• Hi Guest Just in case you were not aware I wanted to highlight that you can now get a free 7 day trial of Horseracebase here.
    We have a lot of members who are existing users of Horseracebase so help is always available if needed, as well as dedicated section of the fourm here.
    Best Wishes
    AR

Using Speed Figures

USFIGS

USFIGS

Yearling
I was involved in a project about 12 years ago with a U.S. Data provider called Thorograph. We came up with what was called 'thoropatterns', i.e. how likely was a horse to run their best figure on a given day. There are almost an infinite # of ways to break this down of course, by track, by trainer, by jockey, etc. The interesting part for me was defining what is a horse's "best"?

Is is their best figure ever? Maybe for a 2YO or 3YO but certainly not a 7YO that had once raced in high class races but is now in the lower level handicaps. I have all sorts of research and theory behind this as it applies to U.S. racing, and BTW it varies by surface. Generally for turf horses, we used the best figure in the last 6 months or last 6 starts, whichever covered the shorter period of time. For a horse returning from a layoff we'd use the best figure within the last 3 starts preceding the layoff. While I have a great deal of subject matter expertise for U.S. horses, I have very little when it comes to European racing. I am curious what good handicappers use when handicapping U.K. & France. There is a lot of theory as to why many of us handicap this way (comparing each horse's best) that I would be happy to explain if anyone was interested.

Thanks in advance.

Ian
about 50 miles North of Gulfstream Park
 
In a nutshell...

What we look for when handicapping a race is

  1. How fast is each horse on their "best" effort
  2. How likely is that to run their best today

Statement 2 is harder to quantify than statement 1. The kinds of things that affect #2 include

a) current form (is a horse improving or regressing in form)
b) ability of a trainer to get a horse's best effort (good trainers have horses that can run near their best regularly, weaker trainer's horses oftentime run their races in a more random fashion
c) ground condition (e.g. some horses can only act on good or better, others run their best with cut in the ground)
d) race composition (likely faster pace, likely slower pace)
e) ability of the jockey to maximize a horse's effort

In a handicapping model I use here, I actually express it in terms of probablility (subjective as that may be). Let's say a horse's best is a TF 120 but he has a bad trainer and will be racing over softer turf that he might like. I might express his probabilities for today as such

Better than 120 TF 5%
110-119 TF 20%
100-109 TF 60%
worse than 100 TF 15%


of course you could use speed figures rather than TF ratings. I do this for each horse and come up with a mean, median and 75% figure for each horse and then compare each horse to the other competitors. This gives me some kind of idea on win probability. This is overly simplified but I think you get the drift.

And yes, my probabilities ARE subjective. I think you'd find though that the more you do this, the more outcomes you see, the better you get at this. What you find is that some horses are just more consistent than others (as well as some trainers). A lot of times you can come up with a fast but more random horse (at a big price) that has the ability to win if they run up to their best but they just don't do that very often. One of the overriding concepts is you want to compare each horse's BEST races (or the practical best they can be expected to run today) against each other horse's BEST races. You DON'T want to compare one horse's good race with a another's bad one. My question for the board is, I have an idea how to find the practical best for a horse racing in the U.S. (I call it its "effective top"). I am not sure if it would be done the same way in Europe, but I think it might.

I hope this helps.

Ian
 
USFIGS USFIGS, how do go about recording s/figures for a race that was supposed to be run on the turf and then gets transferred to the dirt. Do you just ignore the race as effectively it is no longer the same type of race that was initially planed by the trainer. Or would you just treat the race as any other race.


Arkle
 
Generally I try to use the same surface. If I am handicapping turf races I use a turf effective top, dirt a dirt effective top. AW is a little tricky as some AW play closer to dirt and some closer to turf. If I am handicapping a race and the horse's last wasn't on today's surface I will generally skip it unless I have no other lines to rate, e.g. a horse makes its 4th career start, this time on turf with 3 AW races. Obviously I have to use the AW form to assess how fast I think he is and I might adjust the percentages in each group if I think he is likely to run better (or worse) on turf than he has on AW.

I know some will be skeptical that you can assign percentages but that's not correct. For instance, obviously a horse is never 100% to run their best (assessed before the race). Trust me, they're not 90% or even 80% either. You might be surprised how good you can get over time. I have been doing this for almost 20 years now. I believe I was one of the first ones that I know of that attempted to quantify it in a model. I spent a lot of time studying raw figures and talking to Len Friedman (Ragozin) and Jerry Brown (Thorograph) about this since those guys pretty much invented these kinds of figures. I am attempting (more like hoping) to do something similar when it comes to French racing. Part of the problem is good French speed figures (or TF style performance figures) are not easy to come by commercially.
:)
 
Last edited:
USFIGS USFIGS, i know rpjd99 rpjd99 is in to his French racing and also i think TheBluesBrother TheBluesBrother is as well. Personally i to like French racing but i have no speed figures. Have you tried France Gallop you may find some thing there. And also there may be a Australian web site that covers French racing, the name currently escapes my memory.

Arkle
 
I communicated with TheBluesBrother TheBluesBrother in the past. He is very knowledgable and very helpful. Unfortunately for me, I seem to recall he only covered the day's PMU tracks and wasn't interested (at the time) in supplying them commercially to me as it is a hobby not a business for him. I am familiar with the Australian site and have looked at their figures as well which are IMHO, very useful.

I am trying to cover as much ground as possible, so to speak.
 
Hi, USFIGS USFIGS an interesting thread. Out of interest how would you assess a horse that ran its best race last time out putting up its best speed fig. Surely that speed fig would have more significance against other horses in the race who ran their best figs 6 months or so ago. How would you weigh that up as a percentage. Do you use some sort of formula?

Thanks James
 
I would assume that the horse than ran his best is more likely to repeat or go forward (run better than that) than a horse that ran a best fig 6 months ago.

I can only speak for U.S. racing but turf horses that haven't run their best in 6 months still have a chance to run it today assuming that figure was earned fairly (e.g. not due to extreme ground, very favorable race setup, etc.). Depending upon how close he had come to that best fig during the past 6 months I might give him a 25-30% chance to run it today.
 
You must log in or register to reply here.
Back
Top