Over the years it has become very apparent to me that the pace of the race can effect both the final time and the result.
It\'s clearest at the extremes when the early fractions are so slow that the horses are physically incapable of running the last 2-3 furlongs fast enough to produce their normal final time figures. On occasion, the pace can also be so extremely fast that all the major contenders are impacted and the race collapses.
However, I\'ve seen evidence from studying high caliber horses that smaller variation around \"average\" can produce other variances - both favorable and unfavorable. It just seems impossible to put it into a formula.
Now to my question.
Recently, the King\'s Bishop was run on the same day as Pretty Wild\'s allownance victory. (8/23). The final time of the KB was much slower, the pace was MUCH faster, and I would argue that the horses in that race was Waaaaaay better.
Many speed figure men concluded that the track changed speed later in the day (became slower). They bumped the figure up to a higher level than the allowance race. While that is certainly logical from an ability standpoint, I\'m not so sure it reflects reality. IMHO, the pace of the KB may have impacted the final time. The very fact that the fractions were so extremely fast indicates that the track, in fact, had not slowed down.
I see MANY examples of this and IMO when players try to combine trips with these sorts of adjusted figures they wind up double counting or undercounting the impact.
Any thoughts?
Classhandicapper----
Look no furthur than the results of last Saturday\'s Jerome @ Belmont Park.
DURING (who came out of the faster pace KB) beat up on PRETTY WILD by 5 lenghts.
Joe B.
1-- Not to be too harsh about it, but \"it is apparent\" that sometimes fast paced races have slow final times and sometimes slow paced races have slow final times,so the slow times must have been produced by the pace? Certainly extreme slow paces can affect final time for the reason you mentioned, and an EXTREME hot pace can as well (see the recent match race at Del Mar), which is one reason I sometimes have to disregard final time when making a figure. We mark those races with \"pace\" designations.
2-- Logic would indicate that a hot pace would only affect those running the fast fractions, not the ones behind them. Final time for most of the field should not be affected.
3-- Who were the \"many\" figure makers who concluded the track changed speed dramatically between those two races Travers day? Beyer did, I did not-- I used a 1 point different variant, which is a very minor change in keeping with what happens virtually every day at every track. In fact, 5 of the 6 sprints on that day fell into a very tight spread, which is unusual because of the wind (4, below). Pretty Wild ran out of his mind, and bounced the next time. Several solid horses in the KB (Eye of the Tiger etc.) ran to the numbers they had going in, and off the top of my head at least one (Frankel\'s horse) ran a new top. The ones near the hot pace did not run their best figures despite running 1-2, and that is something some handicappers may take into account, and some may not.
4-- There was a significant wind during the races, ranging from 10 to 20 mph, mostly at 10 o\'clock (imagine you are looking down at the face of a clock, with 12 across from the grandstand). The effect of wind is geometric, so the effect of a 20 mph wind is 4 times that of 10 (10x10=100, 20x20=400), and the wind was behind them in that long run down the backside at 7f, producing the potential for some fast early fractions, and into their faces through the stretch.
This wind stuff is something we haven\'t discussed enough here. Since the effect is geometric and since the speed and direction are estimated before and after the race (the observer is watching the running of the race), the data we and Ragozin use can be off a little, so at significant wind speeds the margin of error can get pretty high. Especially since there is a very large structure (grandstand) bouncing the wind currents around. And there is an even greater margin for error for those that don\'t use wind, either for final or (especially) pace figures, or for those (Ragozin)that take the data they get as ironclad fact, and create hard and fast rules limiting the amount of discretion that figure makers have in deciding variants. Ultimately, the only realistic way to make figures is by using the figure histories of the horses.
>1-- Not to be too harsh about it, but \"it is apparent\" that sometimes fast paced races have slow final times and sometimes slow paced races have slow final times,so the slow times must have been produced by the pace?<
Point taken. I guess I wasn\'t clear enough. Built into my comments was the assumption that horses\' speed figures were impacted relative to their norms. There are certainly races that stack up as fast-fast and slow-slow and every other combination possible.
But I talking about a situation where several horses typcially run a figure of \"X\" and a very fast pace causes them to run x-y.
