Here is a grid of the odds for all score combinations up to 9 apiece...
x:0 x:1 x:2 x:3 x:4 x:5 x:6 x:7 x:8 x:9
0:y 19/1 8/1 10/1 13/1 32/1 74/1 207/1 768/1 2499/1 9999/1
1:y 17/1 9/1 9/1 15/1 28/1 70/1 226/1 832/1 1999/1 9999/1
2:y 42/1 19/1 19/1 31/1 68/1 146/1 285/1 1666/1 - 9999/1
3:y 151/1 59/1 61/1 92/1 166/1 369/1 1110/1 - 9999/1 -
4:y 344/1 262/1 178/1 434/1 999/1 1999/1 3332/1 4999/1 - -
5:y 2499/1 1110/1 908/1 1666/1 3332/1 4999/1 - - - -
6:y 4999/1 4999/1 - 9999/1 - - - - - -
7:y 9999/1 - - - - - - - - -
8:y - - - - - - - - - -
9:y - - - - - - - - - -
And here are some summary odds:
Home win: '44/10',
Away win: '1/2',
Score draw: '5/1',
No score draw '19/1'
These are based on 10,000 modeled games and 5000 particles per team. Interested to know if there are other odds people would be interested in.
I compared with Bet365 odds, which are generally fairly similar. My numbers seem to like the idea of a home win slightly more than Bet365. 1 nil to Everton looks like a value bet (although odds are long).
Health warning: I am still validating this model, although I believe the approach is generally solid!
*** Update*** the game finished 0-2 (win for Man City). It stood at 0-1 until injury time, which would have tallied with my most likely result
*** Update*** the game finished 0-2 (win for Man City). It stood at 0-1 until injury time, which would have tallied with my most likely result
No comments:
Post a Comment