Moss Pace Figures

[Tom]

Seems I never hear them referenced, is anyone using them?
Anyone figured them out yet?
I haven't really used them much, but looking at them today in the Risen Star, seems that a couple of the horses getting a lot of talk up are coming out of slow paced races

Horse Pace of Race (Best route)
Indyago 83 - 87
Nowhere 87 - 88
Uno Mas 89 - 86
Friesen 89 - 86
Giant O 85 - 89

Giant Oak ran an 83-89, good balance for a stretch out, but can he came from back in the pack or will he be used trying to keep up early?

Looks like the best early pace any have done is 89-85 (Friesen), 87-86 (Uno Mas), and 85-87 (Nowhere to Hide).

[bobphilo]

Seems I never hear them referenced, is anyone using them?
Anyone figured them out yet?
I haven't really used them much, but looking at them today in the Risen Star, seems that a couple of the horses getting a lot of talk up are coming out of slow paced races

Horse Pace of Race (Best route)
Indyago 83 - 87
Nowhere 87 - 88
Uno Mas 89 - 86
Friesen 89 - 86
Giant O 85 - 89

Giant Oak ran an 83-89, good balance for a stretch out, but can he came from back in the pack or will he be used trying to keep up early?

Looks like the best early pace any have done is 89-85 (Friesen), 87-86 (Uno Mas), and 85-87 (Nowhere to Hide).


Tom, glad you brought this up. That's something I've noticed from the very begining and not just this weekend. The pace figure for a race is almost always lower than the race figure. There seems to be a difference in either the par or scale used for the final time and fractional splits.

To give an example of the contrast with CJ's figures, with his figures the average pace for a race earning a 96 final figure is about 96. With the Moss figures, the pace figure for the same race seems several points lower.

Bob

[proximity]

Seems I never hear them referenced, is anyone using them?
Anyone figured them out yet?


* overall, i give these figures a passing grade. (which is pretty good for me)

*however, i think the plan was too ambitious. shoulda just went with more quality and less quantity. ie. just a single solid 1/2 mile pace figure for all the tracks and distances.

*i am somewhat skeptical of the "formula" for attempting to equate the pace figures for different distances. my hunch is that it may be too tight, causing more "turnback" or "cutback" horses to look like solid early speed contenders.

*here is some par info extracted from friday's delta card. for the 1/2 mile 1 length=1.6 points. all info is for races that earned a 79 at delta. (1/4, 1/2, 3/4, fin)

5f........... 85...81...xx...79
6 1/2f...... 81...80...xx...79
1m ..........74...78...79...79
1m 1/16... 70...75...78...79

i'm not a big student of 1/4 mile times, but they seem to make sense to me. however, there is less than 1 length difference between the half mile speed for 5f and 6 1/2f horses and slightly less than 3 lengths difference between 5f and 1mile horses...... much too narrow to me. also since these figures are meant to reflect velocity, ALL of the par figures should be LOWER than the final figures. for example, the mile pace par for the 2nd call is 79 which is implying the average pace setter in these races is running as fast as horses that run 6f and finish with a 79 final figure. if this was true, route pace setters would be turning back and winning 6f races all the time.

*since p-a is the premier handicapping site on the net, it would be nice to hear from moss, crist, beyer...... outside of blogs and occasional articles.... ie. it would be nice to see the form have some kind of presence here.

just some of my moss/drf thoughts.

proximity

[proximity]

The pace figure for a race is almost always lower than the race figure.

velocity wise, this is actually the way it should be.... and even more than it is with the moss figures (explanation in above post)

[bobphilo]

*also since these figures are meant to reflect velocity, ALL of the par figures should be LOWER than the final figures. proximity

If the figures only purpose were to reflect velocity, the pace figures would be higher since velocity is usually faster earlier at the pace call.
What I want from a pace figure is that it refelect how much faster or slower than par the pace call was in relation to the final time. In a route race I want to know how much faster or slower than par the 1/2 and 3/4 rating were in relation to the given final time. With the Moss ratings, it's almost like he's using the par for a 6 furlong sprint in rating the 3/4 pace rating in a route race and that expains why the ratings are so slow.
Ironically, the supercripts do a better job of that by telling you how fast or slow they went at the pace call based on the par for the SPLIT rather than treating it as the final time for a 6 furlong race, which would obviously be lower.
I think that Moss is trying to do a good job of providing accurate ratings but there is an error in the scale and format which makes the pace figures lower than they should be.
I agree further clarification is needed from the DRF and Moss beyond just saying that the higher the rating the faster the pace or race.

