UFC Fighting Styles:
Clustering Fighters by Style
Based on Historical Fight
Metrics
Visit the web app here.
I make bets with my friends on Ultimate Fighting Championship (UFC) fights. You get one point for the correct winner and bonus points for the correct round and winning method. For this project, I scraped data from UFCStats.com, performed feature engineering and reduction, and ultimately constructed 10 clusters of fighters, each representing a unique Mixed Martial Arts (MMA) fighting style. The results can be explored on my Streamlit application and on a set of Tableau dashboards I designed to provide information to improve betting. My K-Means approach allows for calculating similarity scores in order to construct a percentage breakdown of each fighter into the 10 distinct fighting styles. The top fighting styles of two fighters can then be compared to discern who has the stylistic advantage.
Read. Watch. Review.
-
Watch the powerpoint presentation.
-
Read the full project write-up on Medium.
-
Watch the app demo and code review.
-
Review the project code on my Github.
-
![]()
What defines a fighting style?
-
![]()
Alternate Name: Try Me.
-
![]()
Tactician says, "MMA is a sport, not a bar fight."
-
![]()
One of us is going to sleep.
-
![]()
A brutal game of rock, paper, scissors.
-
![]()
Explore the data for yourself.
-
![]()
Styles can change. Just ask Jose Aldo.
