Speed Rating Message (October 20, 2008)

I receive e-mails every week complaining about the speed ratings being too low at one or more races ... This past weekend was exceptionally bad (more than two dozen e-mails, some from coaches) ... So once again, I decided to give a brief overview of speed ratings (what they are) which addresses most complaints.

For starters ... I do not derive and post speed ratings for any purpose other than one purpose ... I want the ratings for myself so I can use them ... I began making speed ratings public in 2000 because I thought a few local people would be interested (in retrospect, I would NOT do it again) ... If I was not posting speed ratings, I would still be doing it for myself by the same method ... Making ratings too low or too high hurts my purpose, so I do not do it (unless it happens unintentionally).

THIS IS IMPORTANT ... Speed Ratings are determined by runners, NOT race courses ... Speed ratings are a comparison of how fast runners finish a race relative to each other, and NOT a comparison of how fast they ran a specific course in previous races ... Speed ratings are NOT track & field.

Here is a made-up example that illustrates the basic idea behind speed ratings that seems to be causing the complaints ... A large group of runners race at Jamesville Beach ... The following weekend, the same group of runners race again at Jamesville Beach at all of their times are 30 seconds faster than the previous weekend ... How much do their speed ratings go up??

The correct answer is Zero ... Speed ratings compare runner-to-runner ... In this example, the relative position of the the runners stay exactly the same ... Nothing changed runner-to-runner ... The fact that they ran 30 seconds faster does NOT matter since they all did it (and NO, they all didn't magically improve 30 seconds in ability)... The course the race was run on does NOT matter ... What matters is "How fast they ran relative to each other".

Some people do NOT like this concept, some do not understand it ... But it is the basis behind speed ratings for horses and my adapted speed rating method for XC ... It allows runners at different race courses to be compared in terms of speed relative to each other.

I derive speed ratings to be as accurate as possible within the scope of their purpose .... There is always a margin-of-error which I do my best to minimize ... Speed Ratings are NOT meant to be "Precise" because that is not possible (a point or two up or down is acceptably accurate under any situation ... and sometimes a wider margin-of-error is acceptable because that's the best statistics can do).

This past weekend saw excellent race conditions at many races ... this resulted in faster times for many runners ... Faster times do NOT mean higher speed ratings ... Running faster relative to the competition does mean higher ratings.