If I understood correctly you made a point about Ragozin numbers not being equivalent from one circuit to the next. I noticed that when comparing a number for a horse that raced at Belmont to a number for a horse that raced elsewhere the numbers and my result just didn\'t seem to fit. I have rationalized that this is because of the size of the Belmont oval and that 1 turn \"routes\" are just different from 2 turn routes. I asked this question of Len Friedman at a seminar and his answer was that a number is a number.
I would be very curious to know your opinion on this subject. And if possible a further explanation of how racing circuits effect the number making process and, how your approach to racing circuits makes your product a better product than Ragozin.
Thank You
Cozzene
just a quick thought..... i think figs from tracks like kee and pim SHOULD NOT fit in with figs from bel, different ball games to me (i think the shape of the track has a lot to do with it)..... i\'m also interested in hearing TGJB\'s thoughts on this.
I probably won\'t get to this until tomorrow, but I will try to respond at length. It\'s a pretty broad question.
IMHO, a figure tells you how fast a horse ran on a specific day, at a specific distance, at a specific track, but that does not mean the same horse would have run as fast under different conditions.
First of all, there are several reasons a horse may actually PERFORM worse when switching tracks. While I think it doesn\'t happen as often as many people think, sometimes they really do like or dislike a track (Turfway especially), or need one over the track (often those coming from Med to Inaq, for example), or don\'t run well after shipping, or need to acclimate (North to South, East to West). It\'s also true that some guys can get numbers at one place they can\'t get elsewhere (especially Delaware-- it\'s Dodge City down there. Check out the sheets on Pino and Iwinski horses, among many others).
In theory, a number is indeed a number. But that doesn\'t mean the horse will run that number (see above, and others will have other reasons-- pace, etc.-- why a horse may or may not run his race). But beyond that, the numbers can be wrong.
In general, Ragozin does not split the one and two turn races, and as far as I know is the only serious figure maker who does not. As a general rule, when they split, you have to take off from the routes (don\'t know why), sometimes a lot. Since the average American card is sprint heavy, and since your variant is made by trying to have the most horses run about what they run, and since Ragozin is only using one variant (not just for sprints/routes but throughout the day, almost always) this results in making a choice between giving the sprinters what they usually run and robbing the routers, or giving the routers what they run and giving sprinters faster numbers than they deserve. And sometimes they end up BOTH robbing the routers and pumping up the sprinters, if the split is severe enough.
Breeders Cup day was a good example of this stuff-- not only did they rob the early horses on the card when they made the figures for the day, but look at Pleasantly Perfect. Ragozin gave him 7 1/2 points (15 lengths) better than his Goowood figure, 6 1/2 points better than he had ever run before in his life, and he\'s not a young, lightly raced horse. This was because, as I pointed out last Spring (look at April-May in the archives) Friedman does not split the sprints/routes in SoCal, where PP had been running, hence giving the sprinters better than they deseve, and the routers worse. I pointed out a lot of examples back then , and PP is yet another-- he ran a new top alright, but not nearly to the degree Ragozin gave him. On BC day the card was mostly routes, so they did the figure to the routers, some of whom were coming from places where the sprints and routes don\'t split, or don\'t split that much. So they didn\'t rob PP this time (and the whole question was made a lot more complicated by the track speeding up throughout the day).
You mention the 1 turn routes at Belmont. I haven\'t looked at this (just as I hadn\'t noticed the SoCal sprint stuff until Friedman posted the King Robyn race last spring), but my guess is that routers will be given better Ragozin figures at Belmont than at other tracks because those races are done at the sprint variant.
All that aside, figure makers have tendencies, and varying abilities. One may be more conservative, or be more willing to give out good numbers on an off track, willing to slide the day, etc. If a horse who has been robbed of a few points runs against other horses who have been robbed, it\'s no big deal. But if he goes to another track and runs against horses who have been getting what they deserve, there\'s gonna be a problem. Not only will the results look crazy, but you end up giving out 6 1/2 point new tops to older horses.
TGJB wrote,
\"Breeders Cup day was a good example of this stuff-- not only did they rob the early horses on the card when they made the figures for the day, but look at Pleasantly Perfect. Ragozin gave him 7 1/2 points (15 lengths) better than his Goowood figure, 6 1/2 points better than he had ever run before in his life, and he\'s not a young, lightly raced horse.\"
You must be making this up, or is this some type of practical joke.
Think about it a 5YO horse who has been running and winning in Grade 1 company, at the best tracks in the country, suddenly JUMPS UP 6 POINTS ON SCALE.
