OZmium Sports Betting and Horse Racing Forums

OZmium Sports Betting and Horse Racing Forums (http://forums.ozmium.com.au/index.php)
-   Horse Racing (http://forums.ozmium.com.au/forumdisplay.php?f=5)
-   -   Trifectas (http://forums.ozmium.com.au/showthread.php?t=14957)

Punter4211 9th November 2006 09:33 PM

Trifectas
 
Hi everyone, I have a problem.

Im looking for a formula that estimates the trifecta payout based on the tab prices of the runners.. For example three runners, A at $2.50, B at $2.70 and C at $12.00.. You can say that by multiplying the prices together as A*B*C would give you a reasonable estimate, but the combination ABC will give you the same answer as CBA which would logically pay a greater dividend.. Im trying to work out how much more I should put on ABC as opposed to BCA and CBA etc. to return basically the same return.

Simply multiplying them together is not the right answer.. I can access the trifecta pool and the Win and Place Pool so I can work out how many "live" tickets there are on win or place bet, if that helps..

$20.00 prize money to the first respondent to give me a correct answer shown to be accurate to within 10% of the real dividend based on 10 randomly chosen Saturday races...Or a logically calculated formula that I can massage to get the results I want..

Respond to this if you think you have the right answer and Ill get in touch..

ubetido 9th November 2006 10:22 PM

Hi the formula is as follows:

X[(XY-1)(Z+1)
X+1

3[(3*4)-1(9+1)]
3+1

3*11*10
4


So to your dividends
2.5 2.7 12 expected dividend $53


Cheers

brave chief 9th November 2006 10:23 PM

I was working on this one, but gave up in frustration.

the formula i used:

(A x B x C) / (1 - A) / (1 - A - B)

where,
A = winning chance % of 1st
B = winning chance % of 2nd
C = winning chance % of 3rd

After doing the trifecta tables in Excel the file was up to about 39mb in size, which made it almost unusable, prone to freezing etc.

It may be more practical to base your bet size just on the exacta combo's with muliple runners to run 3rd, or some other variation.

Shaun 9th November 2006 11:01 PM

The formula Ubetido put in is correct i have used it before works a treat, divis or normaly higher than the rated prices i will give an exaample from a race today.

The Oaks

Win div of the first 3 places from Unitab

1st $2.50 =a
2nd $2.70 =b
3rd $11.60 =c

take 1 from each price
1.5
1.7
10.6

(1.5x((1.5x1.7)-1)xC)/a)

Payout is $10.79

Trifecta payed $34.10

Punter4211 10th November 2006 07:10 PM

Trifectas
 
Quote:
Originally Posted by ubetido
Hi the formula is as follows:

X[(XY-1)(Z+1)
X+1

3[(3*4)-1(9+1)]
3+1

3*11*10
4


So to your dividends
2.5 2.7 12 expected dividend $53


Cheers
Thanks ubetido,

I'm not sure that I understand the formula so I'll have to play with it till I fathom it out.. It looks good.. Not sure how you derived this formula, but it looks interesting.

Thanks

OzPunter

Punter4211 10th November 2006 07:12 PM

Quote:
Originally Posted by Shaun
The formula Ubetido put in is correct i have used it before works a treat, divis or normaly higher than the rated prices i will give an exaample from a race today.

The Oaks

Win div of the first 3 places from Unitab

1st $2.50 =a
2nd $2.70 =b
3rd $11.60 =c

take 1 from each price
1.5
1.7
10.6

(1.5x((1.5x1.7)-1)xC)/a)

Payout is $10.79

Trifecta payed $34.10
Thanks Shaun,

I'm looking into this and I'll let you know.

OzPunter

Punter4211 10th November 2006 07:15 PM

Quote:
Originally Posted by brave chief
I was working on this one, but gave up in frustration.

the formula i used:

(A x B x C) / (1 - A) / (1 - A - B)

where,
A = winning chance % of 1st
B = winning chance % of 2nd
C = winning chance % of 3rd

After doing the trifecta tables in Excel the file was up to about 39mb in size, which made it almost unusable, prone to freezing etc.

It may be more practical to base your bet size just on the exacta combo's with muliple runners to run 3rd, or some other variation.
Thanks brave chief

I also thought it would be easier that it turned out to be, so I asked you guys..

