Slow Horses

[Ghostzapper04]

So, I've been reading "My $50,000 Year at the Races", published in 1978 recently, and I've noticed from the pp's that Beyer shows in the book, many races have horses running figs in the 100's-90's. I've also noticed that, over the 6 to 7 years, the figs have been in steady decline. This year's Travers was won with an 100 beyer if I'm not mistaken, while an 100 in the Travers 10 years ago would translate to a 6th place finish. I'm wondering why this is. Are horses just plain slower than what we've previously been accustomed to? How much of this is due to the outlaw of steriods and other drugs? Have the beyer figures changed recently?

[therussmeister]

Since no one has responded, I'll take a crack at it.

I consider it a possibility that Beyer's figures for the best horses have inadvertently drifted downward over the years, but I am not at all certain that they have.

The performance gap between 10,000 claimers and graded stakes horses may have narrowed, but not necessarily because stakes horses are slower, but rather claimers are faster. Horses that, in 1978, would have run in 12,500 - 16000 claiming races may now be running for 10,000 due to greatly increased purses along with more racing opportunities at that level. At tracks like Parx I see horses that have earned $75,000 - $100,000 on the year without showing a race above $12,500 claiming in their running lines.

If Beyer's figures for $10,000 claimers have remained static over the years, (I don't know if they have) Stakes horses would earn lower figures, but the reality may be $10,000 claimers should earn higher figures.

[Robert Goren]

When Beyer wrote his book an open 10K claimer was filled slow horses and sore horses. Bute wasn't even legal , let alone lasix in most states. Today the 10k claimer is filled with horses that would have been retired when Beyer wrote his book. They are pretty fast horses who thanks to Lasix won't bled. They are pretty horses who thanks to some marvelous drug(Both legal and illegal) won't feel the pain when they are running. I have not done any research on this in a long time, but my guess is that a 10k claimer winner is at least a second closer to a grade 3 stakes winner today in a 6f race.
Note: by open 10k claimer I mean one that is not restricted by NW of 3L or less. Such things as NW in 6 months do not have the same effect (if any) on the time in a 10K claiming race as they would on the very bottom claiming level at a particular track.

[DeltaLover]

The truth is that for handicapping and gambling purposes the question of the OP is meaningless.

Searching for causes using posterior deterministic arguments that seem rational and applicable only creates theories and conjectures that are impossible to have any empirical prove.

Who was a better horse, Secretariat or Frankel?

Although anyone can have opinion, nobody can prove it and the same applies to the topic of this thread.

[overthehill]

i dont believe the beyer figures are computed the same way. a friend and i both feel that there is now a class component embedded in this calculation somehow.

[cj]

i dont believe the beyer figures are computed the same way. a friend and i both feel that there is now a class component embedded in this calculation somehow.

They are still computed the same exact way they always have been. Beyer just did a segment on Steve Byk's show and discussed exactly how he makes figures.

There are a few reasons I think the figures are shrinking. Some are due to figure making issues, others due to the horses and trainers. We'll never really know though.

[DeltaLover]

Beyer just did a segment on Steve Byk's show and discussed exactly how he makes figures

Is this segmenet available through text or video so we can take a look?

[ceejay]

http://www.thoroughbredracingradionetwork.com/index.php?option=com_events&task=view_detail&agid=1378&year=2012&month=11&day=28&Itemid=35

[DeltaLover]

http://www.thoroughbredracingradionetwork.com/index.php?option=com_events&task=view_detail&agid=1378&year=2012&month=11&day=28&Itemid=35

thx...

[thaskalos]

i dont believe the beyer figures are computed the same way. a friend and i both feel that there is now a class component embedded in this calculation somehow.
I feel the same way.

Beyer can deny it all he wants...but his speed figures are no longer the "pure" speed figures he advertises them to be.

There indeed appears to be some sort of class component in the calculation.

[cj]

I feel the same way.

Beyer can deny it all he wants...but his speed figures are no longer the "pure" speed figures he advertises them to be.

There indeed appears to be some sort of class component in the calculation.

How would he do that? If you are saying he'll break a race out because it just looks "too slow" or even "too fast", I'll agree. But there is no class component.

If you have three horses that have consistently run mid 70s to 80, and the finish fairly close together, what do you do if using the same variant for all other races gives this one a 60?

I look at the pace of the race and see if it was very fast, or very slow, and that may have contributed to the slow time. The reason I do it this way is that I don't think extreme pace scenarios affect all horses the same way. That is the problem with breaking out a race like that.

But if it isn't pace, you almost have to break out the race. The reason is you just can't trust the reported times in this sport. You can't know every wind shift or everything done by maintenance crews at every track in the USA. Even if you get that information, there is no guarantee you know how it will affect final time.

All that said, if you check the race card on any given day and the Beyer figures given, you will see that probably 95% or more follow right in line with the Beyer speed charts. There is nothing to indicate there is any class factor. Try it a few times and let me know what you find.

[thaskalos]

How would he do that? If you are saying he'll break a race out because it just looks "too slow" or even "too fast", I'll agree. But there is no class component.

If you have three horses that have consistently run mid 70s to 80, and the finish fairly close together, what do you do if using the same variant for all other races gives this one a 60?

I look at the pace of the race and see if it was very fast, or very slow, and that may have contributed to the slow time. The reason I do it this way is that I don't think extreme pace scenarios affect all horses the same way. That is the problem with breaking out a race like that.

But if it isn't pace, you almost have to break out the race. The reason is you just can't trust the reported times in this sport. You can't know every wind shift or everything done by maintenance crews at every track in the USA. Even if you get that information, there is no guarantee you know how it will affect final time.

All that said, if you check the race card on any given day and the Beyer figures given, you will see that probably 95% or more follow right in line with the Beyer speed charts. There is nothing to indicate there is any class factor. Try it a few times and let me know what you find.

Allow me to explain...and then tell me what you think:

I don't particularly look for these anomalies, but I have found at least 10 cases over the last couple of years where two horses have run on the same track, on the same day, and at the same distance...but in different races. The races are often only separated by less than an hour of actual time. These two horses would have almost identical final times...but Beyer would award the WRONG horse with the higher figure. One horse would have a final time of 1:10.6 and the other a final time of 1:10.4...but the former horse would be assigned a Beyer figure of 86...while the latter horse would be given a figure of 81. I once saw two horses run identical final times in consecutive races...and yet there was an 8-point Beyer difference between them.

