Meadow calls Sartin con man

[FakeNameChanged]

Originally Posted by lefty359

Doc never said use a single paceline. He said you could use several and settle on the best one. He was the first I remember to explode the myth of the last PL being the best one. I still use best of last 3 comparable lines with success, and yes, when in doubt, I put in more than one and let the prgm tell me which is best.

I never followed the evolution of Sartin's methodology, but he originally said that 90% of the time, your BEST pace line will come from the last two races, and never go back more than five. He said to pick the Best pace line. source: pacemakestherace.com and I've read that else where. I seem to remember he modified that later on.

The process was:

Select 5 Contenders in each race (only “contenders” were handicapped.)
Select a single paceline for each contender.
Based upon the EP-SP-W ranks two horses were selected as bets in each race.
Howard once said that 90% of the time, one of the last 2 races should be picked as pacelines. In later years, I believe he felt differently about that.

He also said that in order to win, there were 3 things the player must do:

Select the right contenders.
Select the right pacelines.
Interpret the readouts correctly.
Essentially, Howard said that if you made mistakes on any of these, you were in danger of losing the race.

So if you don't pick the right contenders, select the right pacelines, or Interpret the readouts correctly, it's on you. Which it is anyway.

[Tom]

Kgen, maybe the best program ever written, solved the paceline selection by displaying them all visually. You could see what lines were abberrent and avoid them.

In Brohamer's MPH, you can put in multiple lines and look how they all compare.

[jay68802]

Originally Posted by Tom

Kgen, maybe the best program ever written, solved the paceline selection by displaying them all visually. You could see what lines were abberrent and avoid them.

In Brohamer's MPH, you can put in multiple lines and look how they all compare.

Are you saying finding the best (fastest) horse in the race is a good idea?

[Tom]

Yes, but when in the race it is the fastest is important.

[dlivery]

When finished the race.
What has to give.

[46zilzal]

Originally Posted by Tom

Kgen, maybe the best program ever written, solved the paceline selection by displaying them all visually. You could see what lines were abberrent and avoid them.

That one was unique

[lefty359]

Synergism6 does that too, which is a real nice thing.

[JHK]

The Meadow article "Unraveling the Sartin Methodology" is reproduced (with permission) in full at http://sports.horse/

[classhandicapper]

Originally Posted by JHK

The Meadow article "Unraveling the Sartin Methodology" is reproduced (with permission) in full at http://sports.horse/

Reading that did nothing to make me want to learn more about their methodologies. It made me glad I didn't spend much time on it to begin with.

[ranchwest]

Originally Posted by classhandicapper

Reading that did nothing to make me want to learn more about their methodologies. It made me glad I didn't spend much time on it to begin with.

Pace and pace lines tell me quite a bit. Sartin doesn't do as much for me as other approaches.

[pandy]

In the late 70's, before I'd ever heard of velocity ratings, I created a velocity ratings system for harness racing, using feet per second, which I eventually changed when I created my Diamond System for thoroughbreds and harness. Both versions are essentially velocity systems but I changed from feet per second to breaking each race down into average 1/16's of a mile. My main formula, however, puts more weight on how a horse finishes than the Brohamer/Sartin Average Pace rating, and the results are better.

The weakness of velocity style ratings is also the strength of the ratings. I mainly use my Diamond Ratings to spot live longshots that I use in exotic wagers.

The weakness of velocity style ratings is as follows.

Okay, you have a $20K claimer who ran 6 furlongs like this: 22.2, 45.1, 1:09 on a track that was a second faster than par, and earned a published speed figure of 90.

A few days later at the same racetrack a $20K claimer ran 6 furlongs like this: 23.2, 47.1, 1:12,on a track that was three and two fifths of a second slower than par, and earned a published speed figure of 92.

Speed figure handicappers are going to prefer the slower horse because the adjusted times show that even though the 1:12 horse ran three seconds slower over the same racetrack in the same class, the horse actually ran about two fifths of a second faster.

With velocity ratings, yes, the ratings will be adjusted accordingly for the track variants, however, in these situations, in most cases, the horse than actually ran much faster is going to have the better (lower) velocity ratings.


