Quote:
Originally Posted by akkopower1
From my experience I'm middling, I generally assume a uniform distribution for totals/scores/whatever's around the median.
For example in NBA the distribution for total game points is quite beautiful, a nice bell shaped curve. I don't think you lose too much information by assuming that points within say 0.1 to 0.2 standard deviations from the mean are all equally likely and thus assume a uniform distribution.
For middle bets that are taken from standard bookies lines (I.e. Prices are close to $2.) the middle you are hoping to hit should be the median and modal score.
for sixes in 2020 and odi, where the two teams are reasonably close to even $$ I generally assume the chance of say 12 6's occurring is 1/12 abd 13 6's 1/13 and so forth.
If one team is a decent favourite (Aus v Sri this Friday) then I will usually decreases those probabilities slightly. Because typically the team batting second won't score too many more sixes than the team batting first.
A certain bookie in Australia used to leave the boundary markets up post coin toss. That was fun!!! Weak team bats first take unders on boundary and vice versa.
I do cricket betting as well, but mainly 2 way prop betting. How is your model working out so far on sixes?