Call me Crazy - but I think " Recency " is making a comeback

[BeatTheChalk]

Just checking over the last month or so. I have seen more than a few hot
favorites - go down. Many of them had not raced for 30 days or more.
The old rule of 30 days became a laughing stock in recent years .. because of the use of ...shall we say ..oh er ahh ..performance enhancement development syndrome.
My rule was always 35 days maximum.period. All but high class stakes and very high allowances. Then I wanted to see some activity on the track to
show me that the nag was at least Ambulatory.
Sooo call me crazy .. but check a few days at your local track - see if I
have lost it :lol: :jump: :D

[HUSKER55]

My grandfather used to tell everone, when I was a kid, that it took 3 weeks of loving care and a lot of work to get a horse back to form after a hard race. I don't know but I hope that helps.

I generally prefer my picks to be under 30 days and I do not see a lot of winners from layoffs more than 45 days, but it does happen.

:)

[Norm]

.
Sooo call me crazy .. but check a few days at your local track

I hope you are right. The hardest transition to make from old style handicapping to the so-called modern version has been the number of days between races. So, a horse made a good move in his last race, then took a 5-week vacation . . . huh ?

[njcurveball]

You might be onto something.

I checked 2007 in my database for dirt tracks (labeled fast) and there were 124 favorites who had been out more than 34 and less than 46.

They lost 5 cents for every $1 bet! ROI of 0.95

All favorites who had been out less than 35 days ago showed 1,006 starters and they lost 16 cents for every $1 bet. ROI of 0.84

Greater than 45 days out showed 176 starters and a loss of 22 cents for every $1 bet. ROI of 0.78

Push your recency elimation to 45 days and you have stats in your favor on fast dirt tracks! :ThmbUp:

Jim

[Overlay]

You might be onto something.

I checked 2007 in my database for dirt tracks (labeled fast) and there were 124 favorites who had been out more than 34 and less than 46.

They lost 5 cents for every $1 bet! ROI of 0.95

All favorites who had been out less than 35 days ago showed 1,006 starters and they lost 16 cents for every $1 bet. ROI of 0.84

Greater than 45 days out showed 176 starters and a loss of 22 cents for every $1 bet. ROI of 0.78

Push your recency elimation to 45 days and you have stats in your favor on fast dirt tracks! :ThmbUp:

I'm confused. I interpreted BeatTheChalk's initial point as being that favorites coming off layoffs longer than the "old rule" of 30 days are once again swinging back to being less profitable (which you seemed to start off by agreeing with). But then your data shows that the most profitable (or least unprofitable) favorites as a group (although a relatively small group) are those with last races from 35 to 45 days ago (which would indicate that layoffs of that duration are still not necessarily a negative factor).

[john del riccio]

There are 2 schools of thought here.


Get the horse right, find the right race, and shoot to win. This lends itself to more time between starts.

They need to earn to eat. This lends itself to less time between starts.

Circumstances play a role but the connections, (tr/own) and the type of horse
(young/old, m/f, sprinter/router) all factor into it.

John

[cj]

Every study I've found finds horses coming off a good race do best running back as quickly as possible. It isn't even close. The longer they wait to run back, the worse they do.

[Tom]

Doing post mortem's on losing races, I am finding that lack of recent action - race or work - is the number one reason I can find for the horse not firing.

[DanG]

The best recent study I have seen on layoffs was by Ken Massa in one of his newsletters.

If the link is inappropriate, please delete it…(Pages 8 – 12)

http://www.homebased2.com/km/pdf/HTRReport-MayJun07.pdf

[nobeyerspls]

You might be onto something.

I checked 2007 in my database for dirt tracks (labeled fast) and there were 124 favorites who had been out more than 34 and less than 46.
Push your recency elimation to 45 days and you have stats in your favor on fast dirt tracks! :ThmbUp:

Jim

Can you do the stats by odds within gender, i.e. winning fillies greater than 5-1 and off 30 days or more. I have found recency to be gender-specific and cash large tickets using that angle. The mantra is fresh fillies/cold colts and it has worked well for me.

[john del riccio]

The best recent study I have seen on layoffs was by Ken Massa in one of his newsletters.

If the link is inappropriate, please delete it…(Pages 8 – 12)

http://www.homebased2.com/km/pdf/HTRReport-MayJun07.pdf

Dan,

Thank You for posting this article. It was truly an eye opener. It certaily does seem to show that giving a horse MORE not LESS time to recover from its previous race is the way to go.

