to be quite a bit faster without this part. I look to early season races as a measure of my improving fitness. I think a lot of folks do. It's very hard to use a trail race - especially one as hilly as Trail Mix - to determine fitness unless you have done the race before and conditions are similar. I have tried to calibrate the times for Trail Mix 25K this year and estimate some sort of equivalency to a full course, with modest success. Let me tell you how.

I had heard the course was about 1/3 of a mile short. I calculated ~ .8 of a mile based upon a rough mapping of the missing segment. That would yield a 'true' course time for me of ~ 5 to 6 minutes slower, but I have reason to believe that this is ambitious. I wanted to see where this race stacked up in general against previous editions. When I checked on archived results, I noticed that the depth of the race has grown considerably in the last few years, so I decided to compare strictly to last year. What I did was to compare time differentials for each runner by place from this year's race to last year's.

I calculated the difference in finish times for each place and graphed the resulting 'improvement' of this year over last year, reasoning that - in the main - the body of runners should run roughly the same times. This yielded the following observation: outside of the first dozen or so runners (who could be considered statistical outliers due simply to the difference in quality of runners that a race of this size sees at the pointy end from one year to the next), the front end of the race saw time improvements in the 7 to 8 (or 9) minute range over last year. This is considerably faster than the 5 to 6 minutes improvement I figured based upon just the shortened course. So, if I were honest, I would assume a full course time roughly 7 and 1/2 minutes slower than clock time. This is harder to swallow, frankly. Apparently, the Ski Hill is a real course discriminator!

One interesting thing that the graph shows after the first hundred or so runners: the finish time differences steadily shrink until they approach zero toward the end of the race. Now, why would that be?

1 comment:

SteveQ said...

The way I compared the years was to use only runners who did both races and who finished in the same order (to eliminate those who had one bad race). Dividing the 2008 times by the 2007 times and multiplying by 31.0 miles gave 29.0 miles almost every time. Later, on the Trail Mix website, they posted what they thought were the distance for the relay, the 25K and the 50K - they don't add up, unless their 28.0 for the 50K is a typo.