New Posts For Thread: Bivariate Poissonhttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poissonDoes anyone on here use such a model for soccer betting?I have been reading the coles and Dixon Modelling Association Football Scores and Inefficiencies in the Football. Betting Market and was wondering about looking into doing something similar. BuI've updated "poisson" to a "bivariate poisson" calculatorCheershttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=543317941#543317941I've updated "poisson" to a "bivariate poisson" calculatorCheersThu, 29 Jan 2015 00:18:06 -0600Dr Alun Owen has written on the subject in the Smartersig magazine and can be found at http://www.smartersig.com/showarts.phphttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542887301#542887301Dr Alun Owen has written on the subject in the Smartersig magazine and can be found at http://www.smartersig.com/showarts.phpFri, 02 Jan 2015 09:44:44 -0600I 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 luckLink: advancedbetcalc.https://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542722857#542722857I 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 luckLink: advancedbetcalc.Fri, 19 Dec 2014 10:29:48 -0600I'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) fohttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542656801#542656801I'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) foSat, 13 Dec 2014 15:34:12 -0600It does fix the issue over the under value of certain scores and the draw doesn't it?https://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542651603#542651603It does fix the issue over the under value of certain scores and the draw doesn't it?Sat, 13 Dec 2014 11:20:27 -0600It 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.https://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542647025#542647025It 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.Sat, 13 Dec 2014 05:30:36 -0600It 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 programminghttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542633913#542633913It 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 programmingFri, 12 Dec 2014 04:17:11 -0600I 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.https://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542626797#542626797I 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.Thu, 11 Dec 2014 12:41:56 -0600I'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 datahttps://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542559037#542559037I'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 dataSat, 06 Dec 2014 15:02:43 -0600Not so simple Gusher.Poisson will always under estimate the chances of a draw.https://community.betfair.com/general_betting/go/thread/view/94082/30393853/bivariate-poisson?post_id=542530567#542530567Not so simple Gusher.Poisson will always under estimate the chances of a draw.Fri, 05 Dec 2014 03:18:08 -0600