Odds @ 100% margin method.

Here you may present your strategy. Comments and suggestions are welcomed.
Post Reply
Posts: 4
Joined: Sat Feb 24, 2018 1:45 pm

Odds @ 100% margin method.

Post by luigi84 » Mon Feb 26, 2018 8:16 pm

Hello Everybody and nice to meet you all.

I am Luigi and it is already some months I am using CGM bets and I am really grateful to Mihai for making it and giving so many interesting study opportunities on statistical betting approaches. Being Italian and based in Russia, my main league focus falls on Italy Serie A, Serie B, Russia Premier and Ukraine Premier.

As of now my current strategy is to calculate from current season results:

- expected goals from goals stats (Scored Home + Conceded Away / Conceded Home + Scored Away)
- expected goals from goals per shots stats (as above)
- expected goals from goals per shots in target stats (as above)

From these data, having both score expectancies, I calculate the table of correct scores and so I do have all projected probabilities (odds) for all markets I want. One thing I would like to stress, and maybe Mihai could take into account, is the draw reduction factor from Dixon and Coles to take into account interdepency between the two poisson scoring variables Home and Away, which infact are not independent each other. (http://www.math.ku.dk/~rolf/teaching/thesis/DixonColes.pdf)

This could help accuracy of extimations. Confronting previous scores we could have a league drf or even a team by team drf. It would be nice to have this calculation incorporated.

In my betting strategy currently I do check team news to have a clearer view of important missing players, I then do take the most restrictive odds coming from the 3 different extimations I do and if there is a margin with the odds offered by the market then I bet.

Anyway, I am not here to write about the betting strategy per se, I am more interested to hear your opinion in how this statistical process of extimations could be helped in order to make odds extimations as accurate as possible; on the match day it would be cool to calculate odds including also:

- missing players
- weather
- stats on ball possession (for example shots could be crossed with ball possession and hence goals crossed with shots)

In the end my goal is to have a reasonably reliable 100% margin pricing that would get no loss and no return on the long term with flat staking on any of the possible outcomes, governed only by goal expectancies and draw reduction factor.

If you think it could be interesting, I could share my excel file that works reading in the same folder data exported from CGM for the leagues I do follow. I am also very eager to hear your critics, thoughts and suggestions.

I worked as sports modeller for bookmakers and odds providers but on those models I worked with a different aim than making odds from statistics: I did make European odds both live and prematch recalculating from Asian lines.
This was an example of live betting odds confronted with bet365, both videos have to be seen simultaneously second by second, from prematch set up to live:
- https://www.youtube.com/watch?v=NDv59ki3bZ8 (Bet365 and SBOBET)
- https://www.youtube.com/watch?v=2WcGbzHQ99o (live algorithm)

Kind regards.


Posts: 5
Joined: Wed Feb 14, 2018 2:18 pm

Re: Odds @ 100% margin method.

Post by mattylee68 » Fri Mar 23, 2018 1:44 pm

Hi Luigi,

Sounds a very interesting project you've got going on there!

I'd really like to take a look at the spreasdheet you mention; can you share that with me please?


Post Reply