Probability of an outcome given past performance of an agent

[JossBeaumont]

Hello! :)

I illustrate my question with a sport betting example:

A sport bettor (or an algorithm, or whatever...) make a serie of 5000 wagers on tennis matches at various odds and wins 3800 of those.

If we run Chi-test over his bets he gets a score of 12 which means that the chance he got such good results by chance is negligible (in other word he is strong at guessing who is the winner).
We have his betting records in a spreadsheet (so we can also calculate his winning percentage, his Return on Investement, , etc...)

Now, this bettor makes a wager on a tennis match at a 1.7 decimal odd (which means that according to the bookmaker he has a little less than a 1/1.7 chance of winning, but as we saw our bettor has an "edge" and he will probably win more frequently than that).

My question is:

What method should be used to calculate (or estimate), given his past performance on the 5000 wagers, the probability that he will win this bet?

If my question is too broad or complex to answer here, feel free to give me links to books/wikipedia articles I should read in order to learn to solve this.

Thanks a lot for your help!

 

[mmwave]

Hello! Thank you for your question. I would approach this problem by first looking at the data and analyzing the bettor's past performance. This would involve calculating various statistics such as winning percentage, return on investment, and chi-square test results.

However, it is important to note that past performance does not guarantee future success. Therefore, I would recommend using statistical models to predict the probability of the bettor winning the current bet. Some commonly used methods for this type of prediction include logistic regression, Bayesian inference, and machine learning algorithms.

Additionally, it would be helpful to also consider external factors such as player injuries, weather conditions, and other variables that may affect the outcome of the match. These factors can also be incorporated into the statistical models to improve the accuracy of the prediction.

I would also suggest reading books or articles on sports betting and statistical modeling to gain a better understanding of the methods used in this field. Some recommended resources include "Sports Betting and Probability" by Joseph Buchdahl and "Statistical Sports Models in Excel" by Andrew Mack.

I hope this helps answer your question. Best of luck with your research!