how to calculate true o/u

[elbciho]

I am trying to figure out the true over/under of a soccer match. For example this game on pinny is over 2.5 2.120, under 2.5 1.806. Assuming this is the correct line the true o/u is somewhere under 2.5. I don't know the formula to find the exact o/u and would like to know what it is.

Thank You

[Pokerjoe]

I used to have a conversion formula, but I just looked and can't find it (I've changed computers a few times since developing it) You should do you're own work anyway, in case I or anyone else giving you a formula is wrong. You're talking about converting a moneyline to a number right? So that if you estimate a total of 2.32, you can compare it to a moneyline to find value? It's doable, and not really that tricky, but sorry to say I don't have it here.

So this has been a colossally useless reply, lol. Sorry. But the fun's in the finding anyway.

[subs]

i'm not a maths guy so i could be really wrong, just trying to stimulate discussion.

i believe goals scored in football is poisson, right? maybe this is misleading, sorry if it is. but i think the poisson formula might help u.

if u find the no-vig implied probability 1st.

in the example above

1/2.12=0.47

1/1.802=0.55

no vig =0.47/0.47+0.55=0.46

then use the poisson formula here

i guess P < 3 = 0.54
x = 0, 1, 2
e = 2.71828
μ is ur answer

P(x; μ) = (e^-μ) (μ^x) / x!

so u need 0.54=[(2.71828^-μ) (μ^0) / 0!]+[(2.71828-μ) (μ^1!) / 1]+[(2.71828^-μ) (μ^2) / 2!]

i prolly messed it up but it was fun

what i have found here is if u post something stupid (prolly like my post above) some1 will troll it and then some1 even smarter will maybe help u out.

bring on the trolls...

[elbciho]

Thanks for the help. I am getting in to this and I don't know what is widely available public and what isn't. For now I just wan't to know how to find the number is for the purposes of live betting. It should be a decimal and shouldn't be too hard to figure out.

[818z_finest]

great info

[usernametaken]

When a team scores 9 runs and 10 runs and 8 runs and goes over game after game, I think it is a good idea to go under after 5 times because of probability of fewer runs and line being adjusted upwards

[subs]

does any1 know if what i wrote above is as retarded as it looks?

Thanks for the help. I am getting in to this and I don't know what is widely available public and what isn't. For now I just wan't to know how to find the number is for the purposes of live betting. It should be a decimal and shouldn't be too hard to figure out.

or just use trial and error with the poisson calculator in the tools.

or just use a simple ratio.

[simona]

I do not think that there is a formula for that.

[Duff85]

As Subs said poisson should do it.

I'm horrible at the numbers but use this thing.

Not sure how useful this will be:

http://annabet.com/en/poissoned/

I use it for moneylines to asian handicaps... but it does also suggest totals prices. I haven't tested the accuracy of it, so you may want to play with it yourself before diving right in.

[subs]

good resource Duff85

apparently, simona is right there is no closed form solution, but it is possible to find the answer numerically or graphically. does any1 know how to do this in excel?

apparently u can tabulate or graph

is all this worth it, or will just using a simple ratio get us close enough. don't know the answer but i guess with something like 0.95/0.05 is going to be less accurate than @ 0.46/0.54. my ignorant guess is that doing it simple is not going to hurt too much in this case.

PokerJoe did u actually get a usable formula or did u do something cool with excel?

good luck

[subs]

i didn't realise just how badly poisson sux until now... or more likely my understanding of poisson sux.

the answer from the poisson formula is 2.515 which is obviously full of shit. the answer from a simple ratio is 2.4. and from trial and error on the poisson calculator 2.435.

closest answer:

sub's poisson formula +50000
ratio +5000
granchrow's poisson tool no price

[CHUBNUT]

for premiership just use 2.71 as your base. If you make the game total 2.51 then the no vig under is -120 Get rid of all that mathematical bullshit. Waiting for a Chi square to surface in this thread, square is definitely the right word. why do people always try to make sports betting look like its quantum physics.

