Do you trust DRF's Beyer figs.

[rufus999]

Originally Posted by Tom

His new book is titled Beyer on Bikes!

And I thought it was 'The Surrey with no Fringe on Top."

[classhandicapper]

Originally Posted by cj

People that elevate slow race performances are asking for trouble. We both know one popular figure maker that does this. The biggest problem is you boost the also rans to look much better than they are.

If I jog 50 meters next to an Olympic sprinter, then we run the last 50, he'll win by about two seconds. If we sprint the full 100 meters, he wins by five seconds. Not only did we not really "race" 100 meters, I had much more left in the tank for the distance we did "race" and thus could stay closer.

In the first example, what do you give the Olympic sprinter on a speed figure scale? The pathetic rating he would deserve and just remember the circumstances, or boost him to his "usual" race. If you do that, suddenly it looks like I could hang with top class female Olympians, which I certainly cannot.

I agree with you totally.

Figure makers often find themselves in the uncomfortable position of choosing whether to assign a very slow final time figure to race that was impacted by an extreme pace or not.

They realize that many of their customers won't understand the situation well if they assign the slow figure and some may even complain about it because they know the horses are better than that.

However, tinkering causes other problems (like you identified).

It's good to have informed customers.

[Cratos]

Originally Posted by alysheba88

I am not being a smart ass. I dont understand what you are saying

Who is "they"?

I am going to guess at what you might be saying. I could be way off so I apologize in advance. If you are saying handicappers should account for pace, running wide, etc in their handicapping then I agree 1000%. If you are saying that other number based programs do attempt to quantify pace, running wide, etc I have no reason to disagree. I am just saying thats not what Beyers do. If people want to tinker with them, go ahead. But people seem to criticize Beyers for NOT taking those things into consideration when it was never the intent to do so. Its like saying One Base Percentage doesnt take into consideration ballparks. Well they arent designed too. But others have. Which is fine and good. Doesnt mean OBP is "wrong" just means someone has taken it to a different level

I apologize for the lack of clarity in my earlier retort to you, but I was using the word “they” as a euphemism for your word “people.” However I do not use BSFs, but I have found in reading books and articles by Andy Beyer far more informative than the BSRs. I believe his statement (and I am paraphrasing): “there is class within class” is a thoughtful and provocative insight into handicapping horses at a perceived class level.

[Cratos]

Originally Posted by rufus999

I'm assuming you prefer pace over speed. The non-linear argument guys usually do. All I really want to know is do you use the figs in DRF or not. Sorry, but there is no door prize for ingenuity on this thread. Maybe next time.

rufus

Sorry to disappoint you, but I don’t use the BSFs, the DRF figures, or anyone’s else figures, all of my quantitative analysis come from myself. However I do use raw data inputs from Equibase and the DRF.

[Cratos]

Originally Posted by rufus999

I feel the longevity and universal acceptance of the system is proof enough of its efficacy but it takes years to truly understand how it can be used to maximize profit. This thread is proof enough of the many different schools of thought on Beyer figs and speed in general. One thing we can all agree on is that Beyer figs continue to evolve and generate interest.rufus

Phil Bull of the Timeform and a mathematician presented a figure methodology long before Beyer and Ray Taulbot, the former managing editor of American Turf Monthly blazed a trail in figure making with his pace calculator before the BSFs.

[Grits]

I don't know who gets more ink at this place, God or Andy Beyer.

Evolution and BSFs.

Both draw pages upon pages. It's endless.

I'm glad I'm neither one.

[rufus999]

Originally Posted by Cratos

Phil Bull of the Timeform and a mathematician presented a figure methodology long before Beyer and Ray Taulbot, the former managing editor of American Turf Monthly blazed a trail in figure making with his pace calculator before the BSFs.

I was aware of pace handicapping back in the '60's. I wasn't all that impressed or thrilled by it, but heck, I was only 10 years old at the time.

rufus

[alysheba88]

Originally Posted by Cratos

I apologize for the lack of clarity in my earlier retort to you, but I was using the word “they” as a euphemism for your word “people.” However I do not use BSFs, but I have found in reading books and articles by Andy Beyer far more informative than the BSRs. I believe his statement (and I am paraphrasing): “there is class within class” is a thoughtful and provocative insight into handicapping horses at a perceived class level.

Dont understand what you are saying in regard to class. Cant follow

Do agree that his books are more illuminating

[WhyWhyWhy]

Originally Posted by cj

There are plenty of days where the variants (route and sprint separate) stay standard across the day.

Or merely appear to. That +10 change in the variant can be hidden if the pace of the later races is a "-10 pace variant." There are plenty of WEEKS where the track variant doesn't change, but I wouldn't recommend using a weekly variant either. I'm speaking theoretically here of course, against using an unrelated event (another race), which if it is "related" by a variant, that's just by chance.