I do have one question that maybe you can answer. It would be very very interesting to know (and quantify), what the horses PREVIOUS performance
before its layoff when put in terms of its overall ability.

John

[boomman]

Doing post mortem's on losing races, I am finding that lack of recent action - race or work - is the number one reason I can find for the horse not firing.

Tom has especially hit the nail on the head when it comes to lack of recent works. As long as you are looking at a track in which a horse has the ability to work (weather not withstanding), you have to take a strong look at that. One of the largest negatives I have found especially when it comes to beating short price horses, which we all strive to do, is a horse coming off a big effort 35 days or more ago and then having no work or anything visible since that race. If you also noticed that the horse did work between races before it says to me vividly that something went awry with this animal. I find that you also often see this kind of horse show up on the track after the lay-off with front wraps for the FIRST time and that sort of "seals the deal" for me in taking a stand against him/her. I think we would all agree that finding vulnerable favorites on a consistent basis is an overall key to beating the game. There are of course, many other factors involved in "beating the chalk" but this is a key one for sure!

[njcurveball]

Can you do the stats by odds within gender, i.e. winning fillies greater than 5-1 and off 30 days or more. I have found recency to be gender-specific and cash large tickets using that angle. The mantra is fresh fillies/cold colts and it has worked well for me.

You might be onto to something here. However, since slicing samples based on one primary factor (betting action) is not in line with doing all of the other slicing.

That is more along the lines of Multiple Regression, weighting, and other modeling techniques which would create many more probabilities and scenarios than desirable for a simple thread like this.

Jim

[njcurveball]

I'm confused. I interpreted BeatTheChalk's initial point as being that favorites coming off layoffs longer than the "old rule" of 30 days are once again swinging back to being less profitable (which you seemed to start off by agreeing with). But then your data shows that the most profitable (or least unprofitable) favorites as a group (although a relatively small group) are those with last races from 35 to 45 days ago (which would indicate that layoffs of that duration are still not necessarily a negative factor).

I was "sugar coating" my reply to try to agree with him. Many times these threads become a debate.

I also found that days away for turf and artificial track actually ENHANCE the profits of favorites on those surfaces.

Everyone wants to believe what they want to believe, even when cold hard facts say different.

A recent race MAY mean a higher win%, but it definitely is more profitable playing horses who have been away longer than 35 days.

Jim

[DanG]

Dan,

Thank You for posting this article. It was truly an eye opener. It certaily does seem to show that giving a horse MORE not LESS time to recover from its previous race is the way to go.

I do have one question that maybe you can answer. It would be very very interesting to know (and quantify), what the horses PREVIOUS performance
before its layoff when put in terms of its overall ability.

John
I’ll try and run that John ASAP.

I have a long macro running on the machine that is set up for impact values. Generic stats on something that groups layoffs with all the duplicate data will be deceiving.

[Tom]

I think we are getting into an area that is ambiguous. When I said recent actin ws the reason I could find for not firing, it was in the scenario of a horse ran a sharp race 35-40 days ago and showed works in the meantime, while another contender worked out twice.

I have no objections to betting horses of longer layoffs since Ken's study, but I want recent works. I would no prefer a horse off 45 days with several recent works over a last race 30 days ago and no works. Again, the usual caveat - be strict with favorites and loosen up on higher odds.

Not would be intersting to see if we have the data to querry days off in conjunction with recent works.

[njcurveball]

but I want recent works..

I totally agree with you Tom. The BIG problem is that many times the horse works out every 5 or 6 days and they only put one workout in the form. Other times they do not even have a workout in the form.

Trainer intent is also a big part of it, but it goes beyond this simple thread.

I have to trust that if the trainer has the horse spotted correctly, they have the horse ready to run, even if there are no published workouts.

And of course there are some trainers who automatically bring up a red flag if their horse isn't back quickly after a good race.

Nothing is set in stone.

Jim

[Robert Fischer]

I have more success with individual cases than looking at a number of days. If I see something unique in terms of recency, I ask some basic questions:


"7 days off huh"... was he a contender here 7 days ago ? what the heck did he do in that race? ... Huge difference between a horse that started slow/rushed up/circled the field 5wide/got whipped on the turn by the jock/etc - "hes gonna have nothing left"!.... than a horse who saved ground in some traffic in that turf race and never fired.


"6 months off huh"... was it an "injury" or a "layoff" , did he run good in his last race? what is the pattern for public works?? what did that bloodhorse article say it was again?? oh the tedon...Can i see it on the replay?.. This trainer just brought that 3yo back from a tendon how'd he do?...

[DanG]

I may not get back to this, so a quick generic look at last out winners & ran last.

