Even when the track speed essentially is staying the same, there are small adjustments we make race to race-- could be because of wind gusts, slight differences in moisture content, etc. One of the interesting things about doing this is that situations come up often where you can see exactly what the race should be, absolutely rock solid, first time through, without looking at the surrounding races at all-- a couple of years ago we posted the Peter Pan on this site (should still be in the archives), and almost the whole field paired their recent tops. There are other times when you can have 2 or 3 races in a row with lightly raced horses, some of them obviously jumping up, and it's a battle.

This is one of those things that is difficult to describe to someone who has not done it. Most of you out there who have made figures have probably used pars, which means you are basically just going off the winners, which is indeed a small sample. The ones who have made projection style figures have not had the advantage of using wind and ground, even if you do use weight. Point being, when you do it the way we do, the sample size (the whole field, lifetime histories) is not that small, and the figure histories of the horses very accurate, which over time leads to a tighter and tighter data base. Unless you make logical misassumptions, in which case everything is up for grabs.



TGJB