One can use past data to determine whether published odds reflect unbiased expectations. It is similar to fundamental investing. Who has ever tried to beat the odds thanks to data analysis?

can you explain better what **types** of STATISTICAL frameworks you actually want to leverage Jad Bouez

In most sports, a game ends with a win, a draw or a loss. Before a given game, one can compile past data by picking similar games, whose results are known. If, in this sample, w% are wins, d% a draws and l% are losses, then the probability of a win in the future game is w/100 (or w%).

The question is: what to include in this sample? All historical confrontations between both teams?

The issue is similar to the one found when conducting a relative valuation of a stock. Which comparables should we use?

Anand Krrish great to see you joined. Jad Bouez welcome Anand to the project, where’s your manners? 😃 haha

Jad Bouez – oooo ok now i get it. You were making it sound crazier than i thought. Now I see. ok so… you’re talking now about our **first step** which is some kind of *understanding what our data set should look like*

–> I think we should contact someone at espn or someone with access to espn sports archives to figure out how to get this data?

https://docs.google.com/document/d/1DhxEEypoPshV5NjzBmGYw9ozgKJK84KzazsOjuPAWIQ/edit

*doing some in depth analysis here*

Using Past Data to compute probabilities, which you can convert into IMPLIED odds, that you compare with the bookie’s odds.

Past data is a team’s past records again team Y for the past 5 years, for instance, or against all teams, or only when playing home, etc… The idea is to find similar past games to the one you want to bet on and average the results in order to find an IMPLIED PROBABILITY of win or loss.