TGJB -

It was not my intent to disparage your figures or your figuremaking in my earlier post (now a couple of days ago) in this thread. You have explained your methods well enough that I know you wouldn't arbitrarily set the variant for a spring 3yo stake race so that half the horses at least pair their tops.

However, as you have indicated, the projection method of figuremaking requires that you have some preconceived notions about what is likely to have occurred in a race (e.g., they didn't all run five point new tops) that allows you to decide whether to break a race out and, when you do, to pick the variant that makes the most sense in light of the fixed beaten-length-and-ground-loss relationships between the horses. As you suggest, the alternative is to be very reticent about giving different variants to individual races absent an obvious change in track condition. I'm inclined to side with you that your approach is more accurate than the alternative as a means of creating a useful handicapping tool.

I think you have to be careful, however, about using your figures to validate your underlying preconceived notions (such as that spring 3yo stake horses at least pair their tops roughly 50% of the time). Because of your preconceived notion, your figuremaking will tend to push the figures in that direction. I'd concede that you couldn't push a square peg into a round hole (so that 50% is probably not an unreasonable estimate). The extent to which your bias might affect things would depend in large part on how often you break races out and by how much.

Having exhausted my thoughts on that subject, I do have a couple of questions:

Len's argument that all permutations are possible raises one of the questions. It's fair to guess that somewhere, sometime, there was a spring 3yo stake race in which all of the horses either ran an off race or Xed (I'm reminded of the movie "Magnolia": strange things happen all the time). Do you think your methods would allow you to identify such a race? Or do you think an occasional error in assessing an individual race is a small price to pay for getting most of the races right? Or do you think something else entirely?

My second question relates to the Fountain of Youth. Without having seen subsequent figures for the entire field, it strikes me that a number of horses coming out of that race posted markedly better figures in their next races. Would something like that cause you to revisit your figures for the race?

Respectfully,

BitPlayer