I initially thought that these were mistakes, or that these figures were the result of a split variant. But then I made a curious discovery. In each and every case, the horse who had been given the higher figure, in spite of it's slower time, was running at a higher class level than its seemingly faster counterpart. So I am forced to believe that some sort of class adjustment was in the works.

I know that what I say here might be hard for some to believe...but it is all true. And I have mentioned it here a couple of times in the past.

[cj]

Allow me to explain...and then tell me what you think:

I don't particularly look for these anomalies, but I have found at least 10 cases over the last couple of years where two horses have run on the same track, on the same day, and at the same distance...but in different races. The races are often only separated by less than an hour of actual time. These two horses would have almost identical final times...but Beyer would award the WRONG horse with the higher figure. One horse would have a final time of 1:10.6 and the other a final time of 1:10.4...but the former horse would be assigned a Beyer figure of 86...while the latter horse would be given a figure of 81. I once saw two horses run identical final times in consecutive races...and yet there was an 8-point Beyer difference between them.

I initially thought that these were mistakes, or that these figures were the result of a split variant. But then I made a curious discovery. In each and every case, the horse who had been given the higher figure, in spite of it's slower time, was running at a higher class level than its seemingly faster counterpart. So I am forced to believe that some sort of class adjustment was in the works.

I know that what I say here might be hard for some to believe...but it is all true. And I have mentioned it here a couple of times in the past.

Honestly, unless you checked to see if the variant was split or have a specific example I can't really say. My guess would be that at least 8 of the 10 times you saw it, the variant was split. I no longer check the Beyer figures as often as I did in the past, but when I do, the times a race is simply "broken out" from the others is pretty small, particularly on dirt. Those times they are broken out, I sometimes agree, and sometimes don't. I have noticed one trend I find disturbing, and that is the tendency to downgrade the figures of runaway winners at all classes.

On turf, I agree with you guys. Beyer turf figures have become more like class ratings than speed figures. Final time is often a secondary factor in the figure assigned. On dirt, not so much.

[cj]

Here are some reasons I think Beyer figures have shrunk:


Claiming horses are better due to increased purses. Horses are more readily dropped, and they are also kept in better health. So, assuming the top horses stay relatively stable in ability, the gap has narrowed. Figure makers can either hold the top horses steady and raise the claiming figures, or hold claimers steady and drop the top horses.
The Beyer charts are outdated. Horses don't want to race as long as they did in the past, they aren't bred for it. The goal of Beyer was to equate performances at different distances with his speed charts. He probably did in 1975, but they need updating in a bad way.
Synthetics and increased turf racing have changed the way riders race, carrying over that form to dirt much more often. Snatching back and strangling horses is much more prevalent now than it was even 10 years ago. Horses won't run fast times when they are held for large portions of the race.
Horses start less and train less. They can't possibly be as fit as they were when racing often and training hard was the norm.
I'm sure there are more, but that is a start.

[Stillriledup]

Allow me to explain...and then tell me what you think:

I don't particularly look for these anomalies, but I have found at least 10 cases over the last couple of years where two horses have run on the same track, on the same day, and at the same distance...but in different races. The races are often only separated by less than an hour of actual time. These two horses would have almost identical final times...but Beyer would award the WRONG horse with the higher figure. One horse would have a final time of 1:10.6 and the other a final time of 1:10.4...but the former horse would be assigned a Beyer figure of 86...while the latter horse would be given a figure of 81. I once saw two horses run identical final times in consecutive races...and yet there was an 8-point Beyer difference between them.

I initially thought that these were mistakes, or that these figures were the result of a split variant. But then I made a curious discovery. In each and every case, the horse who had been given the higher figure, in spite of it's slower time, was running at a higher class level than its seemingly faster counterpart. So I am forced to believe that some sort of class adjustment was in the works.

I know that what I say here might be hard for some to believe...but it is all true. And I have mentioned it here a couple of times in the past.

You are 100% correct, Beyer's figures 'skewed' somehow as i've also seen two completely different BF's for horses who ran a simliar time on the same day less than an hour apart, great observations.

[HUSKER55]

In one of Beyer's books he states that his figures are based on $10K Claimers. If class is not involved then why did he mention it?

[cj]

In one of Beyer's books he states that his figures are based on $10K Claimers. If class is not involved then why did he mention it?

This has absolutely nothing to do with what is being discussed.

[cj]

You are 100% correct, Beyer's figures 'skewed' somehow as i've also seen two completely different BF's for horses who ran a simliar time on the same day less than an hour apart, great observations.

So what? That can happen on anybody's figures. It hardly shows that he is changing them due to some class factor.

[thaskalos]

So what? That can happen on anybody's figures. It hardly shows that he is changing them due to some class factor.
Question:

Clear weather and a fast track. The fifth and the sixth races are both run in identical times...but the winners of these two races are assigned speed figures which differ from each other by eight points.

Can you please explain to me how that can be?

[cj]

Question:

Clear weather and a fast track. The fifth and the sixth races are both run in identical times...but the winners of these two races are assigned speed figures which differ from each other by eight points.

Can you please explain to me how that can be?

Maybe that is the exact point where a variant was split. Possible, right?


Like I said, I'm not saying individual races are never broken out. They are, and I've said as much. However, it isn't often. I do it FAR less than Beyer does. If you provide a specific example, I could elaborate more, but until then it is just speculation based on the very limited information you provided. Figure making just isn't that easy sometimes.

[Tom]

So many cases where Beyer "screwed up" but not on real race yet - provide a real race and then discussion makes sense.

[cj]

Maybe that is the exact point where a variant was split. Possible, right?


Like I said, I'm not saying individual races are never broken out. They are, and I've said as much. However, it isn't often. I do it FAR less than Beyer does. If you provide a specific example, I could elaborate more, but until then it is just speculation based on the very limited information you provided. Figure making just isn't that easy sometimes.

Thask,

So, I get nothing in return? Why do I bother?

[PhantomOnTour]

So many cases where Beyer "screwed up" but not on real race yet - provide a real race and then discussion makes sense.
The Damascus Stks and the BrCup Filly sprint were run about an hour apart (both at 7f) and had final times within 0.14 secs of each other but Groupie Doll gets a BSF nine points higher than the Damascus winner.

[cj]

The Damascus Stks and the BrCup Filly sprint were run about an hour apart (both at 7f) and had final times within 0.14 secs of each other but Groupie Doll gets a BSF nine points higher than the Damascus winner.

