MMA-AI
Free AI predictions for UFC from a professional AI researcher
The Model
Cutting edge algorithms are used in an ensemble of machine learning models that are trained on a tabular dataset with thousands of historical fights. The final model ensemble is 69% accurate. The betting odds favorite wins only 66% of the time.
The Data
Data is continuously updated and fed into complex formulae to reveal the hidden statistical relationships between fighter styles, skills, and recent performance.
The Code
Thousands of hours and lines of Python were used to scrape, transform, and test the final output to maximize accurate predictive abilities. All code is unit tested to ensure no post-fight data leakage or overfitting occurs.
Where do you get the data?
Downloaded from various MMA stat sites. We start with 22 base stats like Strikes Landed and engineer those into 10,000 stats. Then we cut those down to about 10-20 of the most influential, predictive stats based on both manual expertise and statistical analysis.
What is the betting strategy you use?
Due to analysis by Chris from Wolftickets.ai, it was discovered that the most profitable strategy for machine learning sports prediction models is to include the odds in the dataset so the model will achieve accuracy better than the bookies then use 2 to 3 leg parlays based on the predictions.
Why is this free?
It may not be forever. This has been a 3 year endeavor to learn machine learning engineering and at this point I believe I am in possession of a world-class process and model.
How do you train and test the model?
For more details check the Predict Sports with AI page at the top. In brief, I scrape fight stats, engineer a ton of features, cut those features down to the 10-20 that are the best, then use Google Vertex AI in a chronological 80/10/10 train/val/test split so the last 10% of data is never seen by the model and all model evaluations are reflective of how well it did on that last 10% of data.
What is your background?
I was a hacker for about 15 years, now I’m an AI security researcher and engineer.
What are the most influential stats to predict UFC fights?
Besides the bookie odds, my highly tuned custom Elo stat is 62% accurate by itself. As for fight-specific stats, age is a huge factor. 33/34+ leads to massive decline. Others include the round 1 takedown stats. It appears that if a fighter is insistent on going for takedowns early in the fight they are more likely to win. Last, defensive stats and your ability to lower your opponent’s stats from their average baseline are very important as well.