Bob

[proximity]

If the figures only purpose were to reflect velocity, the pace figures would be higher since velocity is usually faster earlier at the pace call.


perhaps i shouldn't have used the term velocity. maybe "speed" would be the better word here.

but if we're trying to reflect the speeds that the horses are moving for a given distance(and moss is), then any pace call should always be lower than the final number because the horses are rating at the pace call.

so the pace par for a route race with a final figure of 79 should be lower than 79. at the 3/4 call the horses aren't putting in equal effort to a final time 79 in a 6f race. they are still being rated. again, if a group of horses were running this fast, then all the route horses would turnback to 6f and win with no problem.

raceshape wise for the individual horse, you are definitely better off using cj's. even moreso since his reflect more than just the raceshape of the pace, but also incorporate how the horse performed vs the "win energy" or bias of the track on that particular day.

[cj]

but if we're trying to reflect the speeds that the horses are moving for a given distance(and moss is), then any pace call should always be lower than the final number because the horses are rating at the pace call.


They may be rating, but they are certainly going faster than they are at the end of most races.

[bobphilo]

perhaps i shouldn't have used the term velocity. maybe "speed" would be the better word here.

but if we're trying to reflect the speeds that the horses are moving for a given distance(and moss is), then any pace call should always be lower than the final number because the horses are rating at the pace call.

so the pace par for a route race with a final figure of 79 should be lower than 79. at the 3/4 call the horses aren't putting in equal effort to a final time 79 in a 6f race. they are still being rated. again, if a group of horses were running this fast, then all the route horses would turnback to 6f and win with no problem.

raceshape wise for the individual horse, you are definitely better off using cj's. even moreso since his reflect more than just the raceshape of the pace, but also incorporate how the horse performed vs the "win energy" or bias of the track on that particular day.

I think I see what you're trying to say. The pace call is not the final call and when taking rating into account he would earn a lower speed rating for the 6F, or pace call, in a route race than he would in a a 6F sprint. Agreed. However the final time for a 6F sprint is an unfair standard when calculating a 6f pace call in a route.
The problem is we're not trying to judge a horses time at the 3/4 pace call of a race as if it were a 6f sprint. Why?, because this is not a 6F sprint we are evaluating but the 3/4 split of a route race. We are not interested in how the horse's pace rating in a route compares to a final time in a 6f sprint, but in how it compares to the par time for the 3/4 pace call in a route. If you use the final time for a 6F sprint as your standard for the pace call of a route, of course you will get an unfairly low rating. That appears to be the problem with the Moss pace ratings.

Bob

[proximity]

They may be rating, but they are certainly going faster than they are at the end of most races.

what i am saying is that for every other set distance within the moss ratings, the further said set distance is from the finish of the race, the lower the moss pace par will be. see the chart i posted above for 1/4 and 1/2 mile pars for races from 5f to 1 1/16 at delta.

the pace figures are all going in the same direction.

so we can compare 1/4 and 1/2 mile ratings earned across distances with the moss figures (although i still think they are a little too close)..... however we cannot compare 3/4 ratings across all distances because for some ratings the moss 6f par is higher than the final par. why not go the whole way and have everything match up? that is all i'm trying to say.

[proximity]

. We are not interested in how the horse's pace rating in a route compares to a final time in a 6f sprint,....

