Since Speed is the Long Pole in the Tent

[Capper Al]

Originally Posted by classhandicapper

I've experimented with that kind of stuff endlessly.

My experience is that when you start introducing more factors into a number it can sometimes cause just as much damage as good in terms of accuracy.
That's why I decided to finally test all this stuff with live data and answer some of these questions definitively instead of by trial and error betting experience.

The evidence so far suggests to me that I am better off using my methods of classing horses as the primary driver and using speed and pace figures as a minor component instead of the primary components.

And because of introducing more factors is why I'm doing a rewrite.

[Capper Al]

Bayesian isn't that advanced. It's stats 101.

[raybo]

Originally Posted by thaskalos

The speed figures are not overbet to the extent that some people think. Every day I see cases where the high-figured horses are IGNORED in the betting...mostly because of class or pace concerns. I say, instead of throwing the speed figures out altogether...what if we augmented these figures to include some of the elements that we know the speed figures are missing? What if we infuse the effects of pace into these figures...while also introducing class into the equation?

I have done something similar with sprint races over the last couple of years, and I have come up with an adjustment to the Beyer speed figures...which has me excited to the point where I now wager only on sprints...with an unusual amount of success. I have been trying to combine pace and speed figures for many years now, with varying degrees of success, and I don't usually get overly excited about things...but I honestly feel that I am making my biggest strides as a horseplayer NOW...in my ADVANCED years.

We are never too old to learn.

Thaskalos, since many of us realize that speed figures are not single factors, but rather are composites of raw final time, beaten lengths converted to time, and daily variant which includes bias and perceived surface speed, etc., should we be looking at speed figures as a starting point in handicapping? I think not. Shouldn't we be starting with something that is physically measured (at least all of us who use published data)? Speed figures at not physically measured, raw final time is, speed figures are "interpreted" by the creator(s) of that figure. Shouldn't we be using raw time, first? Or, are we so inadequate that we must trust someone else to do our interpreting for us?

These are some of the questions I asked myself years ago, and came to the realization that I should totally ignore all published figures, and go my own way instead. At least then, any failures I suffered were faults of my own, not blind trust in someone else. My reasoning is that, if I'm not good enough to beat the game using my own stuff, then I shouldn't be in the game in the first place.

If I'm going to use someone else's "composite" figure(s), then I might as well create my own composites that better explain what actually happened in races, what the total performance of each horse was, not just what some contrived speed figure said it was. If I'm good enough at interpreting what actually happened in races, I will be far ahead of those who are dependent on the perceived skill of some figure maker. (CJ not withstanding of course, as I believe that he probably does a better job, long term, of interpreting individual performances than I do, and most others I've heard of or observed.)

[pandy]

Originally Posted by raybo

Thaskalos, since many of us realize that speed figures are not single factors, but rather are composites of raw final time, beaten lengths converted to time, and daily variant which includes bias and perceived surface speed, etc., should we be looking at speed figures as a starting point in handicapping? I think not. Shouldn't we be starting with something that is physically measured (at least all of us who use published data)? Speed figures at not physically measured, raw final time is, speed figures are "interpreted" by the creator(s) of that figure. Shouldn't we be using raw time, first? Or, are we so inadequate that we must trust someone else to do our interpreting for us?

These are some of the questions I asked myself years ago, and came to the realization that I should totally ignore all published figures, and go my own way instead. At least then, any failures I suffered were faults of my own, not blind trust in someone else. My reasoning is that, if I'm not good enough to beat the game using my own stuff, then I shouldn't be in the game in the first place.

If I'm going to use someone else's "composite" figure(s), then I might as well create my own composites that better explain what actually happened in races, what the total performance of each horse was, not just what some contrived speed figure said it was. If I'm good enough at interpreting what actually happened in races, I will be far ahead of those who are dependent on the perceived skill of some figure maker. (CJ not withstanding of course, as I believe that he probably does a better job, long term, of interpreting individual performances than I do, and most others I've heard of or observed.)

