relationship between carrying high weight and beaten lengths

[JPinMaryland]

Originally Posted by kev

See I can't even get into the whole if this horse is dropping 7 pounds from his last race he's going to run faster and trying to do study's on it is alittle out there. Reason being, me being a sheet user, horses there just not going to run the same speed fig. ever time out.....

But look at what I am saying: I am comparing Fager, a superhorse,running his greatest race vs Arazi another fine horse also running a great race.

And I am comparing them running the same distance, over the same track on a day when each one set a track record.

Isnt it reasonable to say that these performances are quite comparable?

ANd then I am going further and saying, if we take 22 lbs off Fager and make his wt the same, is reasonable to think he will beat Arazi by 3 lengths or 8 1/2?


Where do you disagree with this logic? or maybe you dont and you are just reponding to the othe method. threaded discourse often makes for confusion.

[Figman]

The unmeasurable factor as related to weight on or weight off is a simple one.

Every time a horse competes its form is affected.
Each race either sharpens a horse's form OR dulls a horse's form.

As has been written many times, they are living animals and not machines capable of carbon copy efforts each and every time out.

[JPinMaryland]

all right Artax was carrying a measley 114# on that day in 1999 when he set the 7f AQU record.

Fager was carrying 139, so lessee I am guesssing 4 lbs per length for sprints. 25 lbs difference divide by 4, equals about 6 lengths or 0.85 seconds. Subtract .85 from 1;20.2 and we get 1:19.35 for Fager vs 1;20 for Artax. Difference of .65 sec. or 5 1/2 lengths. Hmm, seems a bit high but perhaps within reason.

That Century dream site has a sprint race with the all time greats.. Fager wins and Artax comes in about 4 lengths back or so. hmm

[Overlay]

In Winning at the Races, William Quirin applied linear regression to 30,000 past-performance lines, and said the results showed that one pound is worth .09 lengths in favor of the horse carrying the higher weight. He also noted that horses carrying 120 pounds ran, on the average, one length (1/5 of a second) faster than did horses carrying 110 pounds. Horses dropping three or more pounds from their last start tended to run 1/5 of a second slower than they had in that start. Horses adding three or more pounds from their last start ran slightly more than 1/5 of a second faster than they had in that start.

[Tom]

Let's assume for a minute that somebody found a statistical relationship between weight carried and speed ratings. Let's further say that we have strong data to suggest that 5 pounds = 1 length. So now knowing that the average horse will speed up or clow down by a length for every 5 poinds change in weight. The problem is that when you look at the entries for a race, you won't find Average as a bettable horse. You have to choose a specific horse.

Do you end with the horse whose speed rating varies by a head, or the one whose rating varies by 2.5 lengths?

These are living creatures, not data points.

[JPinMaryland]

Okay that is a fair pt. tom, but I never said I was doing this in order to find a betting angle. I was more concerned with history, how fast can such and such horse run? what sort of time can we expect to be a track record at such distance? etc. so that is where I am coming from.

I agree that if you were to use this sort of thing as a betting angle it would seem pretty useless, especially if the wt assignments are anything close to reasonable.

[valueguy]

I primarily play small Cdn and Us tracks and its been my observation that weight seems to have an effect on horses running on an off track.They get out slower at the start (sprints ) or fade quicker in routes .I give well meant horses a little better odds than usual if they are dropping weight compared to the other horses." Everything is known by comparison" (Russian Proverb)
I have no data base to prove this observation but i have a trainer friend who
concurs with me.

[ryesteve]

Originally Posted by Overlay

In Winning at the Races, William Quirin applied linear regression to 30,000 past-performance lines, and said the results showed that one pound is worth .09 lengths in favor of the horse carrying the higher weight.

That's because the horses assigned the most weight are generally the better horses, which is why his data would show that they ran faster. I seriously hope that Quirin wasn't implying a cause and effect whereby weight speeds up a horse.