"we" may not be, but mr moss is.

the speed of the pace figure is his first priority.... ahead of the race shape.
see his answer to the "summer doldrums" question on the drf site.

i'm not saying any one way of calculating pace figures is better than another here and i do understand your frustration with the moss figures being potentially cumbersome for what you're trying to do in your handicapping.

but overall, he's went one direction with the figures: that a 76 half for a 1 1/8 race equals a 76 half for a 6 1/2f race and equals a 76 half for a 5f race... and all i'm suggesting here is that he goes all the way and makes these numbers equal an "all out" 76 half for a 4f race.... which of course would be the final time of the race.

hypothetical moss style par chart for 76 speed figure:

distance...... 1/2 mile par.... 3/4 mile par.......final figure
4f...............76.................n/a...................76
5f...............75.................n/a...................76
6f...............74.................76............ ........76
7f...............73.................75............ ........76
8f...............72.................74............ ........76
9f...............71.................73............ ........76
10f..............70................72............. ........76

he could still have the raceshapes AND POSSIBLY IN THE FUTURE THE FORM COULD GIVE YOU AN OPTION OF MAKING "ENERGY" STYLE FIGURES WITH THOSE RACE SHAPES.

for example, using my hypothetical chart above: you have a 9f race that goes 71-76 which would be an "even" raceshape. the form could give you an option of making the figure "76-76" which would better represent to you how the horse used his energy for the 1 1/8.

another horse ran 5f in 71-76, which would have a raceshape of "slow 4" according to the chart. (4 points slower than the normal "75" for this final figure) the "energy" option would then show you this as 72-76.... and so on.

sorry for the long post.:)

[classhandicapper]

The Moss figures are correct to the extent that the track variants on any given day are correct.

The reason that the pace figures tend to be lower than the final time figure at the top of the scale is because the difference between the very best and very worst horses in the world is much less at 2F and 4F than at the end of the race.

In other words....

We sometimes see 10K claimers than get run 2F in 21.3 because they are extremely fast, but we never see 10K claimers that can run 6F in 109 because they can't carry their speed.

At the same time , we sometimes see G1 horses run 2F in 22.4 because there's a lack of pace, but we never see them run 6F in 113.

The result is a compression of the difference between the horses at the pace calls that expands at the ends of races.

The result is that the PARS (averages) for each call expressed in Beyer numbers winds up looking something like this at the extremes. (not exact)

Grade 1 107 111 115

10K Claimer 88 84 80

Because of this reality, Moss provides the little + and - indications as to whether a particular pace was fast or slow relative to the final time.

A Grade 1 race that goes in 115 115 115 actually had a faster pace than average.

A 10K claimer that goes in 80 80 80 actually had a slower pace than average.

The reason I know he is right about this is that I did a very detailed study on this 20 or more years ago when I used to make my own Beyer Scale pace figures for every call in NY. In fact, I discussed some of this with Randy Moss when the project was being implemented.

It can be kind of confusing to work with this at the beginning, but if you look at the + and - notations (or verbal comments) it will start making sense. There are also some very good explanations at the DRF web site.

[bobphilo]

So far 2 different explanations have been posted for why the pace figures are consistently lower than the final time figures in the Moss DRF PPs. One will also find that the Moss pace figures are lower in relation to the final figure than other figure makers ratings. The reason that I suspected from the beginning, as put forth by Proximity is that the figures used as the basis for the for the sectional splits of 2,4 and 6 furlongs are not the actual pars for these splits, but rather the pars for final times of races of 2, 4 and 6 furlongs respectively. Obviously, since a final time par for a given distance (such as a 6 furlong sprint) will be faster than the same distance as a fraction of a longer race (6 furlong split of a route race), using this higher standard will produce lower pace figures.
I am still at a loss as to why Moss adopted this method over more effective methods showing energy distribution. The only possible advantage I can see is that they would be more useful than traditional pace figures in predicting a horse’s final time in a sprint from his pace figure when cutting back from a route. In any case, I’ve found it’s easy enough to derive traditional pace figures from Moss’ figures by simply adding or subtracting the superscript for a given pace call to the final time figure.


