"3-- Ragozin and I started in the same place with a wind formula, which was to have someone do it on paper (in my case a friend who is a meteorologist). In theory, there is a square involved in the formula, which makes the effect of higher winds on final time very dramatic-- the impact of a 10 mph wind is not doubled at 20mph, it's 4 times as great (10 squared is 100, 20 squared 400)."


I'll assume that the effect is programmed correctly and that your example was pulled off the top of your head. But, to be rigorous...

Your example is only true for an object at rest. Drag is proportional to wind density times its velocity squared. But the velocity of interest is that of the wind relative to the horse, not the wind relative to the ground. Since the horse is already traveling close to 40 mph, a headwind of 20 mph makes the velocity of interest 60 mph (40+20), while a head wind of 10 mph makes the velocity of interest 50 mph (40+10). For this example, the effect is not nearly 4 times as great - it is only 1.4 times as great (60**2/50**2).

The overall effect of wind sounds like a pretty straightforward problem, given a reasonable amount of data. I suspect the correlation could best be validated by looking at results from straight races (quarter horses) over a wide range of head and tail winds.

I suspect that the effect of density would normally be small, but I will ask a question. What happens to your variants in the rain, snow, and fog?