Performance correlated to days since last race

[kid4rilla]

Searched the site for this information, but couldn't come up with anything.

I remember seeing in the Equibase programs a pie chart and break down of how horses with varying days since their last race fared by percentage. Kind of like....

Raced within last 7 days: x%
Between 7 and 14 days: x%
Between 14 and 30 days: x%

...

Has anyone done any similar calculations, or know where I could see something along these lines?

[cj]

I can only speak for my figures, but horses coming off of good races perform best the sooner back they run, by a good margin.

[Overlay]

Mike Nunamaker's data from 111,700 horses in 12,815 races in Modern Impact Values (1994) showed the following:

1-5 days: 863 horses; 99 winners (11.47%); 191 2nd or 3rd
6-10 days: 15,768 horses; 1,887 winners (11.97%); 3,828 2nd or 3rd
11-21 days: 49,354 horses; 5,831 winners (11.81%); 11,952 2nd or 3rd
22-30 days: 16,730 horses; 2,009 winners (12.01%); 3,978 2nd or 3rd
31-60 days: 13,069 horses; 1,539 winners (11.78%); 2,906 2nd or 3rd
61-90 days: 3,120 horses; 321 winners (10.29%); 950 2nd or 3rd
91 days or more: 12,796 horses; 1,129 winners (8.82%); 2,138 2nd or 3rd

William Quirin's earlier (1979) data from a smaller sample (17,708 horses in 2,021 races) in Winning at the Races:

1-5 days: 558 horses; 72 winners (12.9%)
6-10 days: 6,086 horses; 806 winners (13.2%)
11-21 days: 7,079 horses; 797 winners (11.3%)
22-30 days: 1,565 horses; 157 winners (10.0%)
31-90 days: 1,584 horses; 137 winners (8.6%)
91 days or more: 836 horses; 52 winners (6.2%)

[bellsbendboy]

"Smoking or next available"? If your horse is jumping out of his skin, or your date is positively starving, you may settle for next available.

On the other hand if your horse had a tough race and needs more time, or if your companion has the next day off, you may wait for non-smoking.

My point is using the condition book, which is available for most tracks at each of their websites, often provides clues, (available races and extras) that makes runback decisions much easier! BBB

[bobphilo]

"Days since last race" used to be a major elimination factor for me but recently many trainers specialize in getting their horses to run their best race fresh. I'm a big believer in using stats for predicting how a certain category of horse will perform in general but this is one factor where it pays to look at how the particular horse or trainer does off the lay-off.

Whenever I come across a horse coming off a lay-off I check to see what the trainer's win percentage is when coming back since about this same number of days. If it is as good or better than his overall percentage, the horse qualifies. If I can find a similar lay-off in the horse's PP's, I'll look to see if his figures improved or declined in his comeback race. Recent works in good time at 5F or more or also good signs when in doubt, but not as important as how the horse or trainer deals with this particular lay-off.

Class also can determine if days since last start is important. A claimer that is inactive too long may have soundness problems while, with a classy horse like Discreet Cat, they just may be picking his spots.

BTW, Discreet Cat's performance in the upcoming Dubai World Cup may be an interesting test of how recency of last start may be meaningless when applied to certain horses.
Bob

[46zilzal]

Horses are individuals. There is no hard and fast rule for recency.

Imagine trying to tell which major league pitcher was "in shape" in April over the same condition in June......No hard or fast rules in either situation.

[BIG49010]

I think that Condition Books have changed in the last few years, you see a 21 day spacing for most of the better races. If a horse gets excluded from a race, then you have to pick the wrong distance, surface, class for your horse.

[kid4rilla]

Thanks for the replies. I know there is no hard n' fast rule, just wanted to see that data for a starting point.

[Robert Fischer]

there is also some days/performance correlation done with individual sires...

[misscashalot]

The following is fr the NYRA circuit and includes the average time coming back for a win only. Not losers.

Claimers 27.12 days (11,282 winners)

Maidens not incl FTS 37.06 (12,013 winners)

Allow fr GI to Starter Allowance 35.74 (15,734 winners)

I post this just to point out that different class racing produces different averages.