The upside for CJ (and Beyer) is that, on *many* days, he will get some insight by comparing a race to others on the card, but for that to occur, the track condition has to remain constant from mid-day to sundown and through the work of a track crew. John uses the same approach I saw Russ Harris using, where everything used to make the figures is contained in one race.

A lot of it has to do not with disputing the validity of the variant concept, but rather final time itself. In Europe, a track variant would be difficult to construct the "American" way because they don't care about time over there. I'm beginning to think American horsemen don't care about time either.

CJ uses a projection method to make his numbers. If he's like Beyer, he'd look down on someone who used par times for variants, even though one could make the same arguments that he makes for using projections and a daily variant in that the pars are less precise. A variant is going to be less preciese than a race-based approach for the same reason, i.e., the variant is a shortcut, just like the par time is a short-cut, and both allow the handicapper to skip a more complex task. I think the most notable aspect of the comparison is that CJ's figures would be much more difficult to make if the races weren't timed, whereas John (or me) would barely be affected.

It is easy to see, especially when you are also evaluating the pace of the races while making the variants. There are days, however, where making separate race variants make sense as well.

The use of pace figures is race-based, so CJ is incorporating race-specific factors. His speed figures are actually useful in that they show the mistakes the public is likely to make, since that's how public figures are done. CJ isn't publishing a "grand unification number" (a performance rating), but he is calculating one if he adjusts his speed figure through his pace figure. His wagering advantage is likely sufficient that it's not worth the risk to search for greater precision, as that could backfire if the new idea doesn't work out.

Just as a speed figure can be modified by race-specific factors, you can also do this vertically by examining the history of the horse (or even the connections). For example, if Kiaran McGlaughlin debuts a horse finishing second with a Beyer of 44, I don't care what the pace figure or anything else was, I know it'll run a 97 second time out and win by thirteen lengths at 5-2.

Both can get you in trouble over time. You have to be versatile and adapt on the fly to be successful.

I don't see how race-variants are such a problem, since they inject not only pace, but class, into their numbers. I may not know that some horse earned a 105 in the sixth race when I'm rating the fifth race, but if I know that no horse in the field has broken 85 ever before, and the class par is 79, something is probably wrong with the 105, and if I use a variant for the day, I could mess up not only that race, but all the others on the card.

That Beyer doesn't know (or didn't publish) how to do this doesn't mean that it is not required to win in this more competitive era. You didn't have to throw into double-coverage in the NFL thirty years ago either, but now you can get cut from the team if you don't know how.

Red Queen 101

[cj]

Originally Posted by WhyWhyWhy

Or merely appear to. That +10 change in the variant can be hidden if the pace of the later races is a "-10 pace variant."

I thought I said I consider pace, but maybe not. I make speed figure variants based on the other horses and the other races for each race. I then a make a pace variant. I then go back and look at the speed figure variants for each race again with the pace figures alongside. Sometimes I adjust them again, and sometimes not.

One example was War Pass in the BC Juvenile. Given how Pyro and him had run previously, it was easy to see how he could be given him a huge speed figure, and I initially did. However, when I then saw the pace figure, I knew it was simply not possible that the horses ran that fast. So, that race was considered alone.

There are times it makes sense to adjust all races, sometimes only races of a certain distance, and sometimes only one race.

To me, it doesn't make sense to disregard the other races as a possible source of information. I'm not saying you must use them, but most times they help. Making figures is a very complicated process, so the more information you have,the better.

[WhyWhyWhy]

Originally Posted by cj

People that elevate slow race performances are asking for trouble. We both know one popular figure maker that does this. The biggest problem is you boost the also rans to look much better than they are.

Only if you us a linear figure method that always gives the winner a higher performance rating than the also-rans. That too is flawed, as no two horses are running the same race. The beaten-lengths chart assumes equal trips, which also never occurs, and is another "blanket adjustment" that is used for convenience.

The best example of this is the 1978 Jockey Club Gold Cup. Does anyone really think Exceller's "speed figure" should have been higher than Seattle Slew? Maybe his figure should be, but not his performance rating. The same could be said if you're comparing Sham's 1973 Belmont to Twice a Prince.

The calibration guy talked about the "high quit" horses (that was Ragozin's innovation, btw) have very high pace figures but pack it in, then improve. That pace figure can be merged with the speed figures to create a performance figure that makes the "improvement" next out expected. If the high-quit race proved the horse's superior ability, any figure that doesn't reflect that ability in the race it supposedly appeared is working with a flawed figure method. We can make these adjustments subjectively, but we can also make variants subjectively, yet we don't. Guys like Beyer (and CJ) force other guys to delve deeper into these mathematics in order to get an edge.

If I jog 50 meters next to an Olympic sprinter, then we run the last 50, he'll win by about two seconds. If we sprint the full 100 meters, he wins by five seconds. Not only did we not really "race" 100 meters, I had much more left in the tank for the distance we did "race" and thus could stay closer.

