Making your own speed figs? It's multi-dimensional now or bust!

[GaryG]

Originally Posted by thaskalos

If a closer gains 5 lengths during the stretch run of a given race, and this closer is credited a full second for his stretch gain...then the stretch runner has gotten an adjustment advantage that he doesn't deserve, relative to the speed horse that he is trying to catch.

This is very true....you can see how true by looking at the Trakus times.

[Capper Al]

thaskalos,

Okay, I wasn't reading all that. But it's there. My focus was more on the side that whatever number one chooses 1/5 or 1/6 second isn't that big of a deal after the times are made parallel for speed not necessarily pace. Matter of fact, after discussing this I'm going to adjust what I'm doing a bit. I can see how the pace boys are concerned about lengths at different stages of the race. There is big difference between first call feet per second and last furlong. Is it significant over a thousand races? I don't know. When I get to pace in my rewrite, I'll be more able to discuss these details. But now for a speed fig, I'm leaning as to not too terribly different what one uses after the times are made parallel.

[Capper Al]

Originally Posted by fmolf

Very well said.....In this way every horse in every race is judged against the same standard. Simple!...I am a firm believer in K.I.S.S methods.

I believe in K.I.S.S also, but with handicapping I find it difficult not to complicate matters. As some say around here, there are just too many confounded variables.

[Dave Schwartz]

Al,

What do you mean by "multidimensional" ratings?

[Capper Al]

Originally Posted by Dave Schwartz

Al,

What do you mean by "multidimensional" ratings?

How I use the term for handicapping is that I have many ways to look at factors like speed, class, pace, connections etc. I formulate one common thread for each factor and build a comprehensive view with a final weighted score. Yes, if I used my time flipping hamburgers at McDonald's I would make more money.

[whodoyoulike]

Originally Posted by Capper Al

How I use the term for handicapping is that I have many ways to look at factors like speed, class, pace, connections etc. I formulate one common thread for each factor and build a comprehensive view with a final weighted score. Yes, if I used my time flipping hamburgers at McDonald's I would make more money.

It would have been clearer if you used "comprehensive" ratings rather than "multidimensional".

Have you been able to develop a comprehensive rating based on speed, class, pace, connections etc.?

If you have, I think you need to remember it is just one rating which needs to be considered in relationship with your other ratings.

[Dave Schwartz]

Thank you for the explanation.

At what point do these ratings (multidimensional, composite, comprehensive, etc.) no longer become "speed ratings?"

As WhoDo, said, ...

It would have been clearer if you used "comprehensive" ratings rather than "multidimensional".

In the HDW system, Jim Cramer has a very powerful rating called "Projected Speed Rating." However, in truth, it is not really a speed rating at all. Kind of like BRIS' Prime Power rating.

So, my question to you Al (meaning no disrespect whatsoever) is:

"Are you trying to create a more accurate speed rating for each past race, for each horse, or are you trying to take speed ratings as well as other ratings/metrics/facts and integrate them into a hit rate?"

[clemkadiddle]

Originally Posted by thaskalos

Clem made the astounding assertion that everyone but him adds a fifth of a second to the horse's time for every length that the horse is behind the leader at the various points of call. That simply isn't true. A great many handicappers have realized that a length isn't worth a fifth of a second...and the pace handicappers have largely abandoned this "convenient" adjustment measure...because inaccurate adjustment measures DO make a difference...even when the measurement is a CONSTANT, and is applied to ALL the horses equally.

In this game...the pace handicapper is comparing speed horses to closers. If a closer gains 5 lengths during the stretch run of a given race, and this closer is credited a full second for his stretch gain...then the stretch runner has gotten an adjustment advantage that he doesn't deserve, relative to the speed horse that he is trying to catch. When a pace handicapper is comparing speed horses to closers...then a fifth of a second "mistake" during a fraction of the race makes a difference.

I have no real argument to make against the notion that simplicity is worth a small measure of inaccuracy...nor can I really argue with the guy who states that precision is an impossibility in this game, since even the employed method for reported lengths behind in this game is fraught with inaccuracies. The horseplayer makes his own adjustment decisions...and he lives with them. But Clem's post seemed to me to be decidedly smug and egotistical. EVERYONE does "this", but Clem does "that"...because he is a 25-year programmer, and a 40-year horseplayer. That's an odd way to talk...especially when you don't really know what everybody else is doing.

Correct...I don't know what everyone else is doing. My library consists of books at least 30 years old...and this is the basis of a lot of work in it. Even Beyer's Speed Figures are based on 1/5 of a second per length combined with parallel time. Beyer's "Picking Winners" divulges the formula. The Beyer figure is based on a segment consisting of 1/10th of the overall distance and the projected percent that the horse's final average speed intersects this segment. (Trust me on this one.)

