Probability

I am complaining that you won\'t allow your customers (of which I am one) to discuss ALL the potential issues behind large differences in figures for races like the JCGC. You just deleted one that was short and direct and that didn\'t mention any forbidden topics. Imagine if I got specific.  

[TGJB]

CH-- I allow my customers to discuss things WITHIN REASON. You fill this board with hot air, constantly and repetitively, and I\'m not going to allow it any more. The people who post and read here are not doing it to listen to you navel gaze and say the same things ad nauseum. If you don\'t get that-- and it is clear that you either don\'t or do not recognize any one else\'s interests as being of any consequence-- tough. I suggest you set up your own website, where you can have further conversations with yourself.

[bobphilo]

Michael,

So in denying equivalence you are saying that there is no justification in saying that Silver Train, in winning the BC Sprint, turned in a better performance than A Bit \'O Gold did in the Classic, and if somebody gave BOG a better figure than ST it would make sense?
Studies of time and class pars of thousands of races all show that the realtionship of winning times (and speed figures) to class levels hold up across all distances. That means that if the winning figure for the average Grade 1 stake at 10 furlongs is, let\'s say, a 0, the average winning figure for Grade 1 stakes at 6 furlongs will also be 0. Same story for every other commonly run distance. Not speculation, the reasearch bears it out. Figures would lose much of their great utlity if they didn\'t reflect equivalent performances at different distances. Figures would then be no more useful than final times and Jerry, and all the other figure makers, would just be wasting their time.
Hey, Jerry, you haven\'t just been wasting your time, have you? I don\'t think so.  
Anyway, be glad to continue discussion, Michael.

Bob

[Michael D.]

Bob,
re the \"studies\" and \"research\" - sounds interesting. where can i find that info. thanks.

[Michael D.]

and Bob, we have discussed the evolution of the breed many times on this board. Jerry HAS NOT said that the breed has developed equally at different distances. He has said he is not sure. Therefore, there is no way he would (or possibly could) blindly assign equal figures to top horses at different distances without taking into account what is actually happening on the track.

[bobphilo]

Michael,

Anyone who makes figures has to first compile a data bank to research and establish the equivalent figures for all distances at every class level.
Beyer in most of his books lists his results as does Quinn in \"Figure Handicapping\". Dr. Quirin, a mathematics professor, conducted a multi-year computer study at his university including over 100 north American tracks and reports them in his books, \"Winning at the Races - Computer discoveries in Thoroughbred Handicapping\", and his later work, \"Thoroughbred Handicapping - State of the Art\". Gordon Pine does an annual compelation of par times for all distances and class levels for all tracks as well, though I\'m not sure if he translates times to figures (it\'s available from the DRF). Charles Carol in \"Handicapping Speed\" gives a very detailed analysis showing the equivalnce of times across class levels and increasing distances.

I\'m sure Jerry himself has such a data base and he can help you with questions about that. Furthermore, from my own experience, when once upon a time I had much more free time, I used to make speed figures and first researched the relationship of times and figures to class. I actaully developed a mathmatical formula  to predict equivalent times at different distances and it matched the actual par times and speed figures from results taken from NY tracks across all class levels. This showed that if one correctly compiles speed ratings, they should show the equivalnce of performance for the same class level at all distances.
So yes, the research shows equivalence of performance for a class level across all commonly run distances.

Bob

Michael,

The studies I have done on average winning figure by class/distance suggest that sprint and route figures are equal for each class. Thank goodness for that.  

I am less certain about the very highest level - Grade 1 sprinters and routers.

There are way fewer Grade 1 sprints than routes each year. So there are sample size issues.

Also, IMO some horses are better at one or the other but those that are versatile enough to do either (Congaree would be an example) tend to become routers because the purses are bigger, the breeding opportunities are better etc.... So I think the very best routers tend to be a little better than the very best sprinters on average.

I know you have some theories about changes in the breed etc... Perhaps the gap has narrowed or flip flopped.

[Michael D.]

Bob,
Thanks for the info. Have the times of 10f races improved as fast as the times of 6f races (over the years)? Thanks.
.....