Hopefully I'll hit on the right formula and I'll share it around when I do..

Regards

OzPunter

jfc 11th November 2006 04:26 AM

Quote:
Originally Posted by OzPunter
Thanks brave chief

I also thought it would be easier that it turned out to be, so I asked you guys..

Hopefully I'll hit on the right formula and I'll share it around when I do..

Regards

OzPunter


This is the correct solution:

A: 2.5
B: 2.7
C: 12
D: =0.8/(E1*F1*G1/(1-E1)/(1-E1-F1)) expected Div
E: =0.84/A1
F: =0.84/B1
G: =0.84/C1


It matches Brave Chief's formula (oft-repeated according to the search function).

It assumes unbiased distributions, Win returns of 84% and Trifecta returns of 80%.

Punter4211 11th November 2006 05:50 AM

Quote:
Originally Posted by jfc
This is the correct solution:

A: 2.5
B: 2.7
C: 12
D: =0.8/(E1*F1*G1/(1-E1)/(1-E1-F1)) expected Div
E: =0.84/A1
F: =0.84/B1
G: =0.84/C1


It matches Brave Chief's formula (oft-repeated according to the search function).

It assumes unbiased distributions, Win returns of 84% and Trifecta returns of 80%.
Thanks JFC, I'll be watching it closely.

Kind Regards
OzPunter

Shaun 11th November 2006 08:55 PM

I plan on running this side by side with the formula i use, i am very happy with the results i get.

Punter4211 12th November 2006 06:52 PM

Quote:
Originally Posted by Shaun
I plan on running this side by side with the formula i use, i am very happy with the results i get.
Hi,
Would you believe I've had so many things on, I havent had a chance to look at them as yet.. But I will do it during the week.

It's a shame there are only two days in a weekend.

Regards
Ozpunter

Punter4211 20th November 2006 06:54 PM

Quote:
Originally Posted by ubetido
Hi the formula is as follows:

X[(XY-1)(Z+1)
X+1

3[(3*4)-1(9+1)]
3+1

3*11*10
4


So to your dividends
2.5 2.7 12 expected dividend $53


Cheers
Hi ubetido,
Thanks for contributing to my request, but unfortunately I couldn't fathom your formula. When I did it on a spread sheet as I interpreted it I got an answer vastly different to the example you've given. Perhaps, if you have time you could walk me through it... The test dividends are A=2.5, B=2.7 & C=12 (dollars of course)

Thanks
Oz Punter

Punter4211 20th November 2006 06:59 PM

Quote:
Originally Posted by Shaun
The formula Ubetido put in is correct i have used it before works a treat, divis or normaly higher than the rated prices i will give an exaample from a race today.

The Oaks

Win div of the first 3 places from Unitab

1st $2.50 =a
2nd $2.70 =b
3rd $11.60 =c

take 1 from each price
1.5
1.7
10.6

(1.5x((1.5x1.7)-1)xC)/a)

Payout is $10.79

Trifecta payed $34.10
Hi Shaun,
Thanks for contributing to my request for a trifecta formula. Your explanation seems to be fairly easy to follow but I still couldn't get your $10.79 example result when I tried the formula out. Perhaps if you have time you could walk me through an example.

Kind Regards
OzPunter

Punter4211 20th November 2006 07:03 PM

Quote:
Originally Posted by jfc
This is the correct solution:

A: 2.5
B: 2.7
C: 12
D: =0.8/(E1*F1*G1/(1-E1)/(1-E1-F1)) expected Div
E: =0.84/A1
F: =0.84/B1
G: =0.84/C1


It matches Brave Chief's formula (oft-repeated according to the search function).

It assumes unbiased distributions, Win returns of 84% and Trifecta returns of 80%.
Hi JFC,
Thanks for contributing to my request for a solution to the Trifecta problem, unfortunately I could not get a sensible result with the example given. Perhaps if you have time you could walk me through the formula to see how you came up with the result.

Thanks
OzPunter

ubetido 20th November 2006 07:55 PM

Hi Oz Punter

On your spreadsheet in cell A2 type in the following:

=B2*(B2*C2-1)*(D2+1)/(B2+1)

If you type 5 in cell B2 and C2 and D2 it should give you 120

Cheers
ubetido


All times are GMT +10. The time now is 03:31 AM.

Powered by: vBulletin Version 3.0.3
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.