You may also want to start calculating a Brier score for yourself and the books. The basic calculation is the average of [(1-odds)2] for winners and [(0-odds)2] for losers.

Basically let’s say your odds say team A has 60% chance of winning and the book says 58%. If team A wins then (1-.6)2 = .16 which is your score. The books score would be (1-.58)2 = .176 However this really only holds weight if you have a lot of games to average together. The lower the score the better you are at forecasting winners.

In my example above the formula would be (1-.6)² if you win or (0-.6)² if you lose. Since it’s squared the negative doesn’t matter.

Here is a more drawn out explanation of the Brier scoring

While browsing the internet and looking for some new inspiration to build an own predictive model, I came upon a very interesting possible feature: the Brier score. The Brier score is a possibility…

beatthebookie.blog

...and even more analytical

https://arxiv.org/pdf/1908.08980.pdf