That race was a VERY tough call. I sided with not breaking the race out, but only because I couldn't find any evidence at all the track changed. In essence, by not breaking it out, I'm giving most of the horses career tops, and these were not lightly raced horses.

So, it kind of depends what you want from the figures. Do you want figures that might indicate an overlay but might also be wrong and cause you bet a horse on a bad figure, or do you want more conservative figures?

I'm more on the bold side, but I also made note of the race and will monitor the horses from the Damascus as soon as they run back. In typical 2012 fashion, not a single horse from the Damascus has run back yet a month later.

[thaskalos]

Thask,

So, I get nothing in return? Why do I bother?

Sorry cj...I didn't mean to ignore you.

It's just that I can't exactly disagree with you, because it is conceivable that a split variant could be the reason for the peculiar figures.

I would like to post a couple of examples...but I don't know how to even begin to look to find them. It's been a while since I last saw one...

[cj]

Sorry cj...I didn't mean to ignore you.

It's just that I can't exactly disagree with you, because it is conceivable that a split variant could be the reason for the peculiar figures.

I would like to post a couple of examples...but I don't know how to even begin to look to find them. It's been a while since I last saw one...

NO worries, but if you see something, bring it up. It is a good discussion and I don't mind giving my opinion on any figures, mine, Beyer, TG, BRIS, or whatever. I know a lot about them all. Even if I don't agree with everything some of them do, I respect their opinions.

[mountainman]

It's become more of a trainer's game, and examining numbers without considering that must lead to misconceptions. To me, the term "super-trainer" connotates not only a staggering win-rate, but also the ability to elicit super performance from unlikely stock.

True, no training method or drug can imbue a talentless horse with the ability to run fast times, but cheap races are littered with horses once capable of big figs, and super-trainers excel at turning the clock back.

Before super-trainers truly came on the scene, the onset of age or infirmity precluded a large percentage of cheap claimers from ever running huge figs.
Now it's not uncommon for aging or apparently unsound horses to run astonishing numbers when in for low tags.

Modern trainers defy father time. Modern trainers banish severe pain. How can those remarkable acts not contort par numbers and parallel times?

[SharpCat]

NO worries, but if you see something, bring it up. It is a good discussion and I don't mind giving my opinion on any figures, mine, Beyer, TG, BRIS, or whatever. I know a lot about them all. Even if I don't agree with everything some of them do, I respect their opinions.


What if the scenario that Thask posted earlier was on firm grass instead of dirt?

[cj]

What if the scenario that Thask posted earlier was on firm grass instead of dirt?

I would look at the same why I would on dirt, to see if an extreme pace scenario had an effect on final time. On turf, it is usually a slow pace that does this. People should remember, horses are ridden to finish first, but not necessarily in the fastest possible time.

Also, turf races are not always reliable when it comes to timing.

[cj]

I have noticed one trend I find disturbing, and that is the tendency to downgrade the figures of runaway winners at all classes.


Beulah, Race 7 on December 1st, was a great example of this. A horse, Final Answer, ran off the screen first off the claim in the 7th race winning by 11. I didn't do the math, but it looks to me like the the variant from the other races would have given this horse a Beyer of at least 10 points higher than the 86 he was given. There was a route the very next race that reverted to the old variant. Final Answer's career best had been a 74.

Oddly enough, the runner up, earned a 75 in his prior start, and a 65 this race. Had the same variant been applied, he would have earned another 75 or so. The show horse, Princini, was given a 51 after earning a 58 in his prior race.

There was another one at Mountaineer on the same day where a horse won by 10 and was a given a 77 with a much different variant than the other races on the card.

For whatever reason, the Beyer guys just seem to arbitrarily knock down these runaway winners. Isn't that when you should expect horses to run huge?

[thaskalos]

Beulah, Race 7 on December 1st, was a great example of this. A horse, Final Answer, ran off the screen first off the claim in the 7th race winning by 11. I didn't do the math, but it looks to me like the the variant from the other races would have given this horse a Beyer of at least 10 points higher than the 86 he was given. There was a route the very next race that reverted to the old variant. Final Answer's career best had been a 74.

Oddly enough, the runner up, earned a 75 in his prior start, and a 65 this race. Had the same variant been applied, he would have earned another 75 or so. The show horse, Princini, was given a 51 after earning a 58 in his prior race.

For whatever reason, the Beyer guys just seem to arbitrarily knock down these runaway winners. Isn't that when you should expect horses to run huge?

The projected variant approach is only as accurate as the figure-maker's competence will allow.

[Jeff P]

One thing I would add... After the Eight Belles incident, a number of track surfaces were purposely made slower.


-jp

.

[cj]

One thing I would add... After the Eight Belles incident, a number of track surfaces were purposely made slower.


-jp

.

Yes, but that should be able to be accounted for by the daily variant. As an example, Game On Dude just won at Hollywood on probably the slowest track in the history of Southern California racing, but was given a Beyer of 109.

[Jeff P]

True, a daily variant "should" be able to account for the track surface. But that's not really what I was getting at.

Imho, the purposeful slowing of track surfaces has had the effect of changing the way racing is conducted.

On old school dirt surfaces, the fastest horses were typically sent from the gate - and if good enough - created enough separation between themselves and the closers - that in order to catch them the closers had to make up a lot of ground. In my opinion this contributed greatly to speed being considered the universal bias.

Purposefully slowed surfaces brought us a completely different type of racing. Because the surface itself negatively impacts speed - when the gate opens, entire fields are hard held and allowed to gallop (well bunched and almost in slow motion) to the top of the stretch. At that point they are cut loose for a short sprint home.

Surfaces that create this type of racing drive me crazy as a bettor. (I tend to avoid them.) I've occasionally dabbled at making my own figures and have to think surfaces that create this type of racing might well drive a figure maker crazy too.

Probably no way to answer it accurately... But considering the two completely different styles of racing involved: Is that 109 on the slowest track in the history of SOCAL racing really comparable to a 109 earned in a truly run race on an old school dirt surface?



-jp

.

[PhantomOnTour]

True, a daily variant "should" be able to account for the track surface. But that's not really what I was getting at.

Imho, the purposeful slowing of track surfaces has had the effect of changing the way racing is conducted.

On old school dirt surfaces, the fastest horses were typically sent from the gate - and if good enough - created enough separation between themselves and the closers - that in order to catch them the closers had to make up a lot of ground. In my opinion this contributed greatly to speed being considered the universal bias.