What Class seems to be saying is that the “slow pace phenomenon” in Moss only exists in higher class races, due to the difference in race shapes between higher and lower class allegedly due to a disproportionate increase in stamina rather than speed as horses increase in class. The example given is that we often see 10K claimers able to speed the 1st quarter in 21.3 but finish in 1:13. However, this says as much about the rider’s poor rating tactics as it does as it does about differences in race shapes. Any jock squandering a 1:13 horse’s limited energy resources with a 21.3 1st quarter is doing a poor job of rating indeed. It’s very hard to draw accurate conclusions about 2 classes’ of horses’ race shapes when the 2 types are ridden differently. It’s a confounding variable.

In any case, even if race shapes change with class, the Moss pace figures remain low relative to the final figures with only the degree varying. The best proof of this is how the pace figure (based on final times) is lower than the superscript figure (based on fractional pars) added or subtracted, depending on the sign, to the final time figure.


Bob

[keilan]

To give an example of the contrast with CJ's figures, with his figures the average pace for a race earning a 96 final figure is about 96. With the Moss figures, the pace figure for the same race seems several points lower.

Bob


Inaccurate

[keilan]

The Moss figures are correct to the extent that the track variants on any given day are correct.

The reason that the pace figures tend to be lower than the final time figure at the top of the scale is because the difference between the very best and very worst horses in the world is much less at 2F and 4F than at the end of the race.

In other words....

We sometimes see 10K claimers than get run 2F in 21.3 because they are extremely fast, but we never see 10K claimers that can run 6F in 109 because they can't carry their speed.

At the same time , we sometimes see G1 horses run 2F in 22.4 because there's a lack of pace, but we never see them run 6F in 113.

The result is a compression of the difference between the horses at the pace calls that expands at the ends of races.

The result is that the PARS (averages) for each call expressed in Beyer numbers winds up looking something like this at the extremes. (not exact)

Grade 1 107 111 115

10K Claimer 88 84 80

Because of this reality, Moss provides the little + and - indications as to whether a particular pace was fast or slow relative to the final time.

A Grade 1 race that goes in 115 115 115 actually had a faster pace than average.

A 10K claimer that goes in 80 80 80 actually had a slower pace than average.

The reason I know he is right about this is that I did a very detailed study on this 20 or more years ago when I used to make my own Beyer Scale pace figures for every call in NY. In fact, I discussed some of this with Randy Moss when the project was being implemented.

It can be kind of confusing to work with this at the beginning, but if you look at the + and - notations (or verbal comments) it will start making sense. There are also some very good explanations at the DRF web site.


good post Class

[proximity]

.
I am still at a loss as to why Moss adopted this method over more effective methods showing energy distribution. The only possible advantage I can see is that they would be more useful than traditional pace figures in predicting a horse’s final time in a sprint from his pace figure when cutting back from a route.


another advantage would be to more accurately compare fractional times in sprint races..... and there are more sprints than routes at most tracks.

compare two horses who set "par" paces: horse A at 5f and horse B at 7f.

using more traditional figures they both may have ran say 80-80. in moss, however A may have gone say 82-80 and B would rate 78-80. so A would have the better pace time and B would have the better closing time. the distance of today's race, the profile of the track for that distance, and the matchup of running styles would then be our guide.

most of john del riccio's pace figures were lower than his final figures as well. i think john's figures "crossed" in the lower 60s, which was near the bottom of his scale. and it should also be noted that he seemed to be a very good "turnback" handicapper..... something that had always been one of my weak points.

[bobphilo]

Inaccurate

How so? Please give reason.

Bob

[PaceAdvantage]

Yes, that would be helpful, wouldn't it Bob?

[bobphilo]

most of john del riccio's pace figures were lower than his final figures as well. i think john's figures "crossed" in the lower 60s, which was near the bottom of his scale. and it should also be noted that he seemed to be a very good "turnback" handicapper..... something that had always been one of my weak points.

That's because John's figures were for the individual horse and not the race pace figure. Johns figures only reflected the way an individual horse distributed his energy and not the race shape. Unless the horse is leading, his pace figure would be lower than the race pace figure.

Bob

[proximity]

That's because John's figures were for the individual horse and not the race pace figure. Johns figures only reflected the way an individual horse distributed his energy and not the race shape. Unless the horse is leading, his pace figure would be lower than the race pace figure.

