We’re pleased to announce version 2.0 of our betting algorithm. This shows significant improvements in the key performance statistics used to measure the quality of our betting system.

At the same time we’re introducing a one number descriptive rating statistic for our betting system to enable comparisons over time and across leagues.

Value betting
Our previous algorithm backed or layed teams based on whether the market odds were over or under priced. However the amount staked was not dependent on how mis-priced the market odds were. Our latest algorithm bets more when odds are mis-priced by a larger amount and less when odds are close to our own internally generated odds. Continued

With inspiration from this weekend’s Grand Prix I’ve run another Monte Carlo simulation.

“Monte Carlo” is the name given to simulations which make use of computer generated random numbers to identify the range of possible outputs a model may generate in the ‘real world’.  Each random number generates an input from a user defined probability distribution which is run through the user’s model to produce a simulated outcome on each iteration of the model.  Run thousands of iterations and you generate a probability distribution for the output of your model. Continued