Turf vs Dirt question

[Cratos]

Originally Posted by cj

I guess I'm misreading, but I don't see how more sprints being carded on dirt has much to do with the question posed by AndyC.

AndyC’s question: “Then why don't we see more closers or horses with stamina win on the dirt?” and the answer of “running style” fits that question in my opinion.

In other words with dirt races being “sprint plentiful” in NA the closer type horse will typically not have enough racing ground to exercise its full talent.

Of course there is exception to this answer and it is class; a class closer would probably win consistently at the sprint distance against inferior opponents.

[raybo]

Originally Posted by cj

This is also what I'm saying!

I'll also add this might have implications when comparing different distances on the same surface, but I'll save that for the next thread and stick to turf/dirt for now.

I agree, I believe it absolutely applies to different distances on the same surface.

[raybo]

Originally Posted by Cratos

Of course there is exception to this answer and it is class; a class closer would probably win consistently at the sprint distance against inferior opponents.

Yup, "class", depending on how you quantify it, and how accurately, will often trump early speed in shorter races, even if the "class" horse has been running longer distances and coming off the pace in those.

[MJC922]

The realization that final time is just not a very reliable indicator of a turf effort in general should be considered. If you recall my statement a long time ago proportional time is the dirty little secret and you're seeing in part why I've stated this is so. As for 1:36 on turf somehow not being equal to 1:36 if beaten by 5, well I would have to disagree there, it is equal, the real issue is 1:36 standalone isn't a reliable indicator for what took place from one turf race to another. I know people who are knee-deep in final time don't want to go there because it invalidates a lot of what they're doing in marketing final time as a 'good' measure, trust me I didn't want to go there either.

As you said Beyer introduced different 6.5f points for beaten lengths in turf races to mitigate against some of this, but it would take different pars, different speed charts and on and on to get this to ever 'line up' properly and even then I suspect it's going to be little more than a cosmetic improvement. Quinn tried to take it in a different direction by amplifying the last fraction, it's crude I agree with you, it creates the cosmetic separation to line up with dirt, the trouble is, late kick is then always more highly rated which in reality is only a band-aid solution at best.

[cj]

Originally Posted by MJC922

The realization that final time is just not a very reliable indicator of a turf effort in general should be considered. If you recall my statement a long time ago proportional time is the dirty little secret and you're seeing in part why I've stated this is so. As for 1:36 on turf somehow not being equal to 1:36 if beaten by 5, well I would have to disagree there, it is equal, the real issue is 1:36 standalone isn't a reliable indicator for what took place from one turf race to another. I know people who are knee-deep in final time don't want to go there because it invalidates a lot of what they're doing in marketing final time as a 'good' measure, trust me I didn't want to go there either.

As you said Beyer introduced different 6.5f points for beaten lengths in turf races to mitigate against some of this, but it would take different pars, different speed charts and on and on to get this to ever 'line up' properly and even then I suspect it's going to be little more than a cosmetic improvement. Quinn tried to take it in a different direction by amplifying the last fraction, it's crude I agree with you, it creates the cosmetic separation to line up with dirt, the trouble is, late kick is then always more highly rated which in reality is only a band-aid solution at best.

I agree with a lot of this, but I also think it is pretty easy to identify when final time isn't reliable, just like it is on dirt. I think it would be more than a cosmetic solution. We'll see though, I could be wrong. There will always be races that aren't rated well with time final time alone, and that will never change.

[Valuist]

Didn't Beyer say in their figure making approach, he started using the beaten lengths chart for 6.5f for grass races between 1 mile and 1 1/8 miles? Believe it was in Beyer on Speed.

Never mind, I guess this was brought up earlier.

[MJC922]

Originally Posted by cj

I agree with a lot of this, but I also think it is pretty easy to identify when final time isn't reliable, just like it is on dirt. I think it would be more than a cosmetic solution. We'll see though, I could be wrong. There will always be races that aren't rated well with time final time alone, and that will never change.

As usual all the best to you man. Respectfully.

[steveb]

time is every bit as valuable on turf as it is on dirt.
it is just 'different'
i am in australia where nearly all races are on turf, but time is far and away the most significant factor in any of my models.

but winners(race) time is treated far differently than non-winners times even if the same.
race time numbers are figured by comparing to other race times, but also ran numbers are figured by comparing to the number of the horse that won the race.
two completely different scenarios that require different calculations

on turf it is generally a fact that nearly every horse in any race can run faster than it actually does.
it is the leader that decides the final time in most cases, not the also rans.