And his trainer isn\'t a drug guy.
Isn\'t this like a 40 year old growing another 4 inches.
TGJB,
I\'d like to ask you a question related to some of your comments.
You laid out the case for route vs. sprint variants very well.
I would assume you have charts that show the relationhips between the sprint distances.
In other words, horses that run 6F in \"X time\" are approximately eqivalent to horses that run 7F in \"Y time\".
Do you find that there are days when the track is especially tiring and 6F in \"X\" = 7F in \"Y\" breaks down and becomes 6F in \"X\" = 7F in \"Y + Z\" or vice versa when the track less tiring than normal?
(same for routes 8F vs 9F etc..)
In this case I am not talking about how uniform the speed of the track is in various sections of the oval. I am also not talking about the impact of wind. I\'m more or less talking about how tiring the track might be on different days changing the relationships.
I think what may be of interest to US racetracks would be the service provided by turftrax.com. They use satellite technology and electromagnetic induction scanning techniques to provide highly accurate soil moisture profiles. Their magnascan project started off as an aid to the farming community but they now scan around 12 UK racecources and put up the going profiles. In truth, I don\'t really have a clue how they do it, but it may be worth a look. Cheers.
We have a standard speed chart, but we have corrections for some distances at certain tracks-- at Pimlico, for example, the 6f run-up is about 5 feet (literally), so the 6f times come up slow versus not only the routes but the 5 and 5 1/2 races. This is an issue to lesser degrees in general-- the longest sprint distance at a track is usually run from a chute limiting how far the gate can be backed up (also true for 1 1/16 inner turf races at Bel depending on the rails) and the slower the horses are moving when they break the beam, the slower the final time. There is also a question of differences in track maintenance-- NYRA superintendent Jerry Porcelli told me they can\'t (or don\'t) get the same equipment into the chute as works on the main track. Some tracks also change run-up distances frequently-- Laurel used to on random days run the 7f races with a MUCH shorter run-up than the 6f ones, which would definitely screw up anyone doing those races together (averaging the variants, creating bad figures for both).
I have not noticed the phenomenom you mention, but that doesn\'t mean it isn\'t happening. I make adjustments to individual races because I don\'t assume absolute accuracy of the information I am getting (the time can be a little off, the wind can gust or be blowing at such an angle as to bounce off the grandstand causing currents no one can see), so I rely very heavily on the past figure histories of the horses and make small corrections even if the variant is effectively the same. It is entirely possible that on slower tracks I am making more corrections in the direction you say and not noticing it because the difference would be small-- an eighth of a mile being only about 13 seconds, a fifth of a second would be a significant percentage of that, but only about a point.
I would also point out that wind might be playing a part if you are making figures that don\'t take wind into account. At most tracks, if a sprint is run into the wind on the backstretch the horses will travel farther into the wind than against it (6f at Bel is the exception) and the final time will be slower, making the track appear slower. Since any additional distance around one turn will also be run into the wind with no offset in the other direction you would have to adjust the figure even more to correct for it.
If what you are suggesting is correct, we might be able to run a study to find out. It would also presumably mean sprints and routes split more when the track is slower. We have a lot of stuff going on now (new products and some other stuff), but if you remind me in mid-December I\'ll see if we can take a look at this.
TGJB,
Thanks a lot.
I have no evidence at all that the phenomenon I am talking about exists. I\'ve seen situations where at Belmont (and elswhere) for example the 6F, 7F, and 8F relationships seemed out of sync with the norms, but I always attributed that to wind and/or the lack of track uniformity. Sometimes it reveals itself in the fractions.
From a theoretical point of view I have just always been curious about that because I believe that tracks can vary once in awhile in their propensity to favor speed or closers. It\'s pure speculation on my part there might also be some impact on the time relationships.
One thing I am fairly sure of from studying pace and watching races is that if the jockeys sense a speed bias they tend to be much more aggressive early. That leads to faster than usual fractions relative to final times - which in effect I suppose could offset any speed bias. It\'s hard for me to pin these things down scientifically and in certain terms, but I remain confident in my observations. With your data it might be possible to prove it.
I definitely agree that \"speed favoring\" biases are a self fulfilling phenomenom-- if the jockeys all send a 6f race becomes a 4f race, and very little happens in the stretch other than the horses galloping to the wire. I have given jockey instructions to do exactly the opposite in those situations because you can often get a great trip.
All else being equal, all races have a built in speed bias to a degree-- less traffic, less ground loss. The bigger the field, the more the number of turns, the more this is true.