Purposefully slowed surfaces brought us a completely different type of racing. Because the surface itself negatively impacts speed - when the gate opens, entire fields are hard held and allowed to gallop (well bunched and almost in slow motion) to the top of the stretch. At that point they are cut loose for a short sprint home.

Surfaces that create this type of racing drive me crazy as a bettor. (I tend to avoid them.) I've occasionally dabbled at making my own figures and have to think surfaces that create this type of racing might well drive a figure maker crazy too.

Probably no way to answer it accurately... But considering the two completely different styles of racing involved: Is that 109 on the slowest track in the history of SOCAL racing really comparable to a 109 earned in a truly run race on an old school dirt surface?



-jp

.
but aren't the closers affected by the slow surface too?
all it does is blunt early speed of E types and not blunt the early "speed" of closers?
IOW...if E types are going slower early aren't the closers going slower early also?

[cj]

but aren't the closers affected by the slow surface too?
all it does is blunt early speed of E types and not blunt the early "speed" of closers?
IOW...if E types are going slower early aren't the closers going slower early also?

Generally, but not always, slower surfaces favor horses that run more evenly. While the actual times might be slower, it doesn't change the way horse expend their energy. If anything, slower surfaces tend to make speed horses expend it even less efficiently.

Surfaces that create this type of racing drive me crazy as a bettor. (I tend to avoid them.) I've occasionally dabbled at making my own figures and have to think surfaces that create this type of racing might well drive a figure maker crazy too.

I mentioned this as one of my "reasons" for horses not running as fast. However, I tend to think it is the proliferation of turf racing and synthetic surfaces that have led to this type of racing more than slower dirt tracks. We'll really never know for sure though.

[eurocapper]

Perhaps whatever the cause is related to the lack of triple crown winners. If there are no real standouts no one will win all three. In that case the figures would be right at least.

[cj]

but aren't the closers affected by the slow surface too?
all it does is blunt early speed of E types and not blunt the early "speed" of closers?
IOW...if E types are going slower early aren't the closers going slower early also?

Another way to look at it is that slower tracks are increasing the length of the race, not a good thing for speed horses. Yes, the distance is the same, but the time is slower meaning the race is, in effect, longer.

[Maximillion]

NO worries, but if you see something, bring it up. It is a good discussion and I don't mind giving my opinion on any figures, mine, Beyer, TG, BRIS, or whatever. I know a lot about them all. Even if I don't agree with everything some of them do, I respect their opinions.


cj
Im familiar with the charts Beyer uses for beaten lengths at different distances..I believe BRIS does this a little differently.
If I can ask, do you have a preference between the two in regards to this?

[cj]

cj
Im familiar with the charts Beyer uses for beaten lengths at different distances..I believe BRIS does this a little differently.
If I can ask, do you have a preference between the two in regards to this?

I believe both do BLs that count more the shorter the race, which makes sense. Beyer, however, has tweaked his BL charts for turf races and his time chart for synthetics.

I have different charts for different surfaces. Both of the aforementioned would be better if they did the same.

[Irish Boy]

I might be remembering incorrectly, but wasn't the original "E=MC^2" of Beyer's formula that the average winner of a $10,000 claimer = 1:12 at 6F = 80? (I might very well be wrong; I couldn't find my copies of my Beyer books).

If that's true, I could see two problems, and I'm sure people more savvy than me could find more.

1.) Maybe the breed really is improving, at least in sprints. As horses reach their genetic potential at the high end, the difference between stakes horses and claimers will diminish. As the difference diminishes, you have two options: upgrade claimers, or downgrade stakes horses. There are a lot more claimers, so it makes sense to downgrade stakes horses.

For the record: I understand that raw times are very, very flawed. But it is my understanding (I could be wrong) that raw times are staying steady or improving at the top end. And 1:12 seems dreadfully slow for a 6F at most tracks (again, maybe i'm wrong). Maybe the gap is just shrinking.

2.) The slots problem. I know CJ has talked about this before. The average 10k claimer is better because the purses are so out of whack. This throws off everything down the line.

[Maximillion]

I believe both do BLs that count more the shorter the race, which makes sense. Beyer, however, has tweaked his BL charts for turf races and his time chart for synthetics.

I have different charts for different surfaces. Both of the aforementioned would be better if they did the same.

Gotcha,thank you....
Im not ALL about figures but I do take a good look at them....as far as Beyer I guess all SFs need to be taken with a grain of salt on turf and AW.

[thearmada]

As a player who is trying to use figures more, I find this thread quite interesting. Definitely interesting stuff.

[Cratos]

So, I've been reading "My $50,000 Year at the Races", published in 1978 recently, and I've noticed from the pp's that Beyer shows in the book, many races have horses running figs in the 100's-90's. I've also noticed that, over the 6 to 7 years, the figs have been in steady decline. This year's Travers was won with an 100 beyer if I'm not mistaken, while an 100 in the Travers 10 years ago would translate to a 6th place finish. I'm wondering why this is. Are horses just plain slower than what we've previously been accustomed to? How much of this is due to the outlaw of steriods and other drugs? Have the beyer figures changed recently?

This is a very interesting thread, but in all due candor and with respect to the thread starter topic, “Slow Horses” using the ubiquitous and flawed speed figure/variant concept is misdirected in many ways in trying understand the concept of “slow horses.”

Racehorses run fast or slow based upon their inherent ability (stride rate and length) and their energy efficiency; and this can be seen across all levels of horses. The reason that a graded stakes horse typically run faster at the same distance than a lowly claimer is because of “class grouping” based on inherent ability.

However breeding is not an exact science and some well-bred horses have not lived up to their performance expectation and on the contrary there have been horses with modest pedigree that have been winners at the highest of stakes levels.

Therefore what is a track variant? The so-called “track variant” by definition would be defined as a variant which measure how data distributes itself about the mean (par) or expected value of a racetrack surface and this variation is the resistance to speed driven by the coefficient of friction.

A racetrack is a static body; it doesn’t speed up or slow down. The coefficient of friction between the horse’s hooves and the track’s surface changes one way or another which allows a healthy horse in fit condition to run faster or slower with respect to its inherent ability. Whereas the horse is a dynamic body and the speed relationship between the two comes through their coefficient of friction interaction.

There are many factors that change the track’s resistance to speed, but the predominant ones are: (1) Environmental conditions, (2) Maintenance, and (3) Usages..

