Exceptions To The Rule - How Deep Can You Go?

[partsnut]

Exceptions To The Rule - How Deep Can You Go?

I worked this race on Sunday and originally, did not come close to the eventual winner. So, I went back after the fact, to see if I could make a case
for the winning horse. I did find that on second glance one might notice that this horse had a better then 20% winning trainer and had a definite pattern of coming back very strong after a layoff. The horses work out pattern were 3 workouts withing 15 days at representative speeds. My biggest question in this race would be how far should I go back with my pace line selection. Normally, I never go back past 90 days but I noticed that this horse had
a very strong pace line going back 139 days.

I believe the rule is to go back and use one of the horses last 3 lines.
Are there exceptions to this rule and how many days should I normally go back?
All opinions would be greatly appreciated.

Here is the RDSS read-0ut for the race in question.

[Seven Furlongs]

That's a good question.

I'm still trying to define what the acceptable parameters, parameters rather than rules, are for paceline selection. Rules can be dogmatic, parameters allow for some flexibility. I approach paceline selection as if it were a bit like Jazz - a bit free form and loose, but there should be some structure, somewhere.

I look at workouts and see if some patterns exist between workouts and performance to get a sense of form. Also, I'd see how the trainer brought the horse back to form under similar circumstances and see if the horse is being brought along the same path if it was successful. I also, and this is just theory as I struggle with it in practice, try and pin a paceline down to what the form cycle of the horse is/will be for the upcoming race.

One of the best quotes I read was by Mark Cramer who said something to the effect that to be successful in handicapping you have to predict change. Sometimes the last three pacelines don't capture the apparent change in form and you might have to dig deeper. Still, my general idea is to try and find a paceline that will reflect the predicted performance of the horse.

Hope this helps, but I am somewhat new to this so maybe my ideas shouldn't be considered as definitive, but rather a work in progress.

[BJennet]

Originally Posted by partsnut

Exceptions To The Rule - How Deep Can You Go?

I worked this race on Sunday and originally, did not come close to the eventual winner. So, I went back after the fact, to see if I could make a case
for the winning horse. I did find that on second glance one might notice that this horse had a better then 20% winning trainer and had a definite pattern of coming back very strong after a layoff. The horses work out pattern were 3 workouts withing 15 days at representative speeds. My biggest question in this race would be how far should I go back with my pace line selection. Normally, I never go back past 90 days but I noticed that this horse had
a very strong pace line going back 139 days.

I believe the rule is to go back and use one of the horses last 3 lines.
Are there exceptions to this rule and how many days should I normally go back?
All opinions would be greatly appreciated.

Here is the RDSS read-0ut for the race in question.

Hi Bill,

I had this one because I look for horses like this - a horse with one of the two highest speed figures in the field going off at 11-1. For me, a no-brainer.
Recency is extremely overbet by the public. If you do a search on Dave Scwartz's posts, you can find his stats on this. The IV for recency is 1.00, which means that, for any given level of recency, horses win exactly their fair share of races. Thus it's a useless handicapping factor. But since the public believes in it, it's hugely negative ROI.

Also higher-class grass horses coming off relatively long layoffs (6-9) months, are flat-bet positive ROI. A 2+-month layoff, as in this race, is nothing for a good grass horse. For the exact stats, I would take a look at Ken Massa's research on this subject, which is on the HTR site. All I know is that I've done very well betting this kind of horse.

Cheers,

B Jennet

[SilentRun]

Originally Posted by partsnut

Exceptions To The Rule - How Deep Can You Go?

I worked this race on Sunday and originally, did not come close to the eventual winner. So, I went back after the fact, to see if I could make a case
for the winning horse. I did find that on second glance one might notice that this horse had a better then 20% winning trainer and had a definite pattern of coming back very strong after a layoff. The horses work out pattern were 3 workouts withing 15 days at representative speeds. My biggest question in this race would be how far should I go back with my pace line selection. Normally, I never go back past 90 days but I noticed that this horse had
a very strong pace line going back 139 days.

I believe the rule is to go back and use one of the horses last 3 lines.
Are there exceptions to this rule and how many days should I normally go back?
All opinions would be greatly appreciated.

Here is the RDSS read-0ut for the race in question.

Hello,

I believe there were some mention of examples of exceptions in the Follow-Up such as:

If the last several races were on a different surface then it is ok to go back
to the first comparable line to today's surface.

Or if the last several lines were run on a different track condition then today...say today's track is fast and the last 3 line were sly or my then it is ok to go back to the 4th or first comparable line.

Specifically for Turf races...Mike Pizzolla in ,"Handicapping Magic" advocates going BACK TO ANY OF the 2 best SR's for the horse. For Turf races I would also add look for the 2 best TPP's lines. Since the race you cited is on the Turf then it is ok to go back all the way and choose the appropriate lines.