The number to the left is the layoff zones referred to in the Massa newsletter ( http://www.homebased2.com/km/pdf/HTRReport-MayJun07.pdf ) and to the right is “1” for last out winners and “5” for ran last. Only fast dirt sprints in this all-burger sample. (Last 730 days)

Now…We all big boys and girls here. We all know everything in racing is applied within context. The newsletter does point out however, that some “traditional” layoff zones are not what they seem. Just as this data needs IV’s to separate the dups and all the other underlying noise that a well thought out query would include.
LAY/FIN COUNT WINS W% WP% ROI AVG$ MAX$
11 3039 615 20% 37% $0.91 $9.00 $62.40
15 4175 263 6% 13% $0.54 $17.26 $76.80
21 16998 3024 18% 33% $0.82 $9.17 $121.00
25 15072 953 6% 14% $0.54 $16.93 $183.80
31 11634 1984 17% 33% $0.75 $8.83 $90.40
35 9126 666 7% 15% $0.69 $18.87 $210.40
41 12397 2071 17% 31% $0.73 $8.71 $85.80
45 11962 953 8% 16% $0.66 $16.47 $182.00
51 2610 349 13% 26% $0.71 $10.62 $83.60
55 5123 365 7% 14% $0.72 $20.19 $226.40
61 2771 304 11% 21% $0.53 $9.74 $62.50
65 5487 357 7% 14% $0.56 $17.25 $205.20

[bellsbendboy]

Clearly most winners have recency on their side as horses that are not "fit" seldom win. As for fillies, they are much better coming off the shelf because they are easier to get fit.

Although first time front bandages concern many players, oft times they are a HUGE plus. Many trainers add fronts if there horse changes surfaces, races on a floated track or simply switch tracks. Bandages are an untapped handicapping topic.

[BeatTheChalk]

I'm confused. I interpreted BeatTheChalk's initial point as being that favorites coming off layoffs longer than the "old rule" of 30 days are once again swinging back to being less profitable (which you seemed to start off by agreeing with). But then your data shows that the most profitable (or least unprofitable) favorites as a group (although a relatively small group) are those with last races from 35 to 45 days ago (which would indicate that layoffs of that duration are still not necessarily a negative factor).

I apologize for my confusing post :bang: But it brought out some great
stats and info. :ThmbUp: Lucky for me it was my first mistake ..ever :lol: :jump:

[classhandicapper]

Every study I've found finds horses coming off a good race do best running back as quickly as possible. It isn't even close. The longer they wait to run back, the worse they do.

:ThmbUp:

I would only add, "if they are coming off a good race".

If a horse is coming off a bad race, it might actually be better for it to not come back so quickly.

I think the key point I am trying to make is that the quicker the horse returns, the more likely his form has NOT changed. Just looking at ALL horses in a category can be very misleading.

[BlueShoe]

For decades my favorite play has been an impressive winner ala Ainslies big win coming back quickly and stepping up in class.Very rare these days,but those returning in 10 days or less get best bet status.Even rarer are those back in 5-7 days.Those get super best bet check marks.When they come back this soon the trainer is telling us that the win took nothing out of the horse and that they are coming right back to grab another purse.

[Norm]

There is a special problem with horses that have been away 30 days. While there are lots of reasons why a horse may need a rest, there is one reason that is more common than others; bleeders. Everyone already knows this but, just to refresh your memory, glance at any DRF and count the number of horses on Lasix. O.k., no surprise, right ? But here's the drill : First time bleeding - barred from racing and on the vet's list 'til a Lasix program is confirmed by trainer's vet. First time bleeding with Lasix - on the track vet's list for 14 days and cannot return to racing 'til track vet observes a timed workout with no bleeding. Second time bleeding with Lasix - on the track vet's list for 30 days and a workout required. O.k., now here's the one trainers fear most : Third time bleeding - on the list for 180 days (the exact procedure and number of days varies by jurisdiction, but most have something similar.)

So, when you see a horse away for 30 days you must wonder, is he rested and had a sharp workout, or is he coming off of the vet's list, had a pre-qualifying workout and may be one race away from a long vacation ? How often does this scenario play out ? . . . count again the number of horses on Lasix. This is one reason why horses don't run back as often as in the old days. Get on the list and lose a lot of expensive training time and the horse may be retired before anyone wanted.

But, wouldn't it be nice to see the vet's list and be informed ??

[dav4463]

As long as they don't have that DRF "double line", a layoff of a year or more, I will still consider the horse, especially with some solid recent works.