A horse being a dynamic body (i.e., has movement) is equally constrained predominately by: (1) Environmental conditions, (2) Health, (3) Physical Fitness, and (4) Human Interactions. There are many other factors, but the aforementioned ones tend to predominate the adversity or success of a horse’s performance and taken combinatorially with the factors affecting the race track surface there could be as many as 5040 possible outcomes; hence no unique variant.

For those heavily into speed handicapping I suggest you look at turn impact, especially angular velocity and centripetal acceleration.
Also understand that a limit in horizontal force should result in a trade-off between acceleration in the direction of movement and centripetal acceleration since as vectors they combine to a single horizontal acceleration

[Magister Ludi]

So, I've been reading "My $50,000 Year at the Races", published in 1978 recently, and I've noticed from the pp's that Beyer shows in the book, many races have horses running figs in the 100's-90's. I've also noticed that, over the 6 to 7 years, the figs have been in steady decline. This year's Travers was won with an 100 beyer if I'm not mistaken, while an 100 in the Travers 10 years ago would translate to a 6th place finish. I'm wondering why this is. Are horses just plain slower than what we've previously been accustomed to? How much of this is due to the outlaw of steriods and other drugs? Have the beyer figures changed recently?

http://espn.go.com/horse/columns/misc/1923297.html

http://www.johnpiesen.com/archive.asp?id=00&uniqID=1482

[traynor]

This is a very interesting thread, but in all due candor and with respect to the thread starter topic, “Slow Horses” using the ubiquitous and flawed speed figure/variant concept is misdirected in many ways in trying understand the concept of “slow horses.”

Racehorses run fast or slow based upon their inherent ability (stride rate and length) and their energy efficiency; and this can be seen across all levels of horses. The reason that a graded stakes horse typically run faster at the same distance than a lowly claimer is because of “class grouping” based on inherent ability.

However breeding is not an exact science and some well-bred horses have not lived up to their performance expectation and on the contrary there have been horses with modest pedigree that have been winners at the highest of stakes levels.

Therefore what is a track variant? The so-called “track variant” by definition would be defined as a variant which measure how data distributes itself about the mean (par) or expected value of a racetrack surface and this variation is the resistance to speed driven by the coefficient of friction.

A racetrack is a static body; it doesn’t speed up or slow down. The coefficient of friction between the horse’s hooves and the track’s surface changes one way or another which allows a healthy horse in fit condition to run faster or slower with respect to its inherent ability. Whereas the horse is a dynamic body and the speed relationship between the two comes through their coefficient of friction interaction.

There are many factors that change the track’s resistance to speed, but the predominant ones are: (1) Environmental conditions, (2) Maintenance, and (3) Usages..

A horse being a dynamic body (i.e., has movement) is equally constrained predominately by: (1) Environmental conditions, (2) Health, (3) Physical Fitness, and (4) Human Interactions. There are many other factors, but the aforementioned ones tend to predominate the adversity or success of a horse’s performance and taken combinatorially with the factors affecting the race track surface there could be as many as 5040 possible outcomes; hence no unique variant.

For those heavily into speed handicapping I suggest you look at turn impact, especially angular velocity and centripetal acceleration.
Also understand that a limit in horizontal force should result in a trade-off between acceleration in the direction of movement and centripetal acceleration since as vectors they combine to a single horizontal acceleration

One of the problems with labeling components of a horse race with numerical values can easily be determined by spending some time at the track, watching the horses critically from paddock to post-race cool down. It doesn't take long to discover that the underlying premise of "putting a number on it" is that each horse in each race is doing the best it can. Astute observation of the actual event (the race) will often indicate otherwise.

[cj]

One of the problems with labeling components of a horse race with numerical values can easily be determined by spending some time at the track, watching the horses critically from paddock to post-race cool down. It doesn't take long to discover that the underlying premise of "putting a number on it" is that each horse in each race is doing the best it can. Astute observation of the actual event (the race) will often indicate otherwise.

Why can't you "put a number on it" while also noting the horse wasn't doing the best it can?

[BIG49010]

Why can't you "put a number on it" while also noting the horse wasn't doing the best it can?

Along this same line, does anybody give more credit to the horse that wins?

I have heard Brad Free I believe on HRTV metion this, but I have never really looked into it.

[cj]

Along this same line, does anybody give more credit to the horse that wins?

I have heard Brad Free I believe on HRTV metion this, but I have never really looked into it.

I don't give more credit, but I do not that a horse won and make it a point to note when the horse "might" have had more in the tank. That is a double edged sword though. Usually easy winners won "easily" because they dominated a field and unless the catch a similar group, they won't run faster when facing tougher competition. I find lightly raced horses are by far the best group for finding these "could have run faster" types.

[traynor]

Why can't you "put a number on it" while also noting the horse wasn't doing the best it can?

I think that might take more direct (and highly subjective) observation and analysis than most bettors are willing to do. It is not just that the winner had something left in reserve, but that any or all of the other entries may have been influenced to exert less than optimal effort.

With all due respect to management (who seem to believe that cavalry charges to the wire are more exciting, or that bettors who notice less than vigorous encouragement of "their" horse by the jockey might be distressed), I think most competent jockeys (the majority of the jockey colony) have little trouble distinguishing between a race in which their horse has a legitimate shot at winning or placing, and a race in which it does not.

[Cratos]

Why can't you "put a number on it" while also noting the horse wasn't doing the best it can?

It is not that you cannot put a number on “it” with “it” being identified as a “speed figure” or “variant” in this context. It is because both become single point estimates which are statistically meaningless in my opinion.

Let’s assume that the speed figure/variant (single point estimates) are obtained by selecting a suitable statistic and their values is computed from a given data sample. Therefore these selected statistics are called the point estimators of X (whatever “X” is defined to be.)

But shouldn’t the speed figure/variant be good estimates that provides a clear enough view of the horse’s performance reality for the race that it allows the bettor to make a good decision about how to wager to possibly hit his/her wagering goals?

However the speed figure/variant (single point estimate) as presently constructed doesn't provide a clear view, I believe that the bettor has to have a clear picture of the horse’s expected performance to understand the best case and the worst case performance estimate. Given such modification, the bettor’s probability of success might be illustrated like this:

Given a speed figure/variant (single point estimate) that was initially estimated for horse X to be an 85 speed figure" is meaningless, then how should the speed figure/variant be stated? I would suggest using ranges and rephrasing the statement to "the estimated speed figure/variant for horse X should be 85 ± 5.

