Beyer Value of a Length on Turf

[jeebus1083]

In the book Beyer on Speed, Andy and his figure-making partner, Mark Hopkins did a study on the value of a length on turf compared to dirt, and due to the volatile nature of the turf (slower, closer paces, closer, faster finishes), the value of a length in turf routes was greater than those on dirt. They used 2.2 points per length as a starting point, which is higher than a conventional two-turn route, and it seemed to correct the figures of horses that ran dismal on dirt, yet seemed to run higher numbers on turf. The correction, much like the one made with synthetic tracks, give the high Beyers more weight upward, and the low Beyers more weight downward.

However, Andy failed to provide or explain the new values per length for turf races in the book, thus, the reader (myself) is left in the dark.

Does anyone here know the values of a length on the turf from shorter distances (turf sprints are filling most cards at most tracks these days), to the longer distances?

[rrbauer]

Look at the beaten-length value for 6.5 furlongs. Last that I heard, that was what they were using for turf routes. I can see no reason to use any different beaten-length values for turf sprints than dirt sprints unless there is compelling FPS difference between those surfaces at similar distances to cause a change.

[jeebus1083]

Look at the beaten-length value for 6.5 furlongs. Last that I heard, that was what they were using for turf routes. I can see no reason to use any different beaten-length values for turf sprints than dirt sprints unless there is compelling FPS difference between those surfaces at similar distances to cause a change.

The problem with just using 6 1/2F values is that there are different varieties of turf routes.... 1 mile, 1m70y, 1 1/16m, 1 1/8m... and sometimes even longer. I can't find it plausible that all the turf routes would apply the 1 length = 2.5 pts formula.

As far as turf sprints are concerned, I have some FPS data from Belmont this meet on my laptop at home that I can pull up and give you an answer. NY runs enough 6F and 7F turf sprints to provide a valid answer.

[cj]

The problem with just using 6 1/2F values is that there are different varieties of turf routes.... 1 mile, 1m70y, 1 1/16m, 1 1/8m... and sometimes even longer. I can't find it plausible that all the turf routes would apply the 1 length = 2.5 pts formula.

As far as turf sprints are concerned, I have some FPS data from Belmont this meet on my laptop at home that I can pull up and give you an answer. NY runs enough 6F and 7F turf sprints to provide a valid answer.

Here is the thing...Beyer solved the same problem, if you will, two different ways. On turf, he adjusted the beaten lengths chart, while on synthetics he adjusted his actual speed charts. Neither one is very effective. When a turf race is run with a fast pace, it doesn't make sense to penalize horses more because they lost. If a race is run with a very slow pace on polytrack, it doesn't really help to add 3 points (arbitrary on my part) to the figure of all horses as if they ran the race the same way.

[Bochall]

May be a bit off topic but here goes: for those of you who make speed figs on your own (the method: Beyer, Quirin, etc..) does not matter. Do you have a separate variant for routes and sprints at Belmont even though they are all run around one turn???? Help me-this si something I have struggled with in the past. Currently, I make 2 separate variants...any input would be big help.Thx.

[jeebus1083]

The problem with just using 6 1/2F values is that there are different varieties of turf routes.... 1 mile, 1m70y, 1 1/16m, 1 1/8m... and sometimes even longer. I can't find it plausible that all the turf routes would apply the 1 length = 2.5 pts formula.

As far as turf sprints are concerned, I have some FPS data from Belmont this meet on my laptop at home that I can pull up and give you an answer. NY runs enough 6F and 7F turf sprints to provide a valid answer.

In Belmont 6F turf sprints this meet (from data that I have so far), the average %E is 50.65%. On dirt, it's 51.94% However, the 1Fr on both turf and dirt are about equal. However, the 2nd and final fractions on both surfaces are distinctively different:

2nd FR
Dirt: 57.01 FPS
Turf: 56.82 FPS

Final FR
Dirt: 52.98 FPS
Turf: 55.67 FPS

One can make the argument that grass doesn't have the same tiring effect that deeper, dirt tracks have, yet the trend remains clear: the winning horses on turf still run fast final fractions in sprints as they do in routes.