Bob

no i am talking about the actual race figures for the class of the race, the pace of the leader and the final figure of the winner which john would send out bi-weekly to subscribers. i stored mine in an access table, had a query that calculated the individual horse's figures and would then print the figures out in the "start notes" on tsn's pps.

i actually believe cj's typical pacefigure would be higher than the final figure since he is using the pace of the winner rather than the leader in making pars and variants..... and not every race goes wire to wire. i'm sure it varies by class and surface of race though. cj can correct me if i'm wrong.

proximity

[proximity]

also, john's ratings really had little to do with raceshapes and were more meant to be viewed independently. apples to apples and oranges to oranges.

for example there was a thread last year where i was naively critical of his lexington ratings which were 82-81. this implied that the raceshape was "fast 1" (when the pace was blazing). however, if you queried the database you would see that an 82 was actually blazing fast for an 81 final figure on his system of numbers. his ratings reflected the phenomenon that classhandicapper was talking about.

there was another big thread on here about the met mile. another high class race with a fast pace which he had 82-82. again, querying my database told me that the 82 pace was actually very fast for an 82 final on his system of numbers......

[classhandicapper]

Bob,

If you focus some attention on the lower class races, you'll see that the Moss Pace Figures tend to be higher than the final time figures for those classes. You probably think it's the other way around most of the time because you tend to focus most of your attention on the higher quality races.

The PARs upon which these figures are based are the actual fractions and final times for every race at every distance covering a long period of time.

I don't think there's any good way to get around the phenomenon we have been discussing.

A 1/5 of a second has a value at each distance. If there just happens to be a smaller difference on average between great horses and cheap horses at the fractional calls than at the final time calls (and there is), so be it.

The ratings will be fine, but you have to get used the fact that the average pace rating for every class/final time will not look neat in comparison to the final time rating because THEY AREN'T NEAT, not because the figures are wrong. That's exactly why he gives out the +/- numbers.

[classhandicapper]

no i am talking about the actual race figures for the class of the race, the pace of the leader and the final figure of the winner which john would send out bi-weekly to subscribers. i stored mine in an access table, had a query that calculated the individual horse's figures and would then print the figures out in the "start notes" on tsn's pps.

i actually believe cj's typical pacefigure would be higher than the final figure since he is using the pace of the winner rather than the leader in making pars and variants..... and not every race goes wire to wire. i'm sure it varies by class and surface of race though. cj can correct me if i'm wrong.

proximity

You are correct about CJ's figures.

[bobphilo]

no i am talking about the actual race figures for the class of the race, the pace of the leader and the final figure of the winner which john would send out bi-weekly to subscribers.

I see. I wasn't aware that the bi-weekly reports contained pace ratings for the race - leader's times. Thanks for the correction.

Bob

[bobphilo]

Yes, that would be helpful, wouldn't it Bob?

PA, you can agree or disagree with my postions all you want but your not-too subtle crack implying I'm not trying to support my position with evidence and reason is unfair and way off base.

[keilan]

How so? Please give reason.

Bob


Bob I believe you implied that with cj's figures the race earning a 96 final figure is about a 96 pace figure.

In your opinion would that also be true for "the race earning a 86 final figure"? I'm not trying at all to trap you, I simply would like to understand your level of interpretation of cj's numbers.

[wolsons]

Bob, I think PA was referring to Keilan's post, not yours...

[bobphilo]

Class,

I just wanted to clarify that neither I nor Proximity ever maintained that the Moss ratings are in any way inaccurate. I have tremendous respect for Moss and his work and greeted his figures with great anticipation.
I've just noted that that when comparing them with several other pace ratings they seem consistently low in relation to final speed ratings for the same races. I agree that I am most familiar with his ratings for higher class ratings and trust you when you say the trend is reversed for lower class ones. Whether this is due to different riding tactics or inherent differences in horses is another matter.