(etc..)
>Look no furthur than the results of last Saturday\'s Jerome @ Belmont Park. DURING (who came out of the faster pace KB) beat up on PRETTY WILD by 5 lenghts.<
I had a huge bet on During for that very reason and that\'s why I brought the subject up here.
I know for a fact that Beyer split the variant and bumped up the the figure for the King\'s Bishop - otherwise During would have had a figure of around 92 and the winner would have gotten around a 97. I know another figure make that did the same thing.
Those earned figures were silly relative the ability of those horses. Horses that finished well beaten had even lower figures.
I am arguing that it was incorrect to split the variant in this case because the problem was PACE and not a change in the track speed.
The figure should have been reported as is (LOW), and astute handicappers should have known that everyone close to it (like During for example) was toasted by it.
On the flip side, perhaps Ghostzapper did not run quite as well as it looks because he was Well Off the pace. That\'s is not totally clear to me.
In some of these pace examples, you will often find players using the adjusted speed figure and then compensating AGAIN for the pace - in effect double counting.
>2-- Logic would indicate that a hot pace would only affect those running the fast fractions, not the ones behind them. Final time for most of the field should not be affected.<
That is what my research indicates also, but I have seen enough examples over the years of horses that were further back getting cooked in fast paces when they made middle moves into it etc...
I\'ve also seen examples where all the front runners - regardless of whether their own fractions were extemely fast or not get - bottomed out by chasing it and collapse.
Typically I see fast paces where the early fraction is very fast and the next one is more or less average. That leads to a fast combination because 1 fast and 1 average still equals fast.
There is not enough evidence for me to definitively say that horses that make middle moves into that second quarter are impacted.
However, there are sometimes extremes where there are back to back quarters that are both very fast. In those cases I\'ve seen the front runners completely collapse, but some of the mid pack closers that ran into it also got impacted.
By the way when I say impacted I mean the final times (speed figures) are not in line with what I would have expected given the horses\' ability and speed of the track that day.
>>This wind stuff is something we haven\'t discussed enough here. Since the effect is geometric and since the speed and direction are estimated before and after the race (the observer is watching the running of the race), the data we and Ragozin use can be off a little, so at significant wind speeds the margin of error can get pretty high.<<
I agree completely about the problem. That\'s one reason I don\'t go through the process of actually making numeric pace figures anymore. (there are other reasons). What I do is look at the horses style, the race development (jockey urging, gaps between horses, field size etc..) , glance at the fractions, and form a more general opinion as in fast, very fast, extremely fast etc...
Who is the other figure maker who split the two races? I would be pretty surprised if it was Ragozin-- he doesn\'t do it even when it\'s obvious it is necessary, like when there is a monsoon (the famous Storm Flag Flying day), or they do major work to the track (Chilukki etc.) between races.
>Pretty Wild ran out of his mind, and bounced the next time. <
This is one of those subective things we will have to agree to disagree about.
I think Pretty Wild would have finished up the track in the King\'s Bishop had he run there instead of in the allowance race. I think he pressed an average allowance level pace that day which allowed him to run an impressive final figure for a standout horse. Had he chased 21.3 and 43.3 he would not have run the same final figure. He\'s a darn good horse with a lot of opportunity to improve in the next few months, but he was overrated going into the Jerome. IMO he did not bounce (at least not in any meaningful way). He was chasing better horses.
I bet out substantially on During in the Jerome for that reason and others highlighted.
It is a sevice called Logic Dictates that makes \"Beyer like\" figures (same scale). Their figures for that day were almost identical to the Beyer figures. The guy that makes them told me he and Beyer had independently concluded that the variant for that day should be split staring with the Kings Bishop.
Not arguing that they are right and you are wrong (you in fact seem to have had a better grip on the day\'s figures). I am just using that day as an example of one of the problems I have encoutered over the years making figures myself and observing the work of others.
The pace sometimes impacts the time and figure makers sometimes adjust the variant for the race. That opens a can of worms for trip handicappers that might double count etc....
