Purse Value

[mkkash]

ERROL:

What is the name of the horse?

I will research him online just to see how close the formula is.

Best of Luck

MKK

[RXB]

Originally Posted by mkkash

If it sounds like a duck it might be a Poodle!!!

Every bit of Data is important in horse racing it all means something or
more than one thing.

Think Smart

MKK

GIGO

[sam4022]

Originally Posted by Ocala Mike

What stu said. I used to use the following "average" distributions when figuring this angle (I don't anymore):

1st - 60%
2md - 20%
3rd - 12%
4th - 8%

I have a Purse Level App available for Apple iPhone & iPad at $3.

[raybo]

Originally Posted by Errol

I'm trying to remember the formulae for figuring the average purse value a horse has been racing for.

Example: Eleven starts this year and earned $42,690.
42,690. divided by 11 = 3881 per race.

But I want to know the average purse size.

The horse has:
2 wins; 1 place; 4 shows & 3 4th place finishes.

There is a factor for each 1st, 2nd, 3rd & 4th place finish which Divided into the $42,690 will give the average purse size the horse was running for. This was a powerful factor years ago (no it did not make a flat bet profit).

I will appreciate any help thank, you so much.

Vaya Con Dios
Errol

If you're just wanting the average purse size that the horse has been running for, why not just use its last 10 races (or 12 races, or however many races you have access to in your data), and average each of those races' purses? That won't get you its lifetime average purse, but why would you want to go back that far anyway?

I must be missing something somewhere.

[Errol]

Hi All

I apologize for not getting back sooner. I’ve been running a test on QuickHorse. I use the tote board to make a go/don’t go. I’m still not done testing.

QH does NOT give me Purse values so I need to convince the programmers of the necessity of adding that data to it’s data bank.

So here’s what I came up with(I hope I don’t garble this).

This imaginary Horse has 5 races this year (actually that’s all you need) and 10 races last year 15, Total.

3 wins, 2 2nds, 4 3rds, and 6 4ths everyone of it’s races were for a $9000 purse. It earned $27,000 for the two years.

I set it up like this

3 x .6 = 1.8

2 x .2 = .4

4 x .14 = .56

6 x .09 = .54
________3.3 x 27,000.00 = 8910.00 which is close enough for govment work.

The 8910.00 is a rating. Yes there is slop in the figures. This rating alone will not make a flat bet profit. Years ago I used it as a class indicator. The Kelco Class Calculator works in a similar fashion(I think).

Vaya Con Dios

Errol

[raybo]

Originally Posted by Errol

Hi All

I apologize for not getting back sooner. I’ve been running a test on QuickHorse. I use the tote board to make a go/don’t go. I’m still not done testing.

QH does NOT give me Purse values so I need to convince the programmers of the necessity of adding that data to it’s data bank.

So here’s what I came up with(I hope I don’t garble this).

This imaginary Horse has 5 races this year (actually that’s all you need) and 10 races last year 15, Total.

3 wins, 2 2nds, 4 3rds, and 6 4ths everyone of it’s races were for a $9000 purse. It earned $27,000 for the two years.

I set it up like this

3 x .6 = 1.8

2 x .2 = .4

4 x .14 = .56

6 x .09 = .54
________3.3 x 27,000.00 = 8910.00 which is close enough for govment work.

The 8910.00 is a rating. Yes there is slop in the figures. This rating alone will not make a flat bet profit. Years ago I used it as a class indicator. The Kelco Class Calculator works in a similar fashion(I think).

Vaya Con Dios

Errol

Ok, but 3.3 x 27000 is 89,100.00, not 8910.00.

[Errol]

I needed to move the decimal point one to the left OY Vey

[jerry-g]

I think what you are trying to do is put a number on class which is a good
approach. However, a word of caution is to not use this as a standalone.
The late Dick Mitchell attempted to put a number on class centered around
class-consistency. He said the number is not very predictive when taken
by itself. A study done by his students showed the horses having the top
five class consistency ratings contained the winner 90% of the time.

He used just three factors, Earnings per start, Win percentage and In-the-
money percentage. Then he weighted them 40-30-30 and then scaled the
results so that the rating would be between 60 and 100 per horse. An Excell
or similar spreadsheet would have the following:

Horse #Starts #Wins #Place #Show Earnings E/S W% In$% CR

Minimum * * *
Maximum * * *

The min and max would be for columns E/S, W% and In$% respectively.

He used the last six races to calculate # of W,P,S and the current year
for earnings unless less than 3 races in which case he would add this
years and last years numbers.

A spreadsheet program makes the calculations and saves time. I would like
to quote the late Dick Mitchell on his definition of class.

"The ability to evaluate the true class of a thoroughbred racehorse is one
of the most difficult skills of handicapping. The reason this is so hard is
because class is a qualitative concept. Class refers to brilliance, determination,
courage, competitiveness and the willingness to prevail."