[JackS]

Sometimes you have to "reckon" the best you can particularly with maidens and claimers.
If the meet is going full steam without off days due to weather or other reasons, claimers should be running back within reasonable time periods.
If the general population of claimers are run back within a 2-3 week time period, this time can be considered normal.
As with everything in handicapping, you might want to give a little i.e- 23-25 days. There could be an acceptible excuse that you and I will never know about so it would come down to the horse and odds.
Lengthier time offs could become important and a possible reason to "toss".
No one is making money on a 10k claimer that hasn't raced in 4-5 weeks or more who was probably laid-up due to physical problems.
No exact answer to this question and the best you can do is try to reason with yourself when you see these types of horses.

[misscashalot]

.........
As with everything in handicapping, you might want to give a little i.e- 23-25 days. There could be an acceptible excuse that you and I will never know about so it would come down to the horse and odds.
Lengthier time offs could become important and a possible reason to "toss".
No one is making money on a 10k claimer that hasn't raced in 4-5 weeks or more who was probably laid-up due to physical problems....


A claimer coming back in 21 days or more, even after a good or winning effort is a poor sign except if the horse has had at least one work since that last race or is dropping in class after a good race. 21+ days No work...then no bet for me.

[JackS]

I used the qualifier "general population". If it's 14 days as an example, 21 days may be a bit much. Rules are good when rules apply. In handicapping it's always generally speaking.

[ranchwest]

The following is fr the NYRA circuit and includes the average time coming back for a win only. Not losers.

Claimers 27.12 days (11,282 winners)

Maidens not incl FTS 37.06 (12,013 winners)

Allow fr GI to Starter Allowance 35.74 (15,734 winners)

I post this just to point out that different class racing produces different averages.

Thanks for those stats. If I remember correctly, I think there's also a difference between sprint and route.

[cj]

One thing I use in my best mechanical play uses time between races as a big factor.

If a horse has my top overall rating, or PF for me, and is in the top 3 in early speed, on any surface, and returns in the PF/3 days or less, it is an automatic play.

Why the PF/3? I tried set limits, like less than 21 days, less than 28 days, etc. I originally settled on 21 days. However, I saw I was missing a lot of winners in higher class races.

The PF/3 gives faster horses a little more time to return. So, a horse in a cheap race with a figure of 60 has 20 days, while a horse with a figure of 100 has 33 days, and a top G1 horse with a 115 has 38 days. I can't say why it works, but it does, which is all I really care about.

[misscashalot]

I used the qualifier "general population". If it's 14 days as an example, 21 days may be a bit much. Rules are good when rules apply. In handicapping it's always generally speaking.

In handicapping, generalities when betting leads to confusion, at least for me. I find a standard in a situation and stick with it if I find it makes me money. For me, in claiming races for 3up and 4up, using the 20+ (19 days is my actual cutoff) , no work, is an automatic throw-out, unless good past race and a drop down, is profitable; and then, if the horse meets the minimal standard, I look at the trainers past successes. Sure I lose some, but in the long run, I do OK with my hard fast rule.

[AAcoolguy]

I used to exclude horses that hadn't raced for 30 days, then 45 days, then 90 days. I found that I was excluding many winners that way. What I am trying now is only excluding horse that have had no activity at all in the past 45 days (1 race + 3 works, 2 races + 2 works, etc.). This will only exclude one or two horses per race, but I also assure myself that I am not throwing out a potential winner.

[Robert Fischer]

I don't do any automatic eliminations , but I can see where that would help with filters using a database.
With horses that you don't really know about, you try to find clues with time off. A long layoff can be a negative. I want to see multiple workouts about 7 days apart without a break. I prefer a class drop. I break down time off into "layoff" or an "injury time off" (meaning was the last race a solid in-form performance? or did the horse quit?)...
with claimers - you don't get a lot of interim works. A break without a work usually means unsoundness so I am hesitant to back these without other explanations.