Sometimes, when uncertainty is big, the range between best and worst case can be significant, but wide range is not a result of bad estimation, it's a result of lack of definition

[Delta Cone]

It is a good discussion and I don't mind giving my opinion on any figures, mine, Beyer, TG, BRIS, or whatever. I know a lot about them all. Even if I don't agree with everything some of them do, I respect their opinions.

As Thaskalos correctly points out, "The projected variant approach is only as accurate as the figure-maker's competence will allow."

My question then is: Which is preferred? A projected variant approach (Beyer) or a more mechanical/computerized approach (BRIS)?

Both approaches have their merits...and drawbacks. A "human" created figure should catch blatant timing errors and other figures that seem unusually flawed. But there is still the issue of the figure-maker's competence and consistency (given that multiple figure-makers are involved with producing Beyer figs).

And of course a computer is more objective, but it cannot easily make the subtle distinctions and judgement calls that seem to be necessary in figure-making.

[cj]

As Thaskalos correctly points out, "The projected variant approach is only as accurate as the figure-maker's competence will allow."

My question then is: Which is preferred? A projected variant approach (Beyer) or a more mechanical/computerized approach (BRIS)?

Both approaches have their merits...and drawbacks. A "human" created figure should catch blatant timing errors and other figures that seem unusually flawed. But there is still the issue of the figure-maker's competence and consistency (given that multiple figure-makers are involved with producing Beyer figs).

And of course a computer is more objective, but it cannot easily make the subtle distinctions and judgement calls that seem to be necessary in figure-making.

Without question, human created figures are WAY better than those done by a computer alone. I use both, but I also have learned how to program a computer to catch things that need human attention.

[cj]

I think that might take more direct (and highly subjective) observation and analysis than most bettors are willing to do. It is not just that the winner had something left in reserve, but that any or all of the other entries may have been influenced to exert less than optimal effort.

With all due respect to management (who seem to believe that cavalry charges to the wire are more exciting, or that bettors who notice less than vigorous encouragement of "their" horse by the jockey might be distressed), I think most competent jockeys (the majority of the jockey colony) have little trouble distinguishing between a race in which their horse has a legitimate shot at winning or placing, and a race in which it does not.

Regardless, I still want to know how fast a horse ran. I much rather know a horse dealt with situations A, B, and C AND ran a figure of 80 than simply know a horse dealt with situations A, B, and C alone. You could tell me 10 reasons why Georgia should have beaten Alabama in football, but if next week they are playing the Atlanta Falcons, they won't mean anything, just as if they were playing the Georgia high school football champs.

[cj]

It is not that you cannot put a number on “it” with “it” being identified as a “speed figure” or “variant” in this context. It is because both become single point estimates which are statistically meaningless in my opinion.


Who makes them single point estimates? What one factor from any race could ever be that?

Speed figures have only ever purported to do one thing. They allow for comparison of how fast horses ran to the wire over different days, tracks, surfaces, and distances. Any other uses you want to assign them are of your own choosing.

[Cratos]

Who makes them single point estimates? What one factor from any race could ever be that?

Speed figures have only ever purported to do one thing. They allow for comparison of how fast horses ran to the wire over different days, tracks, surfaces, and distances. Any other uses you want to assign them are of your own choosing.

If you agree that the speed figure is a statistic, then by definition it is a single point estimate. Also if you can prove that you can independently discern the speed of a racehorse using the speed figure methodology as presently constructed by you or anyone else you deserve the Nobel Prize for applied physics and I will champion your cause.

[Maximillion]

This is my rather crude rule of thumb regarding "easy winners"-
If there is a recent race that is "better" I would be inclined to use that one as probable performance-I also crudely look at the odds the horse has been going off at recently.

If he is one of those "types" that are always overbet(like many frequuent winers) I look elsewhere....if not Im interested.

[cj]

If you agree that the speed figure is a statistic, then by definition it is a single point estimate. Also if you can prove that you can independently discern the speed of a racehorse using the speed figure methodology as presently constructed by you or anyone else you deserve the Nobel Prize for applied physics and I will champion your cause.

Tell me another performance related "single point estimate" that beats speed figures from any one race? I want to hear one thing, be it something to do with raw time, or position gains or losses, or finish position, anything, that does better than a quality speed figure.

For the record, I think it is a measurement, not a statistic. How fast a football players runs the 40 is a measurement, how many yards he gains is a statistic.

[eurocapper]

I'd have to say it's more like card games than other sports, since the main thing is the money. I believe there may be an inverse relationship between speed and class in a given type of racing. If the former applies to it then the latter's role is reduced and vice versa. Pace could apply to both (for class racing when it's made ready-to-run more than as performance).

[Cratos]

Tell me another performance related "single point estimate" that beats speed figures from any one race? I want to hear one thing, be it something to do with raw time, or position gains or losses, or finish position, anything, that does better than a quality speed figure.

For the record, I think it is a measurement, not a statistic. How fast a football players runs the 40 is a measurement, how many yards he gains is a statistic.

To answer your question directly I would contradict myself because I have stated and illustrated in my prior posts in this thread that the single point estimate are meaningless in predictive statistics. To continue this debate would be useless and I hope you don’t take that as an affront.

However I will say that the single most factor constraint on a horse’s winning performance is CLASS which I have defined before on this site as ”the inherent ability of a horse to win consistently at a given level regardless of conditions.” The problem with CLASS is that it is non-parametric and difficult to find. In the words of Pittsburgh Phil: “find class and bet it.”

A speed figure as it is used is a datum and whether it is assumed or given conclusions can be drawn from it, by definition it is a statistic.

Additionally, all statistics are measurements of some sort, but all measurements are not statistics.

[cj]

To answer your question directly I would contradict myself because I have stated and illustrated in my prior posts in this thread that the single point estimate are meaningless in predictive statistics. To continue this debate would be useless and I hope you don’t take that as an affront.

However I will say that the single most factor constraint on a horse’s winning performance is CLASS which I have defined before on this site as ”the inherent ability of a horse to win consistently at a given level regardless of conditions.” The problem with CLASS is that it is non-parametric and difficult to find. In the words of Pittsburgh Phil: “find class and bet it.”

A speed figure as it is used is a datum and whether it is assumed or given conclusions can be drawn from it, by definition it is a statistic.