Thay also appear to often differ from the pace variations indicated by the superscript numbers (which are based on pace pars) within the same Moss ratings. I suspected that this was due to to there being based on pars for final times, which are faster than pace pars for the same fractional distance. This seems to be Proximity's explanation as well.

Actually, the best way to decide this would be just to e-mail Moss himself and just ask him rather than trying to derive an answer. I plan to do that next.
In the meantime, I find that by just adding or subtracting the superscripts to the final speed figure, it works fine.

Anyway, thanks to you and Poximity for the the polite and intelligent dicussion. I hope we were able to throw some light on Tom's original question. Wish I could say the same for all involved.

Bob

[cj]

I'm curious how the Moss pace figures handled Haynesfield. The horse was originally give a 101 when winning the Damon Runyon on December 7th. It has since been lowered to a 93. I personally think the 101 was pretty accurate.

This is how I have his three route races, pace and speed figures:



Aqui 02/07/09 8.5 fst 103 89
Aqui 01/03/09 8.3 fst 108 88
Aqui 12/07/08 8.3 fst 99 102


Now, using the old Quirin model, you can determine pretty well how close up horses ran by averaging the pace figure and speed figure. This gives the horse a 101, followed by a 98, followed by a 96.

The thing with Moss is that they don't make speed figures along with the pace figures. They just use the Beyer figure and change it to a different scale. When Beyer then breaks races out, the pace figure is going to be way off in my opinion. It doesn't make sense to make pace figures independent of final time and just rely on another variant that doesn't consider the pace.

[Speed Figure]

I've never liked moss pace figures. I don't like the scale, they seem far to low, the way they look in the drf is dumb. :ThmbDown:

[proximity]

I don't think there's any good way to get around the phenomenon we have been discussing.

A 1/5 of a second has a value at each distance. If there just happens to be a smaller difference on average between great horses and cheap horses at the fractional calls than at the final time calls (and there is), so be it.
.

i kind of have a solution to this. it isn't necessarily perfect and is kind of long and boring..... but here it is if anyone is interested:

for starters we will just consider the 1/2 mile fraction of any race.

now for races at very short distances, say 4 1/2 furlongs, there will be very little rating going on at the 4f point of the race....afterall there is only 1/16 mile left to the finish. the gap in 1/2 mile par times between the highest and lowest class of horses at the track will be say 10 fifths of a second. as the distances increase, however, the jockeys will be rating the horses more and the gap between the top class horses and bottom class horses (in terms of 1/2 mile par) will narrow. say to 6 fifths of a second at 1 1/4 (numbers not necessarily accurate, just for example)

so for any pace numbers based purely on "speed".... there is an immediate problem: if we say that 1/5 second is the same value (no matter the final distance of the race)then where we start our speed chart has an influence on the "class par" and subsequently whether or not a given pace time for one (final time) distance has a higher or lower figure than another given pace time for another (final time) distance.

the following chart might better illustrate what i'm talking about:

class......4 1/2f...... 1m 1/4
stakes.....110..........106
mid clm....105..........103
md clm....100...........100

arbitrarily start the chart at the md clm level and the stakes par is 4 points higher than at 4 1/2 f than it is at 1 1/4.

class......4 1/2f......1m 1/4
stakes....110...........110
mid clm...105...........107
md clm....100...........104

arbitrarily start the chart at the stakes level though and now the maiden par is higher at 1 1/4. the par for mid level claimers has moved from being 2 points slower at 1m 1/4 to 2 points faster at 1m 1/4.

so where we start the chart has an impact on what these figures mean.... unless we were to increase the value of 1/5 second for the 1/2 mile calls as the distances lengthen. this is why i'm not 100% that you can use pure "speed" for pace figures and that there must (or should) be an energy component to all pace figures.

this would equalize the par charts across the distances and would satisfy energy people, but wouldn't reflect the differences in "speed."

i would "solve" this problem (maybe about as good as it actually could be solved) by studying the figures that pacesetters at longer distances earn when the turnback to shorter distances.

