One of the tricky things with sports betting is that we underestimate or overestimate the probabilities of something happening based on our feelings. Statistics helps to take the subjectivity out of it and ground our estimation in facts.
Bayesian probability takes that one step further by imposing a condition. In today's post, I want to look at the probability of popular Total Goals offered by bookies given a half-time score of 0-0. I'll be looking at Championship data from England.
I downloaded the data from football-data.co.uk. After some processing in Excel, here are the results both on an numbers and percentage basis.
Full Time Total Goal Results (Count)
Full Time Total Goal Results (%)
Thoughts on the data
Unfortunately, I don't have the timing of goals down to the minute level without scraping. But based from this preliminary analysis, betting on the above 0.5 goals would be profitable at odds greater than 1.4.
Greater than 1.5 goals would need odds between 2.5 to 3 and more than 2.5 goals needs 8.33 odds given the half time score.
Personally, I think the over 0.5 goals starts to look interesting around the 60th minute if you can get odds over 1.6 around that time.
Hope you found this post interesting and if there's interest, i'll expand the analysis into more leagues and probably use Python instead. Excel's fine for one time analysis but gets tedious if you have to do it multiple times.
