Bivariate Poisson

tried it it don't work its just gambling at the end of the day

The Poisson is a much better model than most people believe. The reason so many try to apply it and fail is they have no idea how to adjust for heterogeneity in the data. (all data is heterogeneous to some extent).
Depending on which sample set chosen there is a small and sometimes significant correlation coefficient between numbers of goals scored. However there is a massively significant negative correlation between the underlying rates at which goals are scored by the two teams and here is where the Poisson model falls down at least to some extent. Having said that I have never seen this discussed on here and I have no intention of enlightening the forum further

Luckily we don’t need to be fully enlightened to the concept, the formula or how to calculate it because Microsoft Excel has a formula which can work out Poisson automatically. All we really need to know is that it can be used to calculate the probability of outcomes for a football match, which in turn can be turned into odds which we can use to identify value in the market. This covers a number of goal based markets such as Match Outcome (1×2), Correct Score, Over / Under Match Goals, Both Teams To Score and Asian Handicap. Although it has its limitations and faults, Poisson is a useful starting point to understand the fundamentals of creating your own odds. It can work as a standalone model which you use to advise your betting, or it can be used to understand the basics before going on to explore further, more complicated methods.

What you need to do is take historical results to calculate the number of goals teams score and concede. These averages are compared to the league average and used to create values for attacking strength and defensive strength for every team, which are then turned into goal expectation figures. This metric is put into a Poisson Distribution formula which works out the probability of every result when two teams face each other. We then take these probabilities to create our own odds, compare these against the bookies’ odds, then identify where there is value in the market because the bookies are offering more generous odds that we’d expect. Simple!

So this is the stuff I would have learned had I gone to the Uni lectures and not been down the bookies every afternoon! GL - sounds fascinationg. Poisson ended for me at being a skewed kind of Bell curve.

Not so simple Gusher.

Poisson will always under estimate the chances of a draw.

I'm with JML here, Gusher isn't anywhere near the mark and not a solitary indication of how he is going to deal with heterogeneity in the data

I think i I am going to use the Dixon and Coles model to begin with and take it from there. I think from my reading that this is a strong starting point and addresses a lot of the issues with standard poison.

It is still a standard poisson, but with maximum likelihood estimation of parameters. It goes a long way to addressing heterogeniety though. You might be able to do this in R otherwise you'll be doing a lot of programming

It is still a standard poisson, but with maximum likelihood estimation of parameters.

Well, not really. 'Standard' poisson only has 1 parameter. D&C have introduced other parameters to render it non-standard.

It does fix the issue over the under value of certain scores and the draw doesn't it?

I'd agree with you Contrarian if the poisson's were not independent, but in that model they are independent.That model will give a poisson lambda(home) for home goals, a poisson lambda(away) for away goals and a poisson lambda(home) + lambda(away) for total goals. The reason there are multiple parameters to estimate is that the rate lambda are a product of other parameters - a multiplicative model, giving one final parameter for each poisson process

I use poisson, of course in combine with "other stuff". I've made this calculator, so you can try it and if you find it useful I'll be glad.. If you have any questions about using it, or the math behind it - just ask :)
good luck
Link: advancedbetcalc.com/Downloads.aspx

Dr Alun Owen has written on the subject in the Smartersig magazine and can be found at http://www.smartersig.com/showarts.php

I've updated "poisson" to a "bivariate poisson" calculator

Cheers