so in the end we might end up with a chart like this:

class......4 1/2f......1m 1/4(speed)......1m 1/4 (energy)
stakes.....110..........103....................... .110
mid clm....105..........98.........................105
md clm.....100.........93..........................10 0

not saying this stuff is a perfect solution to all pace figure problems, just some thoughts.

proximity

[proximity]

They may be rating, but they are certainly going faster than they are at the end of most races.

also, while in post #30 i have proposed an increase in the value of 1/5 second for a set pace distance as the total distance of a race lengthens, there would not necessarily be a proportional increase in the value of a beaten length.

this would depend on the point in the race at which the riders would typically stop rating and begin to shift gears for an all out drive to the finish. at this point in time the horses certainly could be going faster than they would be at the finish of a race (at the same distance as this pace call) where they could very well be finishing on fumes. if the pace call we are talking about is at or past this point then the value of a beaten length would proportionally be less than it otherwise would since each "length" would be covered faster: the horses ran slower to the point in question, but are actually running faster at the point in question.

[PaceAdvantage]

Bob, I think PA was referring to Keilan's post, not yours...Bingo! Apologies to Bob that I was somewhat unclear in my post.

[classhandicapper]

I'm curious how the Moss pace figures handled Haynesfield. The horse was originally give a 101 when winning the Damon Runyon on December 7th. It has since been lowered to a 93. I personally think the 101 was pretty accurate.


This is how I have his three route races, pace and speed figures:









Aqui 02/07/09 8.5 fst 103 89
Aqui 01/03/09 8.3 fst 108 88
Aqui 12/07/08 8.3 fst 99 102


Now, using the old Quirin model, you can determine pretty well how close up horses ran by averaging the pace figure and speed figure. This gives the horse a 101, followed by a 98, followed by a 96.

The thing with Moss is that they don't make speed figures along with the pace figures. They just use the Beyer figure and change it to a different scale. When Beyer then breaks races out, the pace figure is going to be way off in my opinion. It doesn't make sense to make pace figures independent of final time and just rely on another variant that doesn't consider the pace.






CJ,

I don't know anything about the specifics of this horse, but Moss WILL break out the pace figures from the Beyer variant if he thinks it's the right thing to do.

If there is wind (or other factor) that impacted the fractions differently than the final time, a seperate pace variant will sometimes be made. If Beyer ups/lowers a single final time figure relative to the rest of the day, Moss can probably use his disgression to decide whether it was a pace issue or the track simply slowed or got faster uniformly. My guess is that he usually goes along with the Beyer variant, but I never asked him (and should).

I think Beyer has the bigger issue and that probably makes life tougher for Moss. It's an issue that impacts almost all figure makers. It's one we've discussed here and elsewhere (I have discussed this with Beyer and Moss via e-mail also).

If a fast/slow pace impacts a final time, does the figure maker give the customer the real final time figure or a figure that reflects the ability of the horses better?

Most seem to opt for giving out a figure that reflects ability better. That probably helps the least informed customers, but has the potential to screw up some of the more sophisticated customers because they risk double counting the impact of pace in their handicapping.

I think one of things that really complicates this is that some figure makers aren't as sensitive as you to the impact of pace on time other than at the extremes. They also don't really understand the relationships between pace and final time well enough to do a good job even if they tried. So they more or less build pace into their final time figures (in many cases without even understanding what they are doing). One exception would be Ragozin because they rarely break races out from the rest of the day except at the super extremes. Then I believe they use a home grown formula. I actually like the way the Ragozin's handle it because I don't want pace built into a final time figure.

[classhandicapper]

proximity,

It's going to take a while for me to wrap my head around your solution, but I understand the issue you are trying to correct, including where figure makers arbitrarily start their chart. I think most people start it in the middle of the extremes at the circuit they play to limit the issue, but that is not really a solution.

I want to know two things.

1. The actual speed the horses ran adjusted for run up and turns so I can tell who is fastest early. I don't want it adjusted for jockey rating because I think that's a very individual thing I might be able to estimate by watching the races and looking at the horse's overall record.

2. Whether the pace was fast/slow for that class and relative to the final time at that distance at that track. This is a kind of quality and trip assessment relative to the norms at that track.