[Greyfox]

Originally Posted by thaskalos

.because inaccurate adjustment measures DO make a difference...even when the measurement is a CONSTANT, and is applied to ALL the horses equally.

.

Sorry Thask.
But if you are adding or subtracting a CONSTANT for a length to every horse, depending on their beaten lengths, you'll get the same rank order and that will be accurate.
The math dictates that is so.
If you're working with raw numbers, of course they won't reflect reality, but the rankings will come out the same.

[clemkadiddle]

The overall objective of my work was to equate efforts at different distances...even over different surfaces...where parallel time fails.

Parallel time has been accepted as the standard for all speed handicapping. However, what I have seen in my calculations is that horses are more consistent than one might have guessed. Even with 5 to 7 point fluctuations in speed ratings, summing up the results of each segment for the races in question have produced almost identical totals when horses are in form. That is why I am of the opinion that the concept of parallel time has become obsolete.

Basically, the method:

1. Reverse-engineer the final time where the 100 point score was awarded. The only problem with this I have is BRIS tends to classify routes from 1 mile and longer. In a card where there is a mile race and perhaps a couple of longer races, BRIS will say that the routes are playing 2 seconds slower. When they incorporate this into the speed rating for the mile race, the variant becomes skewed.

2. Once having the 100 point time, compare this to a standard parallel time chart. Use the average FPS for both the actual race as compared to that on the chart and derive a ratio that can be used in the pace calculations.

3. Parse the race into individual segments. Use 8 feet for 1 length. Calculate distance traveled, making sure to include 8 feet for each length closed...or lost. This will reveal the FPS for the segment; multiply times the variant ratio to get an adjusted FPS.

4. Plug this FPS into the mathematical model and multiply by the number of furlongs for the segment.

5. The mathematical model is an exponential expression involving the degree to which the horse exceeds 44 FPS...that's all I am going to say at this point because this is the basis of 8 years of research that involved throwing a whole lot of mathematical darts at this subject. It wasn't until I had this idea...and started pursuing it...I was amazed with the results.

[thaskalos]

Originally Posted by Greyfox

Sorry Thask.
But if you are adding or subtracting a CONSTANT for a length to every horse, depending on their beaten lengths, you'll get the same rank order and that will be accurate.
The math dictates that is so.
If you're working with raw numbers, of course they won't reflect reality, but the rankings will come out the same.

So...you are saying that it makes no difference if a length is really a fifth of a second or a TENTH of a second. As long as we apply the same "constant" to all the horses across the board, then we are fine...right?

[Greyfox]

Originally Posted by thaskalos

So...you are saying that it makes no difference if a length is really a fifth of a second or a TENTH of a second. As long as we apply the same "constant" to all the horses across the board, then we are fine...right?

If you are looking for a bulls eye accurate representation in raw figures, it will make a difference.
If you are looking at rankings the horses will come out in the same order.
In my own instance, all of my numbers in my private program are converted to percentages.
Using tenths or fifths will give different percentages, but not different rankings.
But the top horse will be the top horse and the bottom one the bottom one.
The rankings will be the same - as long as you use the same "constant."
Obviously, it's nice to be close to what is really happening, but if you are using a constant, for ranking purposes it won't change the outcomes.

[thaskalos]

Originally Posted by Greyfox

If you are looking for a bulls eye accurate representation in raw figures, it will make a difference.
If you are looking at rankings the horses will come out in the same order.
In my own instance, all of my numbers in my private program are converted to percentages.
Using tenths or fifths will give different percentages, but not different rankings.
But the top horse will be the top horse and the bottom one the bottom one.
The rankings will be the same - as long as you use the same "constant."
Obviously, it's nice to be close to what is really happening, but if you are using a constant, for ranking purposes it won't change the outcomes.

All I can say, Greyfox...is that I don't agree. And neither would any pace handicapper that I have ever come across. Pace handicappers don't just look at horse "rankings"; they also look at running styles and race shapes. They break down individual fractions...and they demand a certain level of precision. Even in an imprecise game.

[classhandicapper]

clem,

One problem I have with looking at segments is that IMO it tends to overrate "the chances" of horses that use their energy inefficiently relative to those that don't even if it's measuring their performances properly. Horses will tend to run the same way in future starts.

What I think you want to look for is extremes because then there's a good chance the horse will use it's energy more efficiently in its future starts and improve the end result.

[Greyfox]

Originally Posted by thaskalos

All I can say, Greyfox...is that I don't agree. And neither would any pace handicapper that I have ever come across. Pace handicappers don't just look at horse "rankings"; they also look at running styles and race shapes. They break down individual fractions...and they demand a certain level of precision. Even in an imprecise game.

I can't disagree with what you are saying here.
(P.S. I also look at running styles and race shapes.)