Bayes Theorem and E/W Betting

[cavello]

[muppet77]

Re: Bayes Theorem and E/W Betting looks interesting mate. how did you use bayes to come up with all this?

[cavello]

Re: Bayes Theorem and E/W Betting If we take a 3 horse race with the odds of Evens, 3/1,3/1(100% book). We have probabilities of winning of 0.5, 0.25 and 0.25. To the asess the chance of say the favourite coming 2nd we have 50/75 (odds of placing 2nd to winner of 3/1)x 0.25(chance of one 3/1 winning). We then repeat this for each combination. In this case 2 which equals=0.33 or 33%.. We then know that the favourite chance of placing 1st ot 2nd is 83.33%. I have just written coding to cover races up to 20 runners and the first 3 places and must now extend this analysis to 16-20 runner handicaps with 4 places. The slapdash method.

[cavello]

Re: Bayes Theorem and E/W Betting I have just run a database to check the returns of backing odds-on favourites E/W for 1112 races. -0.05 lose win only. -0.04 E/W. Therefore it would seem that backing odds on E/W is not necessarily crazy particularly if included in E/W doubles. Small profit on 2nd and 3rd favourites backing all E/W.