Well, of course, the question then becomes, is it worth your time? Almost all horseplayers are casual bettors who play horses as a form of entertainment. They certainly aren't going to spend several hours a day making up their own performance or speed figures. I did it for years and although it was a good learning experience, it also kind of sucks the fun out of playing the horses.

And, if you do spend a lot of time creating your own performance rating, it had better be a lot more proficient than the best published speed figures or Bris Prime Power, etc., otherwise you're wasting time.

Of course you may have it all computerized and it doesn't take much time, but one thing that makes figures like CJ's or Beyer's valuable is that they do most of the work manually, so they can make decisions regarding track variants, for instance (such as splitting variants because of weather changes), that a computerized program can't.

[raybo]

Originally Posted by pandy

Well, of course, the question then becomes, is it worth your time? Almost all horseplayers are casual bettors who play horses as a form of entertainment. They certainly aren't going to spend several hours a day making up their own performance or speed figures. I did it for years and although it was a good learning experience, it also kind of sucks the fun out of playing the horses.

And, if you do spend a lot of time creating your own performance rating, it had better be a lot more proficient than the best published speed figures or Bris Prime Power, etc., otherwise you're wasting time.

Of course you may have it all computerized and it doesn't take much time, but one thing that makes figures like CJ's or Beyer's valuable is that they do most of the work manually, so they can make decisions regarding track variants, for instance (such as splitting variants because of weather changes), that a computerized program can't.

Of course my methods are computerized, and the "effort" was expended years ago, costing me only some time and data. For all except high class races, the win contenders are produced in less than a second (if the race is even playable in the first place). I test individual tracks in an automated system, usually using a year or more of past results from each track versus my automated rankings methods. Hit percentage is only half of the equation, average price being the other half. My reasoning? The long term is the only thing that counts, so individual results are not of prime importance to me (even though they help determine the results, I do not focus on them except in the case of dealing with outliers). As I stated, I believe that CJ is a bit better at interpreting race performances than I am, but in the long term, my method produces profit, so evidently my method does well enough at interpreting individual performances to produce net profit, and that's all that counts, IMO.

[Cratos]

Originally Posted by Capper Al

Bayesian isn't that advanced. It's stats 101.

There are some very good busiiness/engineering schools at CMU, U of R, MIT and many others that would differ with you, but you are entitled to your opinion

[Cratos]

[QUOTE=raybo]Thaskalos, since many of us realize that speed figures are not single factors, but rather are composites of raw final time, beaten lengths converted to time, and daily variant which includes bias and perceived surface speed, etc., should we be looking at speed figures as a starting point in handicapping? I think not. Shouldn't we be starting with something that is physically measured (at least all of us who use published data)? Speed figures at not physically measured, raw final time is, speed figures are "interpreted" by the creator(s) of that figure. Shouldn't we be using raw time, first? Or, are we so inadequate that we must trust someone else to do our interpreting for us?

These are some of the questions I asked myself years ago, and came to the realization that I should totally ignore all published figures, and go my own way instead. At least then, any failures I suffered were faults of my own, not blind trust in someone else. My reasoning is that, if I'm not good enough to beat the game using my own stuff, then I shouldn't be in the game in the first place.

If I'm going to use someone else's "composite" figure(s), then I might as well create my own composites that better explain what actually happened in races, what the total performance of each horse was, not just what some contrived speed figure said it was. If I'm good enough at interpreting what actually happened in races, I will be far ahead of those who are dependent on the perceived skill of some figure maker. (CJ not withstanding of course, as I believe that he probably does a better job, long term, of interpreting individual performances than I do, and most others I've heard of or observed.)[/QUOTE
Raybo, you have written an excellent retort.

[classhandicapper]

Originally Posted by Capper Al

Bayesian isn't that advanced. It's stats 101.

I took a couple of stats classes in college, but I can't remember any of it. To me, anything beyond average, median, and standard deviation is advanced stats.

