Pace and Cap - Sartin Methodology & The Match Up - View Single Post - 'For the Lead' Guidelines

zimal2 · 02-08-2018, 01:13 PM

I was offline a few days, but wanted to reply to your response to my posting FTL. It is as Ted says an HTR spreadsheet, and it is formatted a little differently than you suggest, but I believe the info and my statement is correct:

For all races last 12 months:

Factor Plays Win W+P WPS . 02/08/2018
---------------------------------------------------------------------------------------------------------
Play All 311261 13% 26% 39%

This means there were over 300,000 horses that raced, and since no filter was applied there were 13% winners or alternatively a little less than 8 horses per race on average since each race has a winner.

Now using the filter of first after layoff only---all races.

Play All 25262 11% 21% 33%
There were 25,000+ horses who were 1pL and of those there were 11% winners

There was little difference whatever the surface except for AR with smaller sample size:
Dirt Sprint 12789 10% 21% 33%
Dirt Route 02030 11% 22% 33%
Turf Sprint 02368 10% 21% 32%
Turf Route 02531 11% 21% 31%
Wet Sprint 02705 11% 21% 33%
Wet Route 00484 11% 25% 36%
Artf Sprint 01900 11% 22% 35%
Artf Route 00455 14% 25% 39%

Using a proprietary HTR algorithm called KAT with 1pL:
KAT 6 01206 16% 31% 43%
KAT 7 01284 21% 37% 51%
KAT 8 00549 18% 35% 48%
KAT 9 00464 28% 45% 61%
KAT= 10 00084 33% 50% 62%

There were 3587 horses in that cluster and roughly 717 winners or about 20% winners on average. So I would stick to my contention that throwing out these layoff horses is dicey. I didn't post the WROI because I don't know how to align the longer column lines. It is not profitable without further filtering, but my point has to do with the win% regardless of ROI.

I use both RDSS and HTR. They complement each other, because RDSS shows the pace picture much much easier than HTR, but HTR has many features that relate to form, et al that are easier to reference than RDSS. (It's ironic that HTR isn't easy to grab the pace picture from, since Ken Massa programmed Brohamer's ideas, but Ken has gone away from a pace approach over the years). In addition, HTR has the "ROBOT" which allows one to make complicated queries leading to spreadsheet data in anywhere from 60 seconds to 5 to 10 minutes.

02-08-2018, 01:13 PM	#62
zimal2 AlwNW1X Join Date: Jan 2013 Posts: 19	I was offline a few days, but wanted to reply to your response to my posting FTL. It is as Ted says an HTR spreadsheet, and it is formatted a little differently than you suggest, but I believe the info and my statement is correct: For all races last 12 months: Factor Plays Win W+P WPS . 02/08/2018 --------------------------------------------------------------------------------------------------------- Play All 311261 13% 26% 39% This means there were over 300,000 horses that raced, and since no filter was applied there were 13% winners or alternatively a little less than 8 horses per race on average since each race has a winner. Now using the filter of first after layoff only---all races. Play All 25262 11% 21% 33% There were 25,000+ horses who were 1pL and of those there were 11% winners There was little difference whatever the surface except for AR with smaller sample size: Dirt Sprint 12789 10% 21% 33% Dirt Route 02030 11% 22% 33% Turf Sprint 02368 10% 21% 32% Turf Route 02531 11% 21% 31% Wet Sprint 02705 11% 21% 33% Wet Route 00484 11% 25% 36% Artf Sprint 01900 11% 22% 35% Artf Route 00455 14% 25% 39% Using a proprietary HTR algorithm called KAT with 1pL: KAT 6 01206 16% 31% 43% KAT 7 01284 21% 37% 51% KAT 8 00549 18% 35% 48% KAT 9 00464 28% 45% 61% KAT= 10 00084 33% 50% 62% There were 3587 horses in that cluster and roughly 717 winners or about 20% winners on average. So I would stick to my contention that throwing out these layoff horses is dicey. I didn't post the WROI because I don't know how to align the longer column lines. It is not profitable without further filtering, but my point has to do with the win% regardless of ROI. I use both RDSS and HTR. They complement each other, because RDSS shows the pace picture much much easier than HTR, but HTR has many features that relate to form, et al that are easier to reference than RDSS. (It's ironic that HTR isn't easy to grab the pace picture from, since Ken Massa programmed Brohamer's ideas, but Ken has gone away from a pace approach over the years). In addition, HTR has the "ROBOT" which allows one to make complicated queries leading to spreadsheet data in anywhere from 60 seconds to 5 to 10 minutes.