Can Data Science Predict UFC Fights? Building a Leak-Free Model with Random Forest | by Raphael Schols | Data Science Collective | Mar, 2025

Avoiding a common loss in forecast models and making a leak free model.

Not a member? Click here to read in free

MAchin Learning is everywhere in sports. For example, the Golden State Warriors basketball team uses it to analyze players’ performance, shot selection, and strategy correction. Likewise, Liverpool Football Club Use it to enhance the recruitment strategy. In another objectionable line of business, bookmakers rely on machine learning models to create difficulties and identify fraud.

Machine Learning and AI have also made their way to mixed martial arts (MMA), in which models claim to predict the results of the UFC fighting. If you are a batting person, “Creating difficulties” has engulfed your eye, or if you are a data hobby you may be surprised:

Can a machine learning really predict UFC results?

Well, you will not be the first to ask this question and some claim that their models can predict UFC battles with 70 % to 80 % accuracy. But before you quit your job, start to calculate the values ​​expected, and bet to live – a problem.