arb function

[Qbawler311]

I'm looking to find a function that calculates the ROI of an arb. I found a function that claimed to calculate this, it is: (T - L)/(((L*(T+2))+1) T=take, L=lay, but this functions output differs (although close) with the output of www.scalpulator.com/ , when using the same odds. What is the correct function of an ROI given decimal odds?

[Pythuz]

ROI = (y - x)(1 + x(2 + y)) in american odds of -x and y and where x and y are in terms of unit bets.

See pythuz.wordpress.com for the derivation.

Edit: which of course is the same formula you posted [img]/images/graemlins/wink.gif[/img]

[BirdDog]

I think that formula is incorrect for either american or decimal odds. Did you two get it from the same source? I also have no idea what "in terms of unit bets" means.

Using American odds like pythuz says of -110 and +120 you get x=110 and y=120.

(120 - 110)(1 + 110(2 + 120) = an insanely large number

Likewise, the formula returns incorrect results using decimal odds.

[ImBen]

"I want to make 'easy money' as an arbitrageur, but lack the ability to perform 7th grade algebra. Can someone please hold my hand?"

[BirdDog]

Ben:
I'm not sure I understand the purpose or target of your deriding post.

Qbawler311:
I made this forumula that gives correct results using decimal odds:

ROI = ((x + y) / (2 + x/y + y/x)) - 1

BTW, I created the scalpulator, and I may post all the calculations at the website if I can find some time.

[Pythuz]

Right, sorry, I missed a slash in the above formula, it should be

(x - y) / (1 + x(2 + y))

Also, by per unit bet I mean that in the standard situation of, say -110 american odds, this is for a bet size of 100, so if you are in the -110 and +120 situation then x = 1.1 and y = 1.2, in which case

(1.2 - 1.1) / (1 + 1.1(2 + 1.2)) = approx 2.2%.

As for where I got it from, I derived it myself and the derivation is on my blog, which is listed above.

[SamIAm]

[ QUOTE ]
Ben:
I'm not sure I understand the purpose or target of your deriding post.

[/ QUOTE ]I assume he meant it to be targeted at the OP. (He just used QuickReply.)

I have to admit, I agree with his sentiment. The arithmetic involved isn't complicated. I'm not saying you should be able to do it in your head, but it should be as easy as asking a forum and waiting a few hours for them to do it for you.
-Sam

[Qbawler311]

[ QUOTE ]
Ben:
I'm not sure I understand the purpose or target of your deriding post.

Qbawler311:
I made this formula that gives correct results using decimal odds:

ROI = ((x + y) / (2 + x/y + y/x)) - 1

BTW, I created the scalpulator, and I may post all the calculations at the website if I can find some time.

[/ QUOTE ]

Very nice Birddog and Cheers on Scalpulator.

Pythuz my mistake was using your equation with decimal numbers. The one that I originally posted with the quotient.

Thank you both, I understand that it would have been easier just to insult my intelligence

[Qbawler311]

[ QUOTE ]
ROI = ((x + y) / (2 + x/y + y/x)) - 1

[/ QUOTE ]

What about if you want to get the ROI for win, loss, and ties?

How does the above function change?

[ImBen]

Post deleted by Performify