I just bring up the point because here are the fastest (in terms of times) 10f races of all time:



1:57 4/5
spectacular bid SA 1980

1:58 1/5
noor GGF 1950
quack Hol 1972
in excess Bel 1991

1:58 2/5
affirmed Hol 1979
greinton Hol 1985

1:58 3/5
swaps Hol 1956
round table Hol 1957
j.o. tobin Hol 1977
affirmed SA 1979
turkoman Hia 1986

1:58 4/5
native diver Hol 1967
figonero Hol 1969
gladwin Haw 1970
group plan Haw 1974
pay tribute Hol 1976
tiller SA 1979
go west young man Hol 1980
alysheba Med 1988
martial law SA 1989

 look at the dates of those races. if you track down the fastest 6f races of all time, you will find that the races are much more recent. is the breed evolving equally at different distances?
 

[bobphilo]

Michael,

Jerry has also NOT said the breed has not developed equally at all distances. If there is research evidence for this supposed disparity, I\'d like to see it.
True evolution can take thousands or millions of years - we are still genetically identical to our Cro Magnon ancestors. You are really refering to the subtle genetic changes brought about by selective breeding over a few generations. Actually, Jerry has not said that the improvement of racehorse is a result of breeding. My take is that modern drugs are a more likely explantion, but I\'ll let Jerry speak for himself on that. I see no evidence that sprint records are falling faster than routes.  
Even if one grants that selective breeding for speed is making sprinters faster than routers, the effect of a a few years breeding could in no way create such a disparity that would make a badly beaten horse finishing 8th in a sprint race better than the same level route race daylight winner. Especially since Imperialism isn\'t really that good at 6 furlongs, so his figure should really be even lower than Boreggo\'s, not better as the sheets claim.
It is hardly fair to say that Jerry is \"blindly\" assigning figures when all the research (see my last post) backs up his methodolgy and the equivalance of correctly made speed figures across distances. Especially since his figures come from projections based on the reality of \"everyday racing\".
By the way, in case I\'m sounding like an apologist for Jerry, I do have some other disagreements with his methodologies and am not 100% convinced horses are getter significantly faster (though it\'s possible), let alone faster at some distances than others. That would make the argument against equivalence even weaker.

Bob

[Michael D.]

Bob wrote: \"Jerry has also NOT said the breed has not developed equally at all distances. If there is research evidence for this supposed disparity, I\'d like to see it.\"

Bob,
You would like to see it? Bob,if you want to blindly give equal figures to top horses at different distances, you need to have research on this subject ALREADY, and it needs to support your theory. Everything you have typed in here is based on an assumption that has no merit. Now you opine that the DEGREE of evolution is too small to account for significant differences, again, with no evidence. Think about it Bob. If you breed a small, speedy Carson City filly to Montbrook, you could get a very fast 6F horse, but are very unlikely to get a very fast 10F horse. If you look at the potential times this horse could put up at those respective distances, we are not talking about miniscule differences. If all of the sudden, the Carson City/Montbrook cross starts to resemble the breed as a whole, you definitely don\'t want to blindly give a \"0\" to the best sprinter and a \"0\" to the best 10f horse. Fortunately today\'s breed does not look like that, but if you look at the way Storm Cat and Mr Prospector blood dominates the game today, you might want to ask the question: \"in terms of equivalance at different distances, is this the same breed we had thirty years ago?\" I don\'t have the definitive answer to that question (just an opinion), so I would be very weary of attaching pars that don\'t move in relation to distance.

[bobphilo]

Michael,

If all the research I've cited in my post is not evidence of equivalence of figures across different distances, I don't know what is. I am not giving an idle opinion about selective breeding and evolution but speaking from my training in biology including evolutionary genetics and equine breeding. As I said, evolution is a term that biologists correctly apply to changes in a species that take huge periods of time. Selective breeding speeds up the process a bit but breeding could not account for the huge degree of change that would suddenly make a terrible performance in a sprint superior to a winning performance in a route of similar class. To rate it as such is a clear indication that figures one is assigning are out of touch with reality, which is what this thread is all about. Furthermore, even if your unwarranted assumption that's sprinters have become amazingly superior to router were true, Imperialism is by no means a superior horse at 6 F. To say that his performance at this distance would be superior to Borrego's at his best distance is absurd on the face of it.
 By the way, I am not saying that the errors in the Rags figures come from his failure to understand equivalence, because he does. That is something every successful figure maker knows. That's why figures are so useful. The problem with the contradiction between the Sprint and Gold Cup has nothing to do with the difference in distance. The same inconsistencies are evident even with races of the same distance. It's a flaw in methodology.
If Jerry would make his figures solely on pars then, yes, the concept of equivalence would be self-serving because it's built into the pars, and though there was equivalence initially, just using pars could perpetuate it. But by using projections this huge shift in equivalence you claim would be evident. Since Jerry is the one actually doing this, I would call on him to explain this even better I could.