I recently read a book that had an entire chapter on the subject, but I wouldn't know how apply it to what I am doing. What I am doing is sort of a backdoor regression using trial and error but with decades of handicapping experience guiding the research and parameter values. I'm already a better horse payer from the effort because now I know the right approach for me even if I don't have all the correct weights yet. For years I kept going back and forth experimenting with hybrids. Now that much is settled.

[Show Me the Wire]

How about this novel, ok not so novel idea about composite factors. Answer this question first: Why is the horse in this race?

[thaskalos]

Originally Posted by raybo

Thaskalos, since many of us realize that speed figures are not single factors, but rather are composites of raw final time, beaten lengths converted to time, and daily variant which includes bias and perceived surface speed, etc., should we be looking at speed figures as a starting point in handicapping? I think not. Shouldn't we be starting with something that is physically measured (at least all of us who use published data)? Speed figures at not physically measured, raw final time is, speed figures are "interpreted" by the creator(s) of that figure. Shouldn't we be using raw time, first? Or, are we so inadequate that we must trust someone else to do our interpreting for us?

These are some of the questions I asked myself years ago, and came to the realization that I should totally ignore all published figures, and go my own way instead. At least then, any failures I suffered were faults of my own, not blind trust in someone else. My reasoning is that, if I'm not good enough to beat the game using my own stuff, then I shouldn't be in the game in the first place.

If I'm going to use someone else's "composite" figure(s), then I might as well create my own composites that better explain what actually happened in races, what the total performance of each horse was, not just what some contrived speed figure said it was. If I'm good enough at interpreting what actually happened in races, I will be far ahead of those who are dependent on the perceived skill of some figure maker. (CJ not withstanding of course, as I believe that he probably does a better job, long term, of interpreting individual performances than I do, and most others I've heard of or observed.)

Even a hardcore gambler like myself likes to live a semblance of what could be called a "normal existence"...and making my own figures from scratch would take away from some other things that I also need to do at this time. So...I decided to thoroughly study an easily accessible "pure-speed" commercial figure...figuring that, in time, I would form some sort of opinion about its inherent "weaknesses", and maybe come up with a useful "additive"...which would make these figures more "complete". I initially chose the Equibase figures, and was very happy with them, until someone decided to "recalibrate" them. I now think that the Equibase figures are a joke...especially on the turf.

I admire the dedication of the player who insists on making his own figures from scratch, but, with so many tracks running, I just can't cope with this sort of workload at the present time. It's much easier for me to take the base figure from Beyer...and add my own pace and class adjustments to it.

I am leery about using someone else's "combined" figures...because I am not sure if the figuremaker's pace handicapping philosophy agrees with my own. I have found that, whereas the speed figure philosophy is pretty universal in interpretation...the PACE handicapping philosophies of the individual players vary greatly.

[raybo]

Originally Posted by thaskalos

Even a hardcore gambler like myself likes to live a semblance of what could be called a "normal existence"...and making my own figures from scratch would take away from some other things that I also need to do at this time. So...I decided to thoroughly study an easily accessible "pure-speed" commercial figure...figuring that, in time, I would form some sort of opinion about its inherent "weaknesses", and maybe come up with a useful "additive"...which would make these figures more "complete". I initially chose the Equibase figures, and was very happy with them, until someone decided to "recalibrate" them. I now think that the Equibase figures are a joke...especially on the turf.

I admire the dedication of the player who insists on making his own figures from scratch, but, with so many tracks running, I just can't cope with this sort of workload at the present time. It's much easier for me to take the base figure from Beyer...and add my own pace and class adjustments to it.

I am leery about using someone else's "combined" figures...because I am not sure if the figuremaker's pace handicapping philosophy agrees with my own. I have found that, whereas the speed figure philosophy is pretty universal in interpretation...the PACE handicapping philosophies of the individual players vary greatly.

