Posted by: Ian | June 8, 2012

Cleaned up results for the Tour of Wessex 2012

The results from this year’s Wessex came out a couple of days ago. While the data in the spreadsheet provided is mostly undoubtedly right (and thanks to the organisers for the xls!), there are a few “corrections” that I’ve made for my own benefit and pass on here.

My first cue that not all of the results are reliable came from Paula, who drove out to see us at the Crowcombe feed station on Day 3. While waiting at the bottom of the hill, she directed several tired riders, at their request, onto the A358 in order to bypass the entire Exmoor section; at least some of these riders appear on the results sheet. To be clear, I have absolutely no issue with this: it’s a sportive, for fun, and not a race, and everyone should prioritise their own enjoyment. Personally, on the one occasion when I had to take a short cut on a sportive myself I told the organisers about it and didn’t ride over the finish mat – but I don’t feel that there’s a major moral point at stake.

Having said that, I wanted to weed out the people who didn’t ride the same miles that I did from my results and here’s what I did…

On Day 1 the average speeds of the riders who completed all three days, ordered by their overall three day ride speed, is as follows:

The blue dots look fishy to me so those riders got removed from my dataset. To be fair, the one who finished in 300th place, and maybe one or two others, may have had a great Day 1 and then endured multiple punctures on both Days 2 and 3. But this is marginal and I’m pretty sure that most of the exclusions are appropriate.

Likewise, on the basis of the Day 2 results I removed the two outliers shown (my apologies for reversing the colour scheme):

And on Day 3, when you’d expect more short cuts to be taken, I again removed the outliers:

(If that guy really did average over 35 mph on Day 3 I take my hat off to him, but it’s a big hike from his average Day 1 speed of 14.3 mph.)

I’m aware, of course, that all of these average speeds are affected by the time spent at the feed stations. Over the three days, Joerg and I spent more than 1 hr 40 mins at the feed stations, which reduced our average speed by about 1.5 mph, on a pure arithmetic basis and ignoring the restorative power of the stops. As a benchmark, the elapsed times, as shown on the official sheet, are thus more useful. For the non-fishy finishers the average elapsed times were 6hrs36, 7hrs30 and 7hrs41 on Days 1, 2 and 3, respectively, to the nearest minute.

The result of the exclusions is to reduce the number of three day finishers from 368 to 351. It really makes no difference to Joerg and me, since we were resolutely mid pack. (We did better on Day 1, best on Day 2 and least well on Day 3, relative to the field.) The exclusions have much more impact on the top of the pack, promoting the guy who finished fifth in the official results to first in mine. You can think of the procedure, perhaps, as statistical dope testing for recreational riders.

Advertisements

Responses

  1. Can you post the .xls file if you still have it?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: