Can someone at ThoroGraph please reiterate here exactly what
timeframe the stats from the trainer profiles are taken from?
The RUNS-BASED stats?
The TG FIGURE BASED stats?
The jockey stats?
The sire stats?
Shouldn\'t the \"Last 90 days\" RUNS and FIGURE stats be roughly
the same? Why aren\'t they? [On a recent sheet for Rick Dutrow,
his \"Last 90 days\" RUNS-based starts reads 150... and his \"Last
90 days\" FIGURE-based was 87!]
Also provoking these questions is my recent post inquiring about trainer
Jennifer Pederson at Turfway. She won a stakes race at Turfway - two years
ago? But your \"At TP\" stat showed no wins. Please explain.
Thanks.
JohnTChance
In the same vein, is a race within one point of a top deemed to be a pair-up? It seems that it is for purposes of Thoro-Pattern but not necessaily so for analysis purposes (e.g., the comment in the ROTW that Don\'t Get Mad, who had a top of 4 as a 2yo, hadn\'t \"made it back to his 2yo top\" even though he ran a 5 earlier this year.)
The timeframes for the stats follow:
Trainer Run-Based Stats--Year 2001 through the present--updated daily.
Trainer Figure-Based Stats--Year 1999 through the present--updated daily BUT with following eligibility criteria--horse must have run 6 times and 3 times for present trainer. The horse has to run enough times to generate an effective top.
This explains the difference in the \"Last 90 days\" stats. The run-based stat is real time but the figure-based stat is not, the horse had to run at minimum 6 times and the last 3 starts had to be for the trainer in question.
This eligibility requirement doesn\'t pertain to the maiden 1st time out, 2nd time out stats.
Jockey Profiles--last 12 months--updated daily.
Sire Profiles--Year 1992 on--all North American tracks for which we make figures (when we started this we didn\'t make figures for quite a few outside tracks, i.e., Hastings Park, but over time we\'ve extended out coverage to include virtually every meet held in North America. (There may be some county fair type meets we don\'t cover.)
The Thoro-Pattern definitions are:
Top--more than 1 point better than previous top;
Pair--1 point better or worse inclusive of previous top;
Off--more than 1 but less than 4 points worse than previous top;
X--4 or more points worse than previous top.
To generate Thoro-Patterns we had to determine categories--is a 0-1/2 new top the same as 3 point new top? Is there a difference between a 3 point and 10 point bounce? Do figures have to be identical to represent identical efforts or is there some leeway? After consideration, the above categories are what we came up with. For analysis purposes, however, we may make finer distinctions. That\'s left up to individual handicapper.
I forgot to respond to the Pedersen query. The exclusion was the result of a \"glitch\" in the programming which excluded the pertinent year. That has since been fixed as a result your initial query. So thanks for pointing it out.
We try to do everything correct the first time, but we\'re not infallible, so if anyone sees datum, data that look(s) errant please point it out.