Most of the time, the speed figures are right and the 92 on the slow surface does finish ahead of the horse that ran the 90 on the faster surface. But, sometimes, actually quite often, it turns out that the horse that ran faster but got the lower speed figure wins the race and pays a good price. ROI's on top speed figure horses are usually low.

I've seen literally hundreds of horses that ran the fastest actual times when using velocity ratings win and pay huge prices, beating horses that had higher speed figures.

Sometimes the fastest horse is actually the horse that ran faster. This is even more useful now than ever because of all the turf racing. Now you have days where there are only three dirt races and it's very difficult to make a good track variant, so the speed figures can be way off.

[pandy]

There's an interesting comparison of velocity vs published figures in the 4th at Parx today.

On Beyers, the 4 horse, Let's Celebrate, has the highest Beyer, a 48 from its debut three starts back. And it's last Beyer is 34. The filly has raced three times and the last three Beyers are 34-36-48, an apparent regression, but now the horse changes hands, ships in from Churchill and cuts back to a sprint.

The 9, Angelsfour Michele, has run twice. Her first Beyer was 0 and her last race, she ran a Beyer of 37.

Now on my Diamond Ratings, Angelsfour Michele's last race is a 78 and Let's Celebrate's best rating (which was from her lone sprint at Fairgrounds) is 87. So Angelsfour Michele has a 9 point advantage, which is a comfortable margin. If you're using each fillies best sprint Beyer to rate them, the 4 looks best. If you're using my Diamond velocity style ratings, the 9 looks best. I would imagine that the 9 will also look best using the Brohamer or Sartin ratings.

[delfman]

Originally Posted by classhandicapper

Reading that did nothing to make me want to learn more about their methodologies. It made me glad I didn't spend much time on it to begin with.

I've never been the type of person who wants to gain info from computer programs or the data produced by them.


just like as in golf I take a total feel approach to handicapping, whereas pace projection is a big part of the process. Maybe not so much pace projection as, race flow.



That article was an interesting read and I learned a thing or two by reading most of it. Thanks for posting the link JHK

[classhandicapper]

Sometimes the fastest horse is actually the horse that ran faster.

We've discussed these previously.

I think there may be a fundamental problem with all figure models.

Good jockeys know when a track is deeper and more tiring than average and when it's like a paved highway and carrying speed better than average. They adjust their aggressiveness to the track, which in turn impacts the fractions, which in turn impacts the final times.

Pace and speed figure models can't differentiate between faster/slower times because of the speed of the track or because the riders were more or less aggressive than usual because of the track.

They build both in via the track variant assuming it was all track speed.

If for example a deeper slower track causes all the riders to rate hard early, the early fractions will be slower than usual in part because of the slow track, but in part because of that rider adjustment. The horses may also finish in a slower final time partly because of that adjustment.

The pace/speed figure maker will assume the entire slow time (pace and final time) was track speed related and build both in via the variant, but the horses didn't really run as fast as the figures. Same thing can happen in reverse.

I think in some cases conventional models may tend to overrate the pace/speed figures of horses that ran on slower tracks and underrate the pace/speed figures of horses that ran really fast on a highway. On the flip side those highway tracks can sometimes carry speed horses to ridiculously fast figures they won't duplicate on an average track. It's a complex mess that's hard to prove or solve.

[Track Collector]

Originally Posted by pandy

There's an interesting comparison of velocity vs published figures in the 4th at Parx today.

On Beyers, the 4 horse, Let's Celebrate, has the highest Beyer, a 48 from its debut three starts back. And it's last Beyer is 34. The filly has raced three times and the last three Beyers are 34-36-48, an apparent regression, but now the horse changes hands, ships in from Churchill and cuts back to a sprint.

The 9, Angelsfour Michele, has run twice. Her first Beyer was 0 and her last race, she ran a Beyer of 37.

Now on my Diamond Ratings, Angelsfour Michele's last race is a 78 and Let's Celebrate's best rating (which was from her lone sprint at Fairgrounds) is 87. So Angelsfour Michele has a 9 point advantage, which is a comfortable margin. If you're using each fillies best sprint Beyer to rate them, the 4 looks best. If you're using my Diamond velocity style ratings, the 9 looks best. I would imagine that the 9 will also look best using the Brohamer or Sartin ratings.

Shhh. The #9 is a pattern match for me today. Hopefully I get a winner.