[Dunbar]

That's nice work on the field size, ranger. It makes a good case for your argument.

I can't help but point out that the field sizes 4-6 do better than field sizes 7-13, though. 87.3% vs 84.9%. The only way I can defend dividing "small" and "large" that way, though, is by saying it puts about half the races in each category. (71 races had 4-6 horses, 73 races had 7-13 horses.)

At any rate, I agree with you that whether the fav towers over the other horses is probably more important than field size.

It looks to me like 88% is a reasonable upper bound on the chance of a heavy favorite actually finishing in the money. If all your picks have an 88% chance of finishing in the top 3, there's about a 1 in 240 chance of picking 43 in a row. If you can nudge your picks up to a 90% success rate, the chance of getting 43 in a row is 1 in 93.