Turns out Friedman is a fan of ROTW, and posted a critique on their site. My (sure to be deleted) reply:

I agree that the 2 stats that Alan cited were not really relevant here, but have to add that
you have pulled one sentence out of 3 pages (and dozens of points). By the way, he read the stat on the page wrong, and Rojas had it wrong too—he rode 12 winners at Suffolk. Alan was looking at the trainer-jockey combo, not the jockey at the track.

Much of the data that any of us distribute on any given day isn’t really relevant to handicapping decisions—most of the Drf pp’s (how many of us study every workout, every point of call in previous races, odds, etc.) and even some of sheet histories (most, not all, of a horse’s numbers 2 and 3 years back are not important now). In the case of trainer profiles, I would estimate that on average about 3 stats for the whole field in one race are important. More with maidens, 1st grass, etc., less with older claimers. The recent move-up trainers have made the last 90 day stats pretty important, too.

The point that you made about how trainers do with a long ship is a good one—that’s exactly why we break out the stat of how they do at an individual track. Obviously, it helps if the sampling is large enough to be meaningful. Using stats raw is no more correct than seeing a horse run 8 last time, and giving him another one. Like Ralph Kiner said, statistics are like bikinis—they show a lot but not everything. And yes, they are used to answer questions that pattern analysis and accurate numbers can’t—how a trainer does with 1st and 2nd time starters, how a sire (and dam) do on grass vs. dirt, how a jock does on off tracks (check out Chavez).

I’m glad that you are paying attention such close attention, though. Any time you want to post your comments on our site, feel free.



TGJB