[misscashalot]

I used to exclude horses that hadn't raced for 30 days, then 45 days, then 90 days. I found that I was excluding many winners that way. What I am trying now is only excluding horse that have had no activity at all in the past 45 days (1 race + 3 works, 2 races + 2 works, etc.). This will only exclude one or two horses per race, but I also assure myself that I am not throwing out a potential winner.

This is a complex subject. If you are keeping records, you will see that Allowance (GI thru Starter Allowance) runners win when laying off longer than Claimers. Trainers have their patterns, some bring em back running after long layoffs better than others. Times of the year also are a consideration. It is no doubt that no matter how long the layoff, a work(s) is necessary. The type of work which signifies that a layoff runner is well preped is dependent on the distance of race he comes back to. A 2 turn race return should be precluded by a series of 4 furlong or 5 furlong works within 4 to 15 days of the return. For a one turn return a 3 furlong work 2 or 3 days prior is a wake up call to me. This 3 furlong work after a five furlong work is gold. Look for it in a one turn return. Uncommon works of 6 or 7 furlongs prep a runner for a stretching out. On the New York circuit, Gary Contessa is a master with the 7 furlong or 1 mile work close to race day with runners coming back after a long (30+ days) freshening. This is just scratching the dynamics of the subject.

[BlueShoe]

Back in the '90s an outfit called Sports Stat in Las Vegas published a very detailed study on favorites using a number of factors.One of the surprises was the conclusion that favorites away 30 days or more actually did better than those with a more recent race.By better,that means a smaller loss (roi).While I will play a runner coming back in 30-60 days,this is a rather grey area for me,and absolutely insist on recent works.A runner away more than 21 days with no work or works since is viewed with great suspicion by this player,and animals away more than 30 should show at least two,45 three,and so on.We all agree than better horses can be excused being away,often there is no race in the condidtion book for them,or if there is,the race doenst fill.

[misscashalot]

Back in the '90s an outfit called Sports Stat in Las Vegas published a very detailed study on favorites using a number of factors.One of the surprises was the conclusion that favorites away 30 days or more actually did better than those with a more recent race.By better,that means a smaller loss (roi).While I will play a runner coming back in 30-60 days,this is a rather grey area for me,and absolutely insist on recent works.A runner away more than 21 days with no work or works since is viewed with great suspicion by this player,and animals away more than 30 should show at least two,45 three,and so on.We all agree than better horses can be excused being away,often there is no race in the condidtion book for them,or if there is,the race doenst fill.

On the NYRA circuit

winning fav coming back since last race

1-9 days (702 winners) Odds $1.45
10-19 (3664) $1.46
20-29 (3735) $1.37
30-39 (1410) $1.39
40-49 (743) $1.37
50-59 (326) $1.39
60+ (998) $1.40
FTS (547) $1.41

[JohnGalt1]

If when handicapping first time starters you look up their breeding to see how likely they are to win, you may want to also look up their breeding after a 90+ day lay off as part of your handicapping along with trainer % and pace lines after lay offs.

I do.

A horse you might pass on other factors may have an A or B sire for first time starts can be included in a pick 3.

A horse that looks good on other factors but has a F or D sire may be downgraded from a prime win bet to a pick 3.

[maxwell]

Two simple rules :

pass any horse that is coming off an official layoff ( 45 days ) ... unless the horse won it's last layoff in an odds line that benefits you.

pass any horse that has raced four times in the last 45 days. Trainer desperation has set in.

I no longer look for the Holy Grail of racing or workout conditioning :

6 pts. - raced in the last 15 days.
5 pts. - raced in the last 30 days.
4 pts. - raced in the last 45 days

3 pts. - workout in the last 15 days.
2 pts. - workout in the last 30 days.
1 pt.. - workout in the last 45 days.

You would think horses with 15 -18 - 20 pts. would be better plays than horses with 8 - 10 - 12 pts. Wrong. I did it for a year only to find it made no difference at all.

Speed points and "Respectable Pars" help me much more. I prefer not to play layoffs, and I don't even bother to look at workouts any longer.