[jeebus1083]

Here is the thing...Beyer solved the same problem, if you will, two different ways. On turf, he adjusted the beaten lengths chart, while on synthetics he adjusted his actual speed charts. Neither one is very effective. When a turf race is run with a fast pace, it doesn't make sense to penalize horses more because they lost. If a race is run with a very slow pace on polytrack, it doesn't really help to add 3 points (arbitrary on my part) to the figure of all horses as if they ran the race the same way.

Does it make sense to give a horse that runs -0's to positive single digits on the dirt, high 20s and low 30s on the grass, when they are just as hopelessly inept there as they are on dirt? I think that in a way, it's misleading to the bettor, effectively painting a picture that because this horse runs higher Beyers on turf, even though his ability on both surfaces is about equal, the horse's preferred surface is the grass. That's just horsefeathers, unless the horse only has a few starts under his belt, then you have to reserve judgment.

[cj]

Does it make sense to give a horse that runs -0's to positive single digits on the dirt, high 20s and low 30s on the grass, when they are just as hopelessly inept there as they are on dirt? I think that in a way, it's misleading to the bettor, effectively painting a picture that because this horse runs higher Beyers on turf, even though his ability on both surfaces is about equal, the horse's preferred surface is the grass. That's just horsefeathers, unless the horse only has a few starts under his belt, then you have to reserve judgment.

No, but he is creating new problems while attempting to fix others. While a large majority of dirt races are run with an honest pace, that isn't the case on turf or synthetics. There really isn't any way to "adjust" the speed figures alone without misrating a lot of horses. A speed figure alone is pretty worthless on any surface, but especially so on anything but dirt.

[proximity]

when thinking about this i find it helpful to go all the way back to the beginning.

first forget about beaten lengths. at the finish of the race they are just units of time converted to "lengths" anyhow. just think about the value of 1/5 second.

the only assumption i think you should make is that in the long term the par figure (not par TIMES) for a particular class of horses will converge towards the same value regardless of the distance or surface of the race. (basically, w/o getting into arguments about whether grade 1 turf horses at belmont are really the same class as grade 1 dirt horses at belmont...)

for races at a particular track, distance, and surface, the wider the average gap is between these similar classes of horses, THE LESS VALUABLE A FIFTH OF A SECOND IS REGARDLESS OF HOW LONG THE RACE IS OR WHAT SURFACE IT IS BEING CONTESTED OVER........EVEN IF THAT SURFACE IS DIRT.

what beyer does is "close enough" imo, but the truth is that each specific type of race (track, dist, surf, not just each distance and surface) has its own unique value of a length/5th of a second.

[jeebus1083]

when thinking about this i find it helpful to go all the way back to the beginning.

first forget about beaten lengths. at the finish of the race they are just units of time converted to "lengths" anyhow. just think about the value of 1/5 second.

the only assumption i think you should make is that in the long term the par figure (not par TIMES) for a particular class of horses will converge towards the same value regardless of the distance or surface of the race. (basically, w/o getting into arguments about whether grade 1 turf horses at belmont are really the same class as grade 1 dirt horses at belmont...)

for races at a particular track, distance, and surface, the wider the average gap is between these similar classes of horses, THE LESS VALUABLE A FIFTH OF A SECOND IS REGARDLESS OF HOW LONG THE RACE IS OR WHAT SURFACE IT IS BEING CONTESTED OVER........EVEN IF THAT SURFACE IS DIRT.

what beyer does is "close enough" imo, but the truth is that each specific type of race (track, dist, surf, not just each distance and surface) has its own unique value of a length/5th of a second.

What confused me about the Beyer Speed Figure math is the fact that his speed rating charts are based on the changing value of 1/5th at different distances, and the beaten lengths adjustments are based on the value of 1/6th of a second at different distances. I can convert a Beyer to a Quirin quite easily ((Beyer Fig - 83)/Value of 1/5th at Distance) + 100, but because Quirin bases his figures on the 1 length = 1/5th rule of thumb, if I try and figure the Beyers of the horses behind the winner, a horse 1 length behind the winner will have a Quirin .15 short of a full point.

Unless I adjust Quirin's scale so that 1 length = 1/6th, the relationship between Beyers and Quirins are not going to be alike for the losers.