Additionally, all statistics are measurements of some sort, but all measurements are not statistics.

Another non-answer, shocking. Surely you realize that single point estimates can be used in conjunction with other factors to find valuable bets.

[bisket]

The projected variant approach is only as accurate as the figure-maker's competence will allow.

this is correct. even the most professional figure makers that are forced to project variants on a regular basis can't get them to be accurate when horses from different tracks race each other for the first time...

[Cratos]

Another non-answer, shocking. Surely you realize that single point estimates can be used in conjunction with other factors to find valuable bets.

Please let's move on because we will never agree and I respect yout right to do so.

[cj]

this is correct. even the most professional figure makers that are forced to project variants on a regular basis can't get them to be accurate when horses from different tracks race each other for the first time...

Of course they can, at least a large majority of the time. It was probably a little tougher in 1970, but in 2012, it isn't hard at all.

[cj]

Please let's move on because we will never agree and I respect yout right to do so.

How can we agree when you never actually say anything?

[Cratos]

How can we agree when you never actually say anything?
You appear to enjoy personal attacks. I have said a lot, but apparently your understanding of quantitative analysis as it relates to horseracing is unique and different and you should be applauded for that unique insight.

[cj]

Gotcha,thank you....
Im not ALL about figures but I do take a good look at them....as far as Beyer I guess all SFs need to be taken with a grain of salt on turf and AW.

There was a GREAT example of this on Wednesday at Gulfstream. There were three turf races run, all at the exact same distance of 7.5f.

R4, 1:31.86, Beyer 80
R6, 1:31.22, Beyer 89
R8, 1:31.74, Beyer 69

Obviously, the last one makes no sense in relation to the other two, and yes it was the cheapest race of the three. It was a maiden claimer for 30k, while the other two were open 25k claimers.

The obvious reason for the difference in the times of the races were the pace calls, the first two being slow and the third fast, but speed figures alone can't capture that. So, I guess rather than just tell us what final time figures the horses really ran, they just make up a number.

The problem with that is extreme pace scenarios don't affect all horses the same way. In the 6th race, adding points to the first two finishers, especially the winner, is probably a better reflection of how well they ran as the overcame a slow pace to complete the exacta. Bt with Beyer figures, they also boost the figures of those horses that set the pace and were run down anyway. Even worse, the horse that ran last actually lost ground late and did no running at all, yet gets an artificial boost to his final time figure for no apparent reason. That makes no sense to me.

[CincyHorseplayer]

There was a GREAT example of this on Wednesday at Gulfstream. There were three turf races run, all at the exact same distance of 7.5f.

R4, 1:31.86, Beyer 80
R6, 1:31.22, Beyer 89
R8, 1:31.74, Beyer 69

Obviously, the last one makes no sense in relation to the other two, and yes it was the cheapest race of the three. It was a maiden claimer for 30k, while the other two were open 25k claimers.

The obvious reason for the difference in the times of the races were the pace calls, the first two being slow and the third fast, but speed figures alone can't capture that. So, I guess rather than just tell us what final time figures the horses really ran, they just make up a number.

The problem with that is extreme pace scenarios don't affect all horses the same way. In the 6th race, adding points to the first two finishers, especially the winner, is probably a better reflection of how well they ran as the overcame a slow pace to complete the exacta. Bt with Beyer figures, they also boost the figures of those horses that set the pace and were run down anyway. Even worse, the horse that ran last actually lost ground late and did no running at all, yet gets an artificial boost to his final time figure for no apparent reason. That makes no sense to me.

You know guys like Beyer and Watchmaker are always talking about how pace affects final time and how predicting the probable pace is a big key to a race and they make their own figures,yet they do 2 things;

1)as you said adjust final times based on pace,or basically try to do our thinking for us by interpreting the pace of a race,instead of letting us draw our own conclusion

2)for guys who are aware of pace,that do make figures,they are baffled when a horse like Trinniberg wins coming into the race with huge speed and pace figs.They are quick to the bias call.

I simply don't see how these guys,who are numbers guys,only look at adjusted final time as the only intrinsic ability of a horse,the rest being subjective.It's as if they've stopped themselves in time and refused to take the next step,make the logical conclusion,and expand the concept of basic ability via numbers.Numbers they deal with.As a readers and fan of theirs over the years,I think it's embarrassing for them.

[jeebus1083]

I simply don't see how these guys,who are numbers guys,only look at adjusted final time as the only intrinsic ability of a horse,the rest being subjective.It's as if they've stopped themselves in time and refused to take the next step,make the logical conclusion,and expand the concept of basic ability via numbers.Numbers they deal with.As a readers and fan of theirs over the years,I think it's embarrassing for them.

Agreed. However, Beyer did tackle the issue of pace in "Beyer on Speed" and while basic, did a decent job. For what it is, Beyers are good tools. However, Beyer figs are to horse racing as Budweiser is to American beers: a mass-produced, basic product.

[cj]

Agreed. However, Beyer did tackle the issue of pace in "Beyer on Speed" and while basic, did a decent job. For what it is, Beyers are good tools. However, Beyer figs are to horse racing as Budweiser is to American beers: a mass-produced, basic product.

The problem with his figures now is that you don't know, without doing research, if the figure has been massaged due to pace issues or not. You may see a horse gets an 80 and set a very fast pace and think wow, if this horse gets an easy trip today he could easily run 85 or 90. But what if that 80 is one of these "adjusted figures" and it was really a 65 or 70?

[Maximillion]

The problem with his figures now is that you don't know, without doing research, if the figure has been massaged due to pace issues or not. You may see a horse gets an 80 and set a very fast pace and think wow, if this horse gets an easy trip today he could easily run 85 or 90. But what if that 80 is one of these "adjusted figures" and it was really a 65 or 70?

Have you noticed this on dirt races too or only turf?

[DeltaLover]

The problem with his figures now is that you don't know, without doing research, if the figure has been massaged due to pace issues or not. You may see a horse gets an 80 and set a very fast pace and think wow, if this horse gets an easy trip today he could easily run 85 or 90. But what if that 80 is one of these "adjusted figures" and it was really a 65 or 70?


I do not understand why as a user of Beyer or any other figure I need to know how it is created.

Any figure can be viewed as a derivative of the past performances that can be evaluated as such.

This means that what all we really need is a procedure to evaluate the effect of a figure based in the outcome of the race.

If we find out that this figure adds significant value we use it or we scrap it otherwise.