Bob

Michael,  

\"I don\'t have the definitive answer to that question (just an opinion), so I would be very weary of attaching pars that don\'t move in relation to distance.\"

If you don\'t trust that the classing system is at least reasonably efficient and that the figure pars for sprint and routes being equal for all classes proves the point, there is only one other way I can think of to test it.  

You would have to study the small pool of very versatile horses that are winning at the same class running both long and short distances and look at their figures. Within that group, there would probably still be some biases, but it would be a better group to study than the more random groups of sprinters and routers within the same class.

If you find a speed figure pattern within that smaller group, you may be on to something. In other words, if most of the Grade 1 7F horses that have also won Grade 1 races at 10F have slower figures going 10F (or vice versa), there is something worth investigating further.

[TGJB]

Michael-- an old friend was in town last night, and I had too much to drink and too little sleep, so I don\'t want to get in a long conversation about this now-- maybe in the next couple of days, if necessary. But a couple of things.

1-- There used to be far more 1 1/4 races than now, which gave horses far more chance to run fast.

2-- you will note tht a lot of those fast times were in California, which used to have lightning fast tracks. That is no longer true-- you don\'t see the fast 6f times there now either.

3-- without getting into a long thing about this, speed chart \"pars\" are originally set up based not on a theoretical relationship between distances, but on averages based on the times horses at similar levels actually run at each distance. To put it differently, if things happen to change the breed to make them not run as well going long, the right thing to do would be to change the pars--  you are  supposed to end up with the winning figures for different distances coming out the same. Not for each individual class, and not necessarily with a specific small group of horses, but overall.

\"the right thing to do would be to change the pars-- you are supposed to end up with the winning figures for different distances coming out the same. Not for each individual class, and not necessarily with a specific small group of horses, but overall.\"

You are saying that the time chart would have to be changed if changed relationships could be demonstrated across the entire breed, correct?

That makes sense to me, but it might be more difficult to prove that than to just study specific groups of very versatile horses.

The overall quality of sprinters vs. routers could be different because of different financial incentives and other factors. That might confuse what the relationships should be for versatile horses.

 

[bobphilo]

Michael,

Yes, I wouldn\'t be too surprised to see that there have been more current. improvements in 6F times than 10 F times. Part of this is a function that there are increasingly more sprints offered than routes and there are several more opportunities to lower records at the shorter distance. I\'d like to see what the trend is with average times of all races run at these distances is.
You also cannot assume that sprinting and routing are like totally different sports like basketball and race riding, for instance. In B-ball, height is a crucial attribute. The same factor would be an almost impossible handicap to a jockey. There is also not the extreme difference one sees in track where races range from 100 meters to over 26 miles (the dash is less than .002% of the marathon). On the other hand, 6F is 60 % of 10F and almost 70% of 9F. There is a tremendous degree of overlap of common factors, such as stride efficiecy, cardiovascular function etc, so that such a huge improvement in sprinters alone without some correspounding route improvement, enough to account for the absurd inconsistencies of the figures for the Sprint and Gold Cup, is unbelievable.
You see, it\'s not so much as issue of whether or not there has been an increase.
the research I\'ve sited shows that, at least at the time it was done, equivalency held. The question is, not have sprint performances improved recently, but if so, have they done so to such incredible extent vs. routes to justify throwing the principle of equivalences and thinking that the figures for the BC Sprint Vs. the Gold Cup make sense? The answer is clearly no. The research, genetics and physiology all point to this.

Bob