To me, all the formulas that try to combine pace and final time have some issues, but I do look at them and use them.

One example...

Paces on synthetic tracks tend to be slower than on dirt.

So if I am comparing the average 10K dirt speed horse to the average 10K synthetic speed horse I want my figures to let me know that the dirt horse is usually faster.

The monkey wrench is that even though the synthetic horse might be running slower, that slower pace might be more taxing on synthetic than on dirt. So the formulas for combining the two are different.

A dirt horse that ran 45 110 might have had an easier time of it than a synthetic horse than ran 45.1 110, but the dirt horse ran faster early. I want to know both.

To get both I use multiple sources and my own knowledge of the class averages.

Also, IMO this is part of the "speed/closer bias issue". I think the formula for combining pace and final time into some kind of performance rating changes from day to day as the track changes and it's virtually impossible to measure that exactly.

[cj]

CJ,

I don't know anything about the specifics of this horse, but Moss WILL break out the pace figures from the Beyer variant if he thinks it's the right thing to do.


I think you misunderstood me, or more likely I wasn't clear enough. I have no doubt that he will break out the pace figure and creates a good variant. However, he does not do anything to the final figure. He uses the Beyer, just on a different scale. So, when Beyer "breaks out" a speed figure, even though the pace figure is probably decent, the relationship between the two will be distorted.

Also, I don't see how you can make any sort of pace variant without considering the overall variant and speed figures. They are not, and never will be, separate entities.

[classhandicapper]

CJ,

I just started a dialogue with Randy on this issue. I'll get back to you.

[classhandicapper]

CJ,

You are correct. (should have known ;) )

In those situations Randy opts to publish what he believes are the correct pace figures because that's his product, but he has no control over the final time figure and can't adjust it. So if Beyer changes the variant for one race and Randy thinks it was a pace issue, the relationship between the pace and final time figures will be off. He says it doesn't happen too often. I assume that's because he usually thinks it's track speed issue when Beyer does that.

[proximity]

One example...

Paces on synthetic tracks tend to be slower than on dirt.

So if I am comparing the average 10K dirt speed horse to the average 10K synthetic speed horse I want my figures to let me know that the dirt horse is usually faster.

The monkey wrench is that even though the synthetic horse might be running slower, that slower pace might be more taxing on synthetic than on dirt. So the formulas for combining the two are different.

A dirt horse that ran 45 110 might have had an easier time of it than a synthetic horse than ran 45.1 110, but the dirt horse ran faster early. I want to know both.
.

classhandicapper,

my "solution" would actually handle all this stuff. slower paces on synthetic surfaces, turf,...... whatever surface could be handled as if they were longer races. the situation would be governed by the relationship between the pace pars and final time pars for the specific track, distance, and surface we are talking about.

"step 2" ( the study of the figures that "longer" situation pacesetters run when they turnback to "shorter" situations)of my solution would show you that the dirt horse is usually faster.

"step 1": the "energy" par chart would tell you that the "slower" synthetic horse had potentially earned his final figure the harder way.

or alternatively that (first) your synthetic horse that went 45 110 actually had a "harder time" of it, even though the dirt horse actually ran faster. synthetic could actually have a higher "energy" pace figure, but after the "speed" adjustment is applied, it could show that dirt has the higher "speed" pace figure.

proximity

[bobphilo]

Bingo! Apologies to Bob that I was somewhat unclear in my post.

PA, OK, no problem. Thanks for the clarification.

Bob

[bobphilo]

Bob I believe you implied that with cj's figures the race earning a 96 final figure is about a 96 pace figure.

In your opinion would that also be true for "the race earning a 86 final figure"? I'm not trying at all to trap you, I simply would like to understand your level of interpretation of cj's numbers.

Keilan, sorry about the delay. The answer is a bit involved and I'm dealing with some medical issues right now and will answer your question as soon as I can. Don't want you to think I was ignoring your question.

Bob

[classhandicapper]

proximity,

It's clear we are on the same page. I'll review your ideas some more. I didn't have a chance yet (busy). It has been a pleasure.