[goucla]

nice site gonna save me some time

[hitman09]

i'm not a maths guy so i could be really wrong, just trying to stimulate discussion.

i believe goals scored in football is poisson, right? maybe this is misleading, sorry if it is. but i think the poisson formula might help u.

if u find the no-vig implied probability 1st.

in the example above

1/2.12=0.47

1/1.802=0.55

no vig =0.47/0.47+0.55=0.46

then use the poisson formula here

i guess P < 3 = 0.54
x = 0, 1, 2
e = 2.71828
μ is ur answer

P(x; μ) = (e^-μ) (μ^x) / x!

so u need 0.54=[(2.71828^-μ) (μ^0) / 0!]+[(2.71828-μ) (μ^1!) / 1]+[(2.71828^-μ) (μ^2) / 2!]

i prolly messed it up but it was fun

what i have found here is if u post something stupid (prolly like my post above) some1 will troll it and then some1 even smarter will maybe help u out.

bring on the trolls...

I don't understand what are you talking about 90% don't play O/U, only play in the last couple of reg weeks. They are pretty much go O, and the early weeks go U.

[subs]

for premiership just use 2.71 as your base. If you make the game total 2.51 then the no vig under is -120 Get rid of all that mathematical bullshit. Waiting for a Chi square to surface in this thread, square is definitely the right word. why do people always try to make sports betting look like its quantum physics.

well PokerJoe kinda said there may be a formula, so we tried to find it but looks harder than 1st thought.

can i ask u a question: r u not embarrassed for stiffing a poster here for points? don't get me wrong, what u post is sometimes good but i can't take u too seriously.

why don't u just pay him back? u would know better than me but gamblers don't really like stiffs, right?

[subs]

I don't understand what are you talking about

that makes 2 of us

90% don't play O/U, only play in the last couple of reg weeks

.

They are pretty much go O, and the early weeks go U

do u have any numbers? would like to see them... much appreciated.

[Pokerjoe]

good resource Duff85

apparently, simona is right there is no closed form solution, but it is possible to find the answer numerically or graphically. does any1 know how to do this in excel?

apparently u can tabulate or graph

is all this worth it, or will just using a simple ratio get us close enough. don't know the answer but i guess with something like 0.95/0.05 is going to be less accurate than @ 0.46/0.54. my ignorant guess is that doing it simple is not going to hurt too much in this case.

PokerJoe did u actually get a usable formula or did u do something cool with excel?

good luck

Yes, I got a good formula, but it was derived from database crunching (footballdata.com, I think was the site for building a soccer DB), on excel, combined with some work with Pinny drop down windows (this was long before Pinny left the US market).

I do remember concluding that poisson wasn't optimal.

I didn't pursue soccer totals further because I was never satisfied with my ability to handle the game line's effect on the total. I mean, if you have an EPL team scoring and allowing 1.25 goals per game, and they're playing a League 2 side that's also scoring and allowing 1.25 goals per game, obviously you're not going to have a game with an expectation of 1.25 goals per side. Or, put it this way (and this is how I was putting it) if you have two teams, each with a certain general scoring range, their total will vary as the handicap does. So a game between evenly matched teams will be lower scoring than one in a game with a 1.5 goal handicap. But I was never happy with my way of accounting for that, and so gave up the ghost on soccer totals.

But I do remember being happy enough with my ability to translate score estimates into o/u lines, that wasn't the stumbling block. I don't recall that as being that tricky. But I probably also did it with brute force, not something particularly elegant.

[subs]

thanks for the explanation Joe.

[bettorjon]

90% don't play O/U, only play in the last couple of reg weeks. They are pretty much go O, and the early weeks go U.

oh yeah, says who?

if youre not playing o/u 2.5 then youre missing alot of money is soccer games. ML and ATS are much harder to beat imo?

btw, what league are you talking in regard to your last statement? coz as far as i know, it doesnt happen that way.