[Overlay]

Originally Posted by ryesteve

That's because the horses assigned the most weight are generally the better horses, which is why his data would show that they ran faster. I seriously hope that Quirin wasn't implying a cause and effect whereby weight speeds up a horse.

Of course, you're absolutely right. No one would suggest that additional weight speeds up a horse. The faster time is the result of the horse's better condition which, in turn, caused the horse to be higher-weighted in the first place. I mentioned the finding precisely to illustrate that better horses are assigned higher weights, but the weights come nowhere near their purported purpose of giving every horse in the race an equal chance of winning. However, as far as assigning a lengths value to weight, these findings might allow the development of an inverse relationship past a certain point (that is, the higher the weight, the greater the distance that the horse will finish ahead of a lighter-weighted horse, again based on the superior condition that caused the assignment of the higher weight).

[sjk]

I would look at it differently. If a horse carrying a higher weight tends to outperform what you would otherwise expect him to do and a horse carrying a lower weight tends to underperform, I would not think that weight was having a perverse effect. I would look at other factors that are causing you to improperly evaluate the horses performances.

For example, you might look at recency of the last race. The way condition books are written horses that have not run in quite some time will usually get in with a low weight. At the same time horses that have not run in a long time carry the risk of being out of condition or coming back from an injury with long-term implications. If you are not fully taking into account the fact that horses that have not run recently have added risk of underperformance, it may look like the weight is the cause.

[JPinMaryland]

I was thinking the same thing as value guy: that wts might have more influence on off track, or under a faster pace scenario. Simply because these conditions require the horse to expend more force and the F=MA relationship should mean that wt will be a factor here.

Hey I found Ainslies book in the used bookstore this afternoon and flipped through it, I was hoping to pick up that Conigliaro book but never got around to it.

[JPinMaryland]

I have to correct that formula I posted that was mentioned upstream. According to Ainslie,the "tried and true" formula is: 4lbs per tick at sprints, 3 per tick at 8f, 2 per tick at 9f and 1 per tick at 10f. I think I said 3 per tick at 7f. that was wrong.


Like most horse people Ainslie tries too hard to stretch a pt. He says that wt. cant equalize horses of different abilities and then goes onto to say that: he did a study of 100 major handicapping events of a given year and found that only 26% of the last place horse came within 15 lengths of the winner. And that the average lengths behind to the last place horse was 30 lengths.

he adds an exclamation pt to all this as if it proves something. All it means is that when you're behind in a race its best not to waste the horse. anyhow.

[Rick]

Read Chapter Six in Nick Mordin's book "Winning Without Thinking" for an explanation of why weight changes don't work like everyone expects.

[Bruddah]

1.then track management is foolishly spending good salaries for officials to write conditions for races. 2. Trainers and especially Jockeys have been paying way too much attention to a non issue, these many years.

As Ricky Ricardo might say, " I don't thin soo Loosie".

Just because weight might have an elusive effect, and is very difficult for handicappers to quantify, doesn't mean it's not a major race factor.

The real issue to me is the condition of the animal, carrying the weight, going the distance. A perfect example is Afleet Alex. His superb conditioning was very apparent in the Preakness and Belmont. But, how do you quantify it before the race takes place? It was certainly obvious to all, after the race. Then the question begs, why didn't he also win the Ky Derby?

In my humble opinion, some things will always remain unquantifiable in handicapping horse races. Weight, Post Position, Trainer Intent etc. etc. etc.

Youse makes youse best guess, pays yo money and hopes fo de best. "Oh Loosie, I'm home frum de club".

[Bill Cullen]

Originally Posted by ryesteve

That's because the horses assigned the most weight are generally the better horses, which is why his data would show that they ran faster. I seriously hope that Quirin wasn't implying a cause and effect whereby weight speeds up a horse.

Quirin at the time was a math professor at Adelphi University. I seriously doubt he would revise newtonian physics and imply that more weght causes a horse's speed to increase.

Bill Cullen