In general I Personally also...

look at the trip notes and situations like blocked, steadied....then I would discount that line and go back further.

If the horse ran at a higher class in its last race but performed poorly then I would ignore that last line.

These are some examples are judgement calls that you would have to make along the way.

Ernie

[Ted Craven]

I always first consider the strongest lines within the last 3 or so (excepting non-comparable ones, e.g. surface, distance, bad trouble, etc). I use the Perceptor Total as an initial guide for this (others use Adjusted SR, or Total Energy). You decide on an absolute cutoff date for a representative paceline: e.g. 90 days, 180, 280, but I tend to keep it within 9 months max, but the earlier the better.

THEN, I try to find reasons not to use that best line for a horse (i.e. too high rating, won't be repeated, etc). If it appeared that this horse hadn't done ANYTHING since its 3rd line back, it would be a different story: there would be real questions as to current fitness and ability to duplicate that earlier effort, despite the layoff and 3 (middling) works in 27 days.

But its win 51 days ago tells that (at least then) it was fit enough to set or overcome the pace it needed to. Importantly, we need to decide if we know why the ratings declined since the 3rd line back. Sometimes we don't know; sometimes the horse was clearly going off form. Consider the Pace of Race Total Energy - PoR (2nd right-hand column for the horse). This horse likes to run earlier, but it either chased or set a slower PoR in each of its last 3 races because that's as fast as it needed to run to be in its preferred position. Consider the axiom that 'a horse runs as fast as it needs to, not as fast as it can'.

SO - it ran as fast as it needed to, and the declining Total Energy and Perceptor rankings don't necessarily reflect a tailing off of form. Perhaps it was tailing off - it got a 60 day rest - and perhaps the connections were seeking a good placement for the next outing (or simply waiting for Aqueduct to open).

All of this is debatable subsidiary analysis attempting to answer the question: should I use the best line for this horse. I always attempt to take a more conservative route when thinking how to represent a horse today from its past performance. If we use the 3rd line, note how it ranks on the BL/BL screen - the best by nearly 5 rating points over the next horse, the #3 - that is a lot of BL points! So, a conservative approach would be to ask: how many points less, or how much worse a performance than the 3rd line back would still make this horse a bettable contender today. If the rank positions of the #6 and the #3 (the 1st and 2nd ranked) were reversed, and the odds were still 10-1, the #6 would still be a bet, so we can allow a seriously worse performance today than that 3rd line shows and still bet on the horse at 10-1.

That is some of the thought process I go through when rationalizing using an older paceline. Apply these criterea for every horse though. BJennet's point is well made (and before Massa or Schwartz, Sartin was making the same point) - recency is more or less well priced. The more cautious bettor will receive the more cautious price. You will lose a few of these, because (as John from Cincinnati says) 'some things we know and some things we don't know. But this kind of horse is precisely what the modern Sartin Methodology (i.e. late 1990's +) was designed to identify.

Ted

[partsnut]

I have to say that I'm totally impressed with the feedback I've gotten on this and wish to thank ZipZap, BJennet, SilentRun, and Ted Craven.

I believe you guys are very knowledgeable and commend you for being willing to share you're knowledge with the membership.

[raceman5]

Hi Bill,

I had the privelidge many years ago at a Sartin seminar in albany ny to sit down with michael pizzola and talk handicapping. I asked him about choosing lines for layoff horses. He told me he always chose the layoff line. So on your hoesw you could of went back as far as you want being he does a great job running off a layoff.

Bob

[Bill V.]

I'm trying to avoid a horses best line ever if its over 90 days ago
Horses like # 6 come along for me and its a tough call.
Thanks for this post

GS
Bill

[alydar_ David]

A few thoughts.

First, Mark Cramer said he'd rather use any line other than the last race. His rationale was the horse just ran that race and is not likely to repeat it.

Barry Burkin recommended using a paceline from about the same amount of layoff time and the same class as the layoff.

Personally, I have to see some indication that the horse might fire on the turf. Some of those not so obvious signs might be:

* Layoff from November to spring, two or three bad dirt races, then turf today.

* Layoff, one grass race five or six lengths worse than his good turf race, then turf today.

If either of those conditions exits I have no problem using the "Way back machine."

But I won't just automatically take the horse's best turf race in his past performances simply because he's running on the lawn today.

Ernie and I talked about this at Gulfstream. He's of a different school of thought. That's cool.

:>)

[shoeless]

I hate to bring this up but I will look at the trainers record for horses coming
off a layoff.I know these long layoff horses will win once in awhile but in the
long run they will eat up your bankroll.

I will be more lienent on days off with turf horses.

Ted,

Anyway to shrink your post so that it will fit on my screen instead of
me having to move the bar to read it.

Jeff