- This topic has 8 replies, 5 voices, and was last updated 10 years, 4 months ago by
.
-
Posts
-
I had not thought much about this topic until I saw the stat that the average horse in the average race runs 20 lb below best form. Now that will include tailed off runners, but even so it’s pretty dramatic and makes traditional handicapping based on marker horses etc very problematic.
It seems to me for betting purposes that ideally one needs the full statistical distribution of a horse’s likely performance, especially in handicaps. But apart from inadequate proxy stats like reliability or inconsistency ratings nobody seems to do this in a formal way.
In practice the winner of a large field handicap will need to run at or near best form to win, so I suppose you can do a two stage calculation of prob of best form x prob of winning with best form based on past consistency but it’s fairly crude. You need to know how volatile the form is. For example a horse that usually runs at 90 but does 105 for one race in 10 might or might not be a better bet than one that runs at 95 normally but never goes above 102. Only a full simulation of all horse performances in a race would give an accurate answer. But maybe there are other ways of working it all out. I know Tom Segal prefers unexposed runners in handicaps – I suppose this is backing horses with a wide distribution of possible form, which may make sense if other punters don’t allow sufficiently for this factor.
I had not thought much about this topic until I saw the stat that the average horse in the average race runs 20 lb below best form. Now that will include tailed off runners, but even so it’s pretty dramatic and makes traditional handicapping based on marker horses etc very problematic.
It seems to me for betting purposes that ideally one needs the full statistical distribution of a horse’s likely performance, especially in handicaps. But apart from inadequate proxy stats like reliability or inconsistency ratings nobody seems to do this in a formal way.
In practice the winner of a large field handicap will need to run at or near best form to win, so I suppose you can do a two stage calculation of prob of best form x prob of winning with best form based on past consistency but it’s fairly crude. You need to know how volatile the form is. For example a horse that usually runs at 90 but does 105 for one race in 10 might or might not be a better bet than one that runs at 95 normally but never goes above 102. Only a full simulation of all horse performances in a race would give an accurate answer. But maybe there are other ways of working it all out. I know Tom Segal prefers unexposed runners in handicaps – I suppose this is backing horses with a wide distribution of possible form, which may make sense if other punters don’t allow sufficiently for this factor.
Only a minority of horses in any race run to form, although am a bit surprised the average is 20 lbs below form. However, what is the definition of "running below form"? To have a number there must be a definition. Many horses win a race but don’t quite run up to their very top performance rating, may be did not need to do so, is this still counted as running 3 lbs (or whatever) below form? Even Frankel only ran to his very best performance rating two or three times. Were all the other runs "below form"? Or, is it only previous form that counts? So, if Frankel was constantly improving (bar the St James’ Palace) was he always counted as "running to form"? An older horse may have put up its very best performance as a 4 year old and not quite capable of running to that rating, so – does that mean every run from 4 year old to 8 (including wins) is rated "below form"?
Obviously there are a few it is virtually impossible to forecast inconsistency/reliability due to temperament issues; but with the vast majority most horse’s "inconsistencey" is due to being best with a certain set of circumstances. Some are more suseptible than others to aspects that might not be quite right to produce their very best. Just one (or more) of the following might cause any horse to run below form (suspect I’ve forgotten something): Going, distance, pace, whether the horse itself is in form, trainer form, time of year, draw, right handed or left handed, undulating/flat/sharp/galloping track, jockeyship, the right appendage/s, weight carried, fitness (race needed or come too soon or best fresh), got upset travelling to the course or in paddock/to post, convinced some don’t run well when it’s actually raining etc etc. So may be it is not surprising so few horses run to form and how far below form they run…
But their "inconsistency" is seldom random and their probability of running to form can be allowed for by looking at form. I don’t see why a seperate calculation for inconsistency is needed Kasparov, when the reasons for that inconsistency are in its form.
It really annoys me when people run down Timeform ratings, thinking that the "top rated" is in some way their "tip" (not you Kasparov). Without first looking at all the other relevent details of form. The Timeform Master Rating is what they consider a horse capable of given optimum conditions, there are many reasons (above) why a horse may not run to that Master Rating.