Briefly, because I\'m busy, the logic doesn\'t hold up with Pretty Wild-- his race is clearly in line with the rest of the day, if you did anything at all to the day it would be to upgrade the KB. So he gets the figure he gets (a big one), and a clearly worse one in the Jerome.
I expected we would not agree on Pretty Wild.
You view his performance in the Jerome as very inferior because he ran a much slower final figure. I view his performance as similar (perhaps a bit inferior) to his allowance race. I believe he ran a slower final time in the Jerome primarily because he was chasing the pace of graded stakes horses instead of allowance horses. (perhaps the mile distance also had something to do with it the worse performance)
Great discussion!
Ditto. I agree with the faster pace view, however it is possible that PRETTY WILD did bounce. I believe that he did not because I was never impressed with him to begin with and rate him as a no better than a good allowance horse.
I also subscribe to \"Logic\" and find that he does a terrific job with the pace figs. Speed figs, pace figs, trip notes, etc...they\'re all tools that over the long haul work better together than alone.
Joe B.
For what it is worth, I agree with 2 of the points that I think you were trying to make:
1) That many sheets players tend to read a poor final fig as an \"X\" when actually it may have been a big effort as part of a hot pace.
2) That you have to be careful about playing front runners to run back to big figs earned with soft fractions; particularly, if the fig is isolated and the horse is likely to encounter faster fractions in the race you are handicapping.
However, I am not sure either was the case with Pretty Wild. He looked like a good \"classic sheet bounce\" candidate independent of pace considerations. Who knows which came into play last weekend - maybe a combination of both. I guess the main takeaway is that a horse that looks likely to bounce on the sheets and also looks weak from a pace standpoint is a great play against.
Unfortunately, none of this helped me much in this case as Tafaseel\'s failure to get past During in the stretch drive cost me a pretty nice score in the pick-4 (although I know of at least one other regular on this board who was very pleased by During\'s victory - and my loss was his gain).
The question I have for you is why would someone so interested in pace use the tag \"classhandicapper?\"
Cheers.
Chris
Post Edited (09-18-03 11:02)
Maybe he\'s an old-timer like me who grew up
with Ainslie \"Class prevails\" mentality, yet
my favorite races to handicap are stake sprints
(where pace-is-the-race) and turf routes (class rules) IMHO.
I am not 100% sure that Pretty Wild\'s performance was totally the result of pace either, but I\'m pretty sure it contributed something to the fall off in his speed figures because those fractions were fast relative to the final time.
I thought Tafaseel was also a good value in there. Pretty Wild was my main \"bet against\". I liked During a little better than Tafaseel because I thought he was tad better and it was more hidden.
I definitely agree that if you have two negatives (pace and bounce candidate as in your example) that is a great \"bet against\".
I consider myself primarily a class handicapper that uses speed figures to help explain outcomes rather than measure horses performances in an exact way.
In other words if I rate Horses A and B as approximately equal based on recent finishes against other horses (looking at speed figures too) and A beats B by 3 lengths, the speed figure earned in that race will help me determine in A moved forward, B moved backwards or both. However, I believe there are so many inaccuracies in figure making that I wouldn\'t take the exact number as gospel. I would view the performance from multiple perspectives - each either conflicting or reinforcing the others.
>Maybe he\'s an old-timer like me who grew up
with Ainslie \"Class prevails\" mentality, yet
my favorite races to handicap are stake sprints (where pace-is-the-race) and turf routes (class rules) IMHO.<
Yep. I\'m more speed figure oriented than Tom was because the figures are so much better these days, but I am not a slave to them because I know that even the best figure makers (like Brown) run into difficulties, complexities etc...
Since we were just talking about this-- from Sat DRF, emphasis added by me:
\"Man, there was a headwind\", exercise rider Bobin Smullen said. \"I had to get a little lower to see if I could get out of it. AROUND THE TURN I FELT NOTHING; TURNING DOWN THE LANE I ALMOST GOT BLOWN OFF THE HORSE.\"
So either the wind picked up dramatically during the work (read race), or, more likely, the grandstand was blocking it at some parts of the track. Either way it goes to show that wind readings, even taken by live obvservers looking at flags, ripples on infield lakes, etc., before and after races, can be greatly innacurate (that was an extreme wind Robin was talking about, one that could easily have a 2-3 point effect on time). Wind readings from airports, which another service used proudly for decades and may still use, are taken periodically (hourly), several miles from the track, and are worthless.