[headhawg]

The Handifast program has a version of Mitchell's class-consistency numbers, plus a modified pace-speed and a Power number. Those numbers did really well yesterday at GP. IIRC, it had the first eight races on top including Very Playful in the 7th @$34. I'm just mentioning this in case people don't want to or can't set up an Excel worksheet to do the calculations.

[jerry-g]

Originally Posted by headhawg

The Handifast program has a version of Mitchell's class-consistency numbers, plus a modified pace-speed and a Power number. Those numbers did really well yesterday at GP. IIRC, it had the first eight races on top including Very Playful in the 7th @$34. I'm just mentioning this in case people don't want to or can't set up an Excel worksheet to do the calculations.

Along those same lines I would like to offer to upload to the forum my Excel
spreadsheet template that does all the math for you. I made up fictitious
names for the horses and I warn that it is not to be used to predict the
winner but use it as another tool to help isolate contenders. No doubt it will
pick winners but make sure you have a good strike rate with it. If anyone is
interested in taking a gander at it then let me know and I will post it.

[raybo]

Not to put a damper on this approach, but considering the average field sizes today, wouldn't almost any decent contender selection method have the winners in their top 5 rankings, about 90% of the time, or am I once again missing something?

Narrowing down those contenders, and assigning value to them, has always been the hard part, hasn't it? By combining class, pace-speed, and power aren't we basically doing what the public does? Handi might be doing better as he is using proprietary pace-speed and power numbers. So his would be "different" than the combined public, and "different" is where the value comes from. I do know that in the short "Challenge" we had here on PA a few years ago (6 tracks in 6 days I believe, top pick only to win), his program finished 1st, in the Black Box section, with a 25-26% ROI (not sure of his exact ROI but I was 2nd with 24% ROI). And, I'm not sure his "class consistency" method is what he was using back then, and might even be better now.

[raybo]

Originally Posted by jerry-g

Along those same lines I would like to offer to upload to the forum my Excel
spreadsheet template that does all the math for you. I made up fictitious
names for the horses and I warn that it is not to be used to predict the
winner but use it as another tool to help isolate contenders. No doubt it will
pick winners but make sure you have a good strike rate with it. If anyone is
interested in taking a gander at it then let me know and I will post it.

Jerry, Post it or send it to me and I'll put it in my Excel black box track testing program and run some tests and post them here.

[jerry-g]

Yes Raybo I would have to agree with you. However, I think Mitchell was
trying to be ultra conservative and not mislead anyone into thinking this
is a standalone way to pick winners. I believe the crowd hardly ever goes
beyond the last race and with this method you take the last six for the
computations. If you find it is picking winner in the top 3 or 4 then I think
this is good information to have. I have never used the Handyfast program
so I can't comment on it but I definitely would take Hogs recommendation
that it is good. I am going to go ahead and upload my template for those
that may want to check it out and customize it any way they wish.

[raybo]

Originally Posted by jerry-g

Yes Raybo I would have to agree with you. However, I think Mitchell was
trying to be ultra conservative and not mislead anyone into thinking this
is a standalone way to pick winners. I believe the crowd hardly ever goes
beyond the last race and with this method you take the last six for the
computations. If you find it is picking winner in the top 3 or 4 then I think
this is good information to have. I have never used the Handyfast program
so I can't comment on it but I definitely would take Hogs recommendation
that it is good. I am going to go ahead and upload my template for those
that may want to check it out and customize it any way they wish.

Jerry,

Which column are you using for the contender rankings, "CCR" or "Overall" (they're close to being the same but aren't exactly, due to ties in the "Overall")? I'm assuming the "CCR" as they are unlikely to produce as many ties in the rankings.

[classhandicapper]

"The ability to evaluate the true class of a thoroughbred racehorse is one
of the most difficult skills of handicapping. The reason this is so hard is
because class is a qualitative concept. Class refers to brilliance, determination,
courage, competitiveness and the willingness to prevail."

I would add "stamina" or "reserve racing energy" to that list.

The major problem with all these "earnings' based formulas is that they don't equalize horses across tracks/circuits where the purses AND expenses might be lower (or higher) for the same quality of horse. The other is inflated Statebred purses and the possibility that purses can be inflated at the casino tracks.

I've been struggling with these cross track/circuit classification issues for decades. There is no easy solution. That's why everyone becomes a speed handicapper.

Right now I am trying to build a track to track purse adjustment, but it's not easy. If I knew the cost structure for training, vets, etc... at every track I could probably build a nifty "theoretical" model. But to my knowledge the data does not exist publicly.

The best I've been able to do so far is gather PAR information for multiple tracks. It's tough to get PARs for every class these days (there are so many classes and small sample sizes), but they definitely help. Ideally, I'd also like to be able to track shippers to and from every track, but even that has some issues because shipping is sometimes tied to trainer intent.