Value Is EverythingThe source is the onemilefourandten blog
The source of the data is Raceform Interactive (RFI) with the analysis carried out in the R statistical environment. Flat and NH races are considered separately. Races in GB and Ireland only since 2007 up to mid-January 2014 are considered. Each time a horse runs its Racing Post Rating (RPR) is compared with the maximum RPR the horse has achieved up to and including race date. The difference between the two numbers is defined as the RPR relative (RPRrel). The maximum value RPR can achieve is zero. Following Timeform’s definition, if a horse runs within 5lb of its maximum rating it is considered to have Run To Form (RTF). Horses are classified as having either high or low consistency according to the percentage of times a horse has RTF relative to a cut-off of 50%. The choice of 50% is arbitrary. No allowance is made for the number of times a horse runs, or its age in assigning a consistency classification, even though the older/more often a horse runs the more likely it is to run more than 5lb below its previous best. Races are classified as either handicaps, pattern races or other. For Flat races the analysis is restricted to horses aged 4 and above. The reason for this is to reduce the influence on the analysis of the return to racing after a long break of unexposed, previously immature horses. For similar reasons in NH races the analysis is restricted to horses aged 5 and above. Days off between races are classified into the following 6 buckets: up to 10 days, 10 to 29 days, 30 to 59 days (1m to 2m), 60 to 179 days (3m to 6m), 180 to 364 days (6m to 1y) and 365 to 730 days (1y to 2y).
There are several tables on the website. Average is 19.6 lb.
"It really annoys me when people run down Timeform ratings, thinking that the "top rated" is in some way their "tip" (not you Kasparov). Without first looking at all the other relevent details of form. The Timeform Master Rating is what they consider a horse capable of given optimum conditions, there are many reasons (above) why a horse may not run to that Master Rating."
So folks having a day out at the races and stumping up for a Timeform racecard are supposed to know that and know how to do that?
If Timeform could do it they would do it and make an assessment that under today’s non-optimum conditions the TF rating should be lowered by so many pounds. They don’t and they can’t – so you are annoyed at the wrong people – it should be at Timeform and the like that peddle this ancient snake oil to a gullible public.
The main function of trainers is to disguise ability, occasionally they let the cat out of the bag as it were. It’s these glimpses of ability that need to be followed up.
The main function of trainers is to disguise ability, occasionally they let the cat out of the bag as it were. It’s these glimpses of ability that need to be followed up.
Nonsense. The main function of trainers is to wear silly clothes and warble "he’ll jump a fence one day".
Mike
"It really annoys me when people run down Timeform ratings, thinking that the "top rated" is in some way their "tip" (not you Kasparov). Without first looking at all the other relevent details of form. The Timeform Master Rating is what they consider a horse capable of given optimum conditions, there are many reasons (above) why a horse may not run to that Master Rating."
So folks having a day out at the races and stumping up for a Timeform racecard are supposed to know that and know how to do that?
Oh come off it Robert.
If the Timeform Racecard write up says of the top rated "best on a
soft
surface", and it is
firm
ground; then the "top rated" should
not
be thought of as the
"tip"
.
It’s up to every individual punter/buyer to read their Timeform Racecard and come to their own conclusion over (A) what’s best suited by conditions and (B) what is value to win.
Very few first time racegoers buy Timeform anyway. Vast majority of customers being experienced punters that know how important aspects like going, distance, draw etc can be.
Value Is Everything"It really annoys me when people run down Timeform ratings, thinking that the "top rated" is in some way their "tip" (not you Kasparov). Without first looking at all the other relevent details of form. The Timeform Master Rating is what they consider a horse capable of given optimum conditions, there are many reasons (above) why a horse may not run to that Master Rating."
If Timeform could do it they would do it and make an assessment that under today’s non-optimum conditions the TF rating should be lowered by so many pounds. They don’t and they can’t – so you are annoyed at the wrong people – it should be at Timeform and the like that peddle this ancient snake oil to a gullible public.
Have you thought this through Robert?
When are Timeform supposed to assess "today’s non-optimum conditions" and write the racecard with adjusted ratings?
The day before? When the ground is good-firm…
In the morning? When the ground has changed to good…
Or ten minutes before start of racing? When ground is soft…
Or when they need to change it back to good-firm once it’s obvious the Clerk has made another ball-up with the official description?Ratings would be changing all over the place and subscribers wouldn’t know whether they’re coming or going. Much better to use optimum conditions for a Master Rating, with performance ratings of the last 5(?) performances; so subscribers can see what ratings the horse has put up on the relevent going.
I’d rather they keep it as it is thank you Robert.
Value Is EverythingThe main function of trainers is to disguise ability, occasionally they let the cat out of the bag as it were. It’s these glimpses of ability that need to be followed up.
Nonsense. The main function of trainers is to wear silly clothes and warble "he’ll jump a fence one day".
Mike
”Not today Josephine, he’s too short”, another oft repeated phrase.
- You must be logged in to reply to this topic.