Ultimately wind observations, like ground and track maintenance, are info bits you use. But you make your figures using the figure histories of the horses, and the more you do it the tighter your data base gets, making the figures you then make using it even more accurate with time.
I read that article also and immediately thought about this conversation. Good post.
When speaking of the effect of wind, I\'m thinking that the more accurate word to use was exponentially rather than geometrically. Only someone very special could nail the Derby Trifecta, the Belmont Exacta, kill Saratoga, and make Pythagorus roll over in his grave all in the same year.
The Derby score was last year, and the one who got killed at Saratoga this year was me. Is Pythagorus the same guy as Pythagoras? If not, he\'s probably the one who killed Sar.
Okay, exponential.
classhandicapper...
I like to consider myself somewhat of an expert on the Beyer\'s. I check all the Beyer\'s in each edition of Simulcast Weekly. I see the \"adjusted\" Beyers you speak of all the time. It isn\'t always a split variant either. It is sometimes just one race in the middle of a group of others, not at the beginning or end of the card. I agree the figure should be left as is, ala During should have gotten a 92. Although the figures might represent a more accurate picture of the horses who were up close to the pace, they greatly distort the closers numbers.
Example, horse A runs :43 1:10 while hanging on to win by a nostril. Horse B comes from the clouds to just miss. Applying the same variant that other races were assigned would give Horse A a figure of 90. The problem is, Horse A always runs in the low 100s. So the Beyer guys will arbitrarily bump the race to 100 or so. Probably a true reflection of Horse As ability on a Beyer scale. The problem is Horse B, who should have gotten a 90 Beyer as well, is now showing a 100 also.
For the record, they do the same thing in reverse as well. A filly named Mossflower was given a 101 for an allowance score at Belmont a few years ago when the variant given to all the other races would have given her a 114 or so. She was entered in a G1 Stakes race next out, and paid $11 due to that 101 Beyer in the PPs.
I don\'t mind these things, and actually track them and have been known to make a nice score from them.
I guess the point of all this is it really helps to understand the process of how the figures you use are made. Not just the overall process, but each figure on its own merits.
Thank you for that thoughtful post about pace, speed figures, separate track variants etc... I am happy someone understands and commented on the issues I am bringing up.
It\'s quite a problem unless you pay close attention because you want to make sure you don\'t double count certain aspects of trip that the figure maker might have included in his variant.
I have often seen situations where a figure maker made a separate variant for a race because they thought the track was slower for that race, when in fact the race was slower because of the pace and not the track speed.
They stumble into the right figure for the wrong reason for some horses and a totally wrong figure for other horses in the same race by adjusting it. (just as in your example)
Personally, I think ALL aspects of trip should be left out of the figure and variant process. That includes the impact of extreme paces and paths.
It doesn\'t matter what you think about bias or the impact of being wide etc.... I would prefer to simply have that information separately and quantify the impact separately based on my own study of those issues.
The obvious case is when a horse is 3-4 wide on both turns and another on the rail all the way and they finish close. On a neutral track that means one thing. On a dead rail that means something else. On a golden rail that means something else. Bias is super tough to determine, but it is simply obvious that some tracks do not play the same every day. It\'s easy to observe the extremes even if it\'s difficult to take it to the level suggested by some authors and players. So I defintely pay attention to the extremes. Otherwise the figures are completely worthless for those days. Without that knowledge you\'ll wind up making up all sorts of dumb reasons for the variance of some horses\' form.
Another less obvious example is WHEN a horse is wide and how hard he is being used at that time.
It is pretty clear to me that if you are a closer that is 3-4 wide first turn in the middle/back of the pack and the pace is moderate, that is having MUCH LESS of a negative impact than a front runner that is 3-4 wide first turn into a hot pace. The latter is dead meat and the former is not doing much that will take any starch out of his kick later.
I could go on, but at least we are on the same wavelength.