Actually his trainer bet $1,000 to show on you so the horse ran off the board.

Seriously, CJ just answered his own question by noting that the sprinter could have run faster. One would up the sprinter's figure and leave CJ's untouched, to reflect the superior runner more or less mailing it in. CJ will also have several other races without skewed pace scenarios to use as a backdrop.

In the first example, what do you give the Olympic sprinter on a speed figure scale? The pathetic rating he would deserve and just remember the circumstances, or boost him to his "usual" race. If you do that, suddenly it looks like I could hang with top class female Olympians, which I certainly cannot.

Why is CJ assuming I'm going to bump up both figures? Because Beyer told me I had to?

Any blanket adjustment is generally bad. Just because one could win with them in the past, doesn't make them theoeretically sound, and those looking for new frontiers of profitability are going to zero in on any weakness in order to exploit it. Track variants and pace figures, as well as blanket beaten-lengths adjustments, are three huge vulnerabilities that almost every number-crunching handicapper of this era has.

When they put finish lines up at the quarter and half-mile poles, I might begin using pace figures differently.

[WhyWhyWhy]

Originally Posted by cj

They are much more reliable than unadjusted time.

So is the DRF's method.

Performance ratings are much more reliable than raw speed figures. All that's required to make a performance rating as a premise is that horses don't improve or decline drastically over time (except for predictable reasons). A group of 90s with a 70 speed figure can still be presumed to be 90s, even if Beyer's charts say otherwise for that race. They still competed. They're still the same horses, and they still got the same conditioning/class/whatever benefits from the race. And -- get this - the people who bet on them still won and lost the same amount of money, with the same purses paid.

The "different" pace is a minor difference at best.

[WhyWhyWhy]

Originally Posted by Tom

This is a lot of nutso stuff...everyone KNOWS Andy Beyer only buys and drives HARNESS HORSES!

His new book is titled Beyer on Bikes!

Damn, Robert Duvall could so play him in a movie!

[WhyWhyWhy]

Originally Posted by cj

I thought I said I consider pace, but maybe not. I make speed figure variants based on the other horses and the other races for each race. I then a make a pace variant. I then go back and look at the speed figure variants for each race again with the pace figures alongside. Sometimes I adjust them again, and sometimes not.

One example was War Pass in the BC Juvenile. Given how Pyro and him had run previously, it was easy to see how he could be given him a huge speed figure, and I initially did. However, when I then saw the pace figure, I knew it was simply not possible that the horses ran that fast. So, that race was considered alone.

There are times it makes sense to adjust all races, sometimes only races of a certain distance, and sometimes only one race.

To me, it doesn't make sense to disregard the other races as a possible source of information. I'm not saying you must use them, but most times they help. Making figures is a very complicated process, so the more information you have,the better.

I understand your methodology, but you don't mix the two ratings into a single performance rating, so you are doing what could be done automatically "by hand." Someone who looks at the teletimer is "considreing final time" but I wouldn't call them a speed handicapper. If you did, you wouldn't be saying "113 speed 88 pace" but instead that would convert to a single number. It does not.

Once you inject pace into speed figures, it would not make sense to use a linear, blanket adjustment for the entire field. Even though each horse's pace rating changes due to its unique trip, the "pace figure" itself is made with a blanket, just as the speed figure is. The two numbers you give are both accurate, but they also yield a third number, once one figures out the correct relationship between the two. In the 113/88 example above, what would that third number be?

I should add that I make pace ratings much differently, and they contain a very high class component.

[cj]

I have to be honest, you are too vague for me to try to figure out what you are trying to say. It is 1:30 am and I'm about 2/3rds through a 12 hour shift, so maybe it is just me. I'll try to hit a few things then let it go for now.

The best thing about performance figures is that you can usually be more confident a horse is improving or declining with them than using speed figures alone.

In my example, I was saying it was tough to make a figure for a slow paced race. Sure, you can give the winner what you expect and use a different adjustment for the others. That makes sense...sometimes. What do you do when the horses have little or no history? If the Pyro race was a maiden special weight many of the horses had just a few races, maybe no routes, now what do you do? I prefer to give the horse exactly what I think they ran that day, not what the horse could maybe run if asked.

I don't get " The 'different' pace is a minor difference at best." Sometimes the pace is a minor factor, sometimes it is a huge factor. I'm sure you know this, just not sure what you are trying to say.

The biggest flaw with using only what happened in a specific race to make figures for that race is this...what do you do with the other races with little or no information to go on? Class pars? Please, those are one of the biggest frauds ever passed on to the public. I'd much rather use information from surrounding races than some made up par with no relevance to today's race.

Look at it all...the times of the race, the histories of the horses in the race, all the other races, track maintenance if available, weather changes, troubled trips, trainer changes, pressure, and so on. Then decide what is best. Most races aren't that tough, but some are. You have to do the work on those.