I know what you're saying because I know, somewhat, from what context you are speaking from. I am not talking about making your own speed figures or pace figures, in the traditional ways that most of the data providers do. What I'm getting at is to ignore all that, create your own "performance" rating(s) from scratch, in your own way, using your own knowledge and experience in what is important to such a rating, without accepting someone else's interpretations/opinions. Of course, in your case, that being not utilizing the benefits of automated computer processes, I can definitely understand why you wouldn't pursue such a method at this late date, and the fact that you are already successful anyway. That would be unnecessary in your case.

But, for those others who are struggling to learn their way around racing, or have been struggling for years using and trying to interpret/adjust published pace and/or speed figures, published class designations, etc., etc., they might be better off starting all over with just the physically measured data, and forge their own path. At least they will have some idea of why they may be failing, rather than thinking that it has to be something other than the published figures and never getting to the crux of the problem, the frequent inadequacy of published figures and their own lack of knowledge about racing in general. When you create something from scratch, you know everything that went into that creation, which means you have personal insight into improving it that nobody else has. If you ever get to the point that your creation is good, nobody else has access to it, and that should give you an automatic edge over the public (and I assume you have already figured out that doing it your way is far better than doing it the way everybody else does).

[Cratos]

Originally Posted by raybo

I know what you're saying because I know, somewhat, from what context you are speaking from. I am not talking about making your own speed figures or pace figures, in the traditional ways that most of the data providers do. What I'm getting at is to ignore all that, create your own "performance" rating(s) from scratch, in your own way, using your own knowledge and experience in what is important to such a rating, without accepting someone else's interpretations/opinions. Of course, in your case, that being not utilizing the benefits of automated computer processes, I can definitely understand why you wouldn't pursue such a method at this late date, and the fact that you are already successful anyway. That would be unnecessary in your case.

But, for those others who are struggling to learn their way around racing, or have been struggling for years using and trying to interpret/adjust published pace and/or speed figures, published class designations, etc., etc., they might be better off starting all over with just the physically measured data, and forge their own path. At least they will have some idea of why they may be failing, rather than thinking that it has to be something other than the published figures and never getting to the crux of the problem, the frequent inadequacy of published figures and their own lack of knowledge about racing in general. When you create something from scratch, you know everything that went into that creation, which means you have personal insight into improving it that nobody else has. If you ever get to the point that your creation is good, nobody else has access to it, and that should give you an automatic edge over the public (and I assume you have already figured out that doing it your way is far better than doing it the way everybody else does).

You are on a 'roll' today; keep it up, this thread is in dire need of your type of advice.

[classhandicapper]

Originally Posted by raybo

When you create something from scratch, you know everything that went into that creation, which means you have personal insight into improving it that nobody else has. If you ever get to the point that your creation is good, nobody else has access to it, and that should give you an automatic edge over the public (and I assume you have already figured out that doing it your way is far better than doing it the way everybody else does).

That's the way I feel about class. No one has the numbers I have.

I have some ideas for late speed and early speed ratings that would be totally unique.

I have some unique ideas about how to incorporate pace into speed figures the way thaskalos is suggesting.

But I think final time speed figures have been researched to death.

[pandy]

One of the main reasons why I developed my various handicapping methods is because you have to be able look at races from a different perspective than the average person, otherwise you'll land on the same horses, or favorites. So creating your own set of numbers certainly makes sense, if they're good.

I started testing a new method on my handicappingwinners website yesterday, hit a cold $255 exacta at Indiana last race. I had to sweat out an objection but it turned out it was on the third place finisher. I bet my tests.

This method looks for horses that finish well but are fast enough to stay within striking distance. For the tests, I'm don't look at the pps at all, just speed and pace ratings.

[Capper Al]

Originally Posted by Cratos

There are some very good busiiness/engineering schools at CMU, U of R, MIT and many others that would differ with you, but you are entitled to your opinion

I'm not knocking Bayesian. I'm happy that it is only stats 101. I forgot most everything else from school.