Saratoga 7f

[alydar_ David]

As of August 16, 2010: 100% of the horses in my Saratoga 7f model were either first or second in TE and TS.

Note how narrow the range is for Median%.

[Eddybugsy]

I finished my modeling work last night and my findings were similar.
TE + TS = 1or2 at 73%
and the other 27% or the holes so to speak was filled in with DCL=1

[Eddybugsy]

I also found the thing for 6F

[BJennet]

Originally Posted by alydar_David

As of August 16, 2010: 100% of the horses in my Saratoga 7f model were either first or second in TE and TS.

Note how narrow the range is for Median%.

alydar David,

Thanks for the chart - typical of your excellent posts. Last year I did an informal study of TE based on a sample of 1200 races, and the performance of this factor is extremely consistent. And my results concurred with yours, with regards to the NY and CA circuits, where #1 and #2 TE combined for a win rate of ca. 80%, compared with 65% at all tracks. This factor alone has given me two Pick-6's within the past couple of months, albeit chalky ones. In Sunday's, at DMR, #1 TE won three races, #2 won two, #3 the other, but with a strong DCL.


Re TS - just like to mention something I know you're aware of, but which newer players seem not to know, is that there is an enormous amount of redundancy in the range of Sartin Methodology factors, which can make learning more difficult than it needs to be.

One of the goals of my study was to try to isolate the power of the various factors. It can't be done. TE, TS, TPP, CPR, VDC, and any of the other 'overall' factors are very highly correlated. This may not seem apparent from one race or even a day of racing, but if you look at a couple of hundred races, the results speak for themselves.

Doc himself made this point, which is why he greatly simplified his play in the last few years, as the 'Follow Up' issues from that period demonstrate. During this period Doc often mentioned something revealed in your chart - don't be influenced by positional running style or what he called the 'visual' aspect of the race. As your chart illustrates, running styles are far less predictive than the very tight e/l or %Med models - exactly Doc's point. And as he mentioned, he didn't model anything in those last years except e/l, and simply used it in conjunction with the BL/BL.

Thanks again.

Cheers,

B Jennet

[Ted Craven]

Thanks David and Bill, and thanks BJennet for your supporting macro evidence.

David, Bill, of course such models are dependent on the method used to select the pacelines used to represent the horses (from which the 100% etc stats are drawn).

Could either or both of you say just a few words about what paceline selection strategy you employ for these models.

Thanks,

Ted

[alydar_ David]

Eddybugsy, nice to have reinforcement as models are by their very nature subjective.

BJennet, thanks for yet another erudite post. Glad to hear of your Pick3 successes.

Ted, I'll post a race to illustrate by example. It'll be the third race at Saratoga today in the Selections section.

[alydar_ David]

Here's an update for Saratoga 7f dirt, non-maiden, non-stakes. As you can see 100% of the recent winners are first or second in all three categories: TE, F1 and TS.