To give an example let's assume we have a couple of figures, F1 and F2 and we want to see which one is better in predicting the outcome of a race.

It is easy to create a function receiving as input an array of all the figures, possibly adding more data, such as interval days, distance changes or whatever and optimize it to minimize the delta between the figure ran in the next race minus the calculated one.

Doing the same for both figures we will be able to detect the one that better resembles the outcome.

Again, this is just an example as both the parameters passed as the interpretation of the return value could be very different than what I describe here.

A similar approach can be used to estimate the correlation factor of two figures if we suspect they are measuring the same metric.

Doing some research towards this direction, will reveal that there is little difference in the type of the speed figure we are going to use. Beyer, bris, equibase or whatever else professional figure are all pretty close to each other. This concept has long ago reach its maturity and does not to be the distinctive factor in the creation of a successful approach any more. More R&D is required today towards metrics that still seem obscure, like for example a figure measuring the probability of improvement, a form or class figure or anything that has not been the subject of numerous books and theories and is produced for mass consumption.

[cj]

I do not understand why as a user of Beyer or any other figure I need to know how it is created.

Any figure can be viewed as a derivative of the past performances that can be evaluated as such.

This means that what all we really need is a procedure to evaluate the effect of a figure based in the outcome of the race.

If we find out that this figure adds significant value we use it or we scrap it otherwise.

To give an example let's assume we have a couple of figures, F1 and F2 and we want to see which one is better in predicting the outcome of a race.

It is easy to create a function receiving as input an array of all the figures, possibly adding more data, such as interval days, distance changes or whatever and optimize it to minimize the delta between the figure ran in the next race minus the calculated one.

Doing the same for both figures we will be able to detect the one that better resembles the outcome.

Again, this is just an example as both the parameters passed as the interpretation of the return value could be very different than what I describe here.

A similar approach can be used to estimate the correlation factor of two figures if we suspect they are measuring the same metric.

Doing some research towards this direction, will reveal that there is little difference in the type of the speed figure we are going to use. Beyer, bris, equibase or whatever else professional figure are all pretty close to each other. This concept has long ago reach its maturity and does not to be the distinctive factor in the creation of a successful approach any more. More R&D is required today towards metrics that still seem obscure, like for example a figure measuring the probability of improvement, a form or class figure or anything that has not been the subject of numerous books and theories and is produced for mass consumption.

Beyer figures are final time figures, allegedly. If you are going to blindly use numbers, what you say makes more sense. But if you are using them as part of a more comprehensive approach, it does matter for the reason I pointed out. I'm sure anyone using the figures blindly is not winning, Beyer's or any others.

As for all figures being the same, I disagree, but even if I didn't my argument would remain. There are lots of races with very fast and very slow paces, but only some are "adjusted" while many are not. A comprehensive handicappers would want to know.

[cj]

Have you noticed this on dirt races too or only turf?

It is done on dirt races, but the majority are turf/synth races.

[letswastemoney]

It is done on dirt races, but the majority are turf/synth races.
I remember at Woodbine, I never understood how Joshua Tree earned an equal or higher Beyer (it's too long ago for me to remember the exact numbers) than Siyouma, even though the fractions for the Siyouma race were faster, and had that race been just a little longer, I have to think Siyouma would have finished in a faster final time.

I understand they were not the same distance though.

EP Taylor chart
http://www1.drf.com/displayChart.do?filepath=Download\simulcastweekly\ charts\1210\WO2012101408.pdf

Canadian International
http://www1.drf.com/displayChart.do?filepath=Download\simulcastweekly\ charts\1210\WO2012101410.pdf

[DeltaLover]

As for all figures being the same, I disagree

Is this disagreement based in some analysis or it is just an opinion?

If indeed you have some methodology to compare the effectiveness of various figures can you please present it ?

[Maximillion]

It is done on dirt races, but the majority are turf/synth races.

Thanks....thats a little discouraging (even if its rare on dirt) because it could also have an effect on how you view the others (not to mention the winner)
coming out of those races.

[cj]

Is this disagreement based in some analysis or it is just an opinion?

If indeed you have some methodology to compare the effectiveness of various figures can you please present it ?

It is analysis I have done, but I'm not giving that info away, sorry.

[DeltaLover]

It is analysis I have done, but I'm not giving that info away, sorry.

Why?

Just descibing a comparison althorithm I don't see how can it hurt your IP.

[cj]

Why?

Just descibing a comparison althorithm I don't see how can it hurt your IP.

What is IP?

I just don't see the point in giving away what involved lots of work. I've done a lot of that in the past and to be honest, it is a thankless task. All it does is bring more work for myself because what invariably follows are questions from skeptics or requests for even more data from others. Frankly, I just don't have time for that.

I'll be happy (soon) to post the numbers on my figures. Anyone else can post stats on other numbers as they please.

[bisket]

The problem with his figures now is that you don't know, without doing research, if the figure has been massaged due to pace issues or not. You may see a horse gets an 80 and set a very fast pace and think wow, if this horse gets an easy trip today he could easily run 85 or 90. But what if that 80 is one of these "adjusted figures" and it was really a 65 or 70?

i tend to think they are messaged to make sense of races that show they're off by more than a few points. i'm sure in some cases pace could be the reason, but i'd say that's a 50/50 proposition with races that just don't make sense. you're example of 3 races on the same surface at the same distance on the same day is an anomaly these days..... how can someone make a figure with straight face for a card at full of 2 year old maiden races at churhill? then when those horses move to gulfstream, and it becomes necessary to correlate the figures given at churchill with a figure today that doesn't make sense. i don't think any of this is a figure maker's fault or shows lack of skill. it's just the hand we're dealt from the game today.....

[DeltaLover]

What is IP?



IP= Intellectual Property

http://en.wikipedia.org/wiki/Intellectual_property

[DeltaLover]

Anyone else can post stats on other numbers as they please.

I will try to have something by next week...

[barn32]

Searching for causes using posterior deterministic arguments that seem rational and applicable only creates theories and conjectures that are impossible to have any empirical prove.
My, my, my, my my.

[cj]

The Damascus Stks and the BrCup Filly sprint were run about an hour apart (both at 7f) and had final times within 0.14 secs of each other but Groupie Doll gets a BSF nine points higher than the Damascus winner.

The top two finishers from the Damascus Stakes are in the Malibu. The runner up returned and ran a good second in the Underwood, albeit on Cushion track. This should be interesting.