[cj]

Originally Posted by steveb

time is every bit as valuable on turf as it is on dirt.
it is just 'different'
i am in australia where nearly all races are on turf, but time is far and away the most significant factor in any of my models.

but winners(race) time is treated far differently than non-winners times even if the same.
race time numbers are figured by comparing to other race times, but also ran numbers are figured by comparing to the number of the horse that won the race.
two completely different scenarios that require different calculations

on turf it is generally a fact that nearly every horse in any race can run faster than it actually does.
it is the leader that decides the final time in most cases, not the also rans.

That is interesting and may very well prove to be true here as well. I'm going to try to find out. Either way I'll have better numbers in the end, that is the goal.

[raybo]

Originally Posted by steveb

but winners(race) time is treated far differently than non-winners times even if the same.
race time numbers are figured by comparing to other race times, but also ran numbers are figured by comparing to the number of the horse that won the race.
two completely different scenarios that require different calculations

I think that is why "time per beaten length" must be measured/calculated as accurately as possible. Otherwise, we have an inaccurate method of comparing horses and projecting their relative performances in the future

[steveb]

Originally Posted by raybo

I think that is why "time per beaten length" must be measured/calculated as accurately as possible. Otherwise, we have an inaccurate method of comparing horses and projecting their relative performances in the future

i don't use lengths for any calculations, only always the times of the individual horses, from which the margins are derived anyway.

i am guessing that in the states you are not being provided with those individual times, which is also ridiculously the case here in some jurisdictions.
i have the means to get all those times that the public does not, but feel they should

it makes keeping everything in proportion much easier, to use time instead of lengths.

[MJC922]

Originally Posted by steveb

time is every bit as valuable on turf as it is on dirt.
it is just 'different'
i am in australia where nearly all races are on turf, but time is far and away the most significant factor in any of my models.

but winners(race) time is treated far differently than non-winners times even if the same.
race time numbers are figured by comparing to other race times, but also ran numbers are figured by comparing to the number of the horse that won the race.
two completely different scenarios that require different calculations

on turf it is generally a fact that nearly every horse in any race can run faster than it actually does.
it is the leader that decides the final time in most cases, not the also rans.

IMO we can treat winners time differently than non-winners in all kinds of creative ways but it's only a mask to cover up the inadequate nature of final time measuring the effort. It's then no longer final time it becomes an artificial construct so what's the point in calling it time anymore, call it a projection. Quinn's method for example has no fundamental soundness but probably produces a better strike rate... accidentally. If all that matters is a better strike rate for the model on average then I can't argue with it, however in the case of when the pace is too fast and we're actually amplifying the last fraction to favor the late runners even more then it makes no sense. I have no problem with artificial constructs as long as we're honest about what they're NOT.

[steveb]

there is nothing artificial about it.
it's all non-subjective and proportional and it just 'happens'.

one thing that has me intrigued, i often wonder how far advanced they would be in their time stuff in the states, if they measured in metric rather than imperial.
some things are just staring you in the face with metrics, that one would not likely notice with imperial.

[DeltaLover]

If the subject of the thread is how to calculate the final time of a horse that was beaten by a number of lengths, I cannot see how the surface is playing a role. The related parameters are the final time of the winner of the race, the distance of the race, the number of beaten lengths and the average length of the horse.

The formula I am using for this purpose is very simple and straight forward:

HORSE_LENGTH = 3.


@make_exception_safe
def get_time_from_beaten_lengths(distance, final_time, lengths_beaten):
    """
    :param distance: Distance of the race in yards
    :param final_time: Final time in decimal format
    :param lengths_beaten: Number of beaten lengths in decimal format
    :return: the final time for the given beaten lengths
    """
    return final_time * (1 + lengths_beaten * HORSE_LENGTH / distance)

[MJC922]

Originally Posted by steveb

there is nothing artificial about it.
it's all non-subjective and proportional and it just 'happens'.

one thing that has me intrigued, i often wonder how far advanced they would be in their time stuff in the states, if they measured in metric rather than imperial.
some things are just staring you in the face with metrics, that one would not likely notice with imperial.

The part that 'just happens' in the states we call it class.