2014 Layoff & Paceline discussion
 

All-

I've seen a gentlemen with the handle "For The Lead" post several times about how he throws out horses who are on a layoff of 90 days or longer as they win less than 5% of the time.

At first I had some trouble with this concept, but after mining data on my own, I am able to wrap my head around it. Per my data, horses off over 90 days do in fact win at a very low amount of the time. In a previous layoff discussion, Richie correctly points out that there are 2 approaches (link: http://paceandcap.com/forums/showthread.php?t=628 ).

1. Screw it- if he beats you, he beats you- Sartin Method handicappers only need to win 50% of races to make $. I can handle having that 5% of winners off a layoff being included in the 50% of races I lose.

2. Ignore layoff rules and play the include the horse as a contender anyways.

Over the past weekend at Saratoga and Del mar, there were a few big favorites that were coming off long layoffs. Keying against them was very profitable for me.

MY question:

Since we've established that a horse coming off a layoff often times needs a race to get back into shape, what do we do with a horse that is running a second race off a layoff? The first race off a layoff won't be indicative of how a horse will perform today. Specifically. I have no idea which line to use.

Any insight would be appreciated.

Thanks,
Ryan

Last line is still the line
 

Ryan,

Going back to races prior to the lay off put's you in danger of rating a horse off an old form cycle.

Personally from my records IF the horse is not a contender based off the most recent comeback race its usually a throw out for me.

let your records be your guide!

Paul link

Paul-

I tend to agree with that. I wonder if there is a non manual way to track data of how many winners are off of a 2nd, 3rd, race off an extended layoff. It can be done manually, but I'm not trying to manually enter data into excel.

Regardless- Thanks for the response.

Ryan
The man who says to throw out horses off over 90 days
charts every race at every track every day. Statistically. his records show that ANY horse off over 90 days is a bad bet at all race tracks and all distance and surfaces.
If you follow this guideline then follow it ALWAYS and as Richie says after you lose
- screw it -
If you did decide to chart these long layoff- (over 90 days) winners
I am sure even hand entering the infor wont be too much work
because you will get very few over 90 day layoff horses to begin with.

Its funny how things that cause us to lose a race like a horse off over 90 days winning first back happen far less than we imagine. The pain of the loss
snaps us into negativity.
The snap we should feel is the rubberband on the wrist after we say as Richie P said -screw it- Its a bad bet.
Now snap that rubberband and snap back into what wins

Good Skill
Bill

Bill-

I wholeheartedly agree with you.

I unfortunately don't have the amount of data (I've just started mining data), experience, and resources that he has. I would love to explore that database though.

My intended message was: my small database has confirmed to me that 90-95% of winners have raced in the last 90 days. There's a huge difference between absorbing a comment on an internet message board, and proving it out yourself.

After accepting that fact, I have found that it often times creates an arbitrage opportunity. By throwing out horses that haven't raced in the last 90 days, I have been able to eliminate some high priced favorites over the past week. If I lose to a horse off a long layoff- I am completely OK with it (Although it hasn't happened in my last 30 races). After all, I only need to win 50% of my races to be profitable. I can barely settle on 2 horses to bet now, anything that helps me eliminate favorites as a contender is helpful. If I had an issue with 50% of my races, I'm subscribing to the wrong theories.

I would like to be able to prove out the success rate of a horse on it's second, or third, or even forth race off a long layoff. Right now, the RDSS exports that feed into my database show the last time a horse raced. There currently isn't an indicator of whether a horses is on it's second, third, or forth race off a layoff (and I'm even sure if there's anything useful in that information... yet).

Off subject- what ever happened to all the databases PIRCO built to prove out the corollaries? I would imagine that someone still has them.

One of the problems with the Track Master PPs for RDSS is that they do not provide trainer stats. DRF or Bris PPs would show the trainer's record with horses running 1st off a long layoff.

I generally get the Bris PPs in addition to RDSS for that and other information. I recommend Bris, as I find them to as good as DRF and much cheaper. Basic PPs are $2.95 a card from DRF and $1.00 a card from Bris.

layoffs can be quite tricky...check out FOLLOW UP #81 pages 34-35..and FOLLOW UP #77 page 66. Howard says on page 35 that demanding recency can cost you money. He says that if you care enough that you can do your own research...In order to be a valid contender the layoff horse should rank no worse that third after hides...there again as I've stated before on different factors...KEEPING RECORDS IS VITAL...plus I agree with you Mike..I also used BRIS along with RDSS...if the horse has run well in the past off a layoff...beware..if the price is right it's worth the extra money to bet it.

It is possible to see how a horse has done after a layoff, but it isn't easy. You can go to Equibase and look up the specific horse. All the life time starts are listed there by date. You then have to go through and find the long gaps, say 90 days or more, and see how the horse did on the 1st and 2nd races after coming back.

Problems are that with a younger horse, you may not find any meaningful data, and an older horse may have changed trainers a number of times over the years. And I would argue that the trainer's record with returning horses is as important as the horse's history.

At random, I looked up an older horse that is running in the 4th at Monmouth tomorrow: Taperrific. Click on the "Results" tab under the horse's name to see the full lists of starts.

Tom Brohamer
 

Hi Ryan,

Very interesting thread.

On the Brohamer Lecture Tapes in the audio section Tom discusses the issue of races following a layoff. If you haven't listened to them yet you may like to, but if you do I'd open a Word document and type as you listen. There is hardly any filler in the whole 14 hours.

Tom raises the following points that could be explored as part of a data mining exercise:

1) If a horse runs a strong race following a layoff, then expect it to run below that level in it's second race & then back up to something approaching the 1st race in it's third.
2) The third race back is statistically the strongest race.
3) It is easier for a horse to come back from a layoff to win a sprint than a route. (Think this was from Propacelines audio)
4) Following a layoff, it is easier for an E horse to hang on than it is for a S horse to get up late.

Frank Cunningham (Propacelines audio) recommends passing a race where the strong horse has had a layoff. I think following on from this point, if a horse has had a layoff, but was a bit of a dog anyway then layoff or not it was never going to be considered a contender.

Just to get a quick disclaimer in, I'm paraphrasing these points, so please listen to the tapes if you want to explore this issues.

Bob Purdy raised the point that if you are concentrating on a particular circuit then it behooves you to pay attention to the methods of the trainers, but then as Tom B states you could end up spending more time collating data than cashing tickets & to be fair it isn't the Methodology.

Another point following on from this though are issues Tom raised regarding class movement following a layoff:
1) If a horse is raised in class after a layoff then the trainer clearly thinks highly of the horse.
2) If it is dropped (especially with fast workouts) then the trainer is most probably trying to offload the animal to one of the other trainers.

If you have data mining skills i'd be particularly interested to see the outcome of any research if you are happy to share your findings on here.

One final point, not Methodology now, but Howard did review the work of Joe Takach in the Follow Up. If you are comfortable with the visual inspection of a horse in the paddock then perhaps this may provide a clue as to the horse's readiness to run well or not following a layoff.

Takach has a long series of articles about physicality at his web site, starting HERE. He also has a DVD on the subject that is much more useful, because you can actually see examples of what he is talking about. The usefulness of physicality in handicapping is limited unless you are actually at the track, because TVG or HRTV or track feeds rarely show the horses in the paddock, and coverage of the post parade is hit and miss.