I’ve derived and implemented a custom machine learning algorithm for rating Super Smash Bros. Melee players based on head to head data. Check out the Jupyter Notebook on my GitHub here!

The method doesn’t have anything in particular to do with Melee and could be used to rate players of any one-on-one game like tennis or chess too. My work is hardly very original. It was heavily inspired by both TrueSkill and the Elo rating system, and influenced by my study of deep learning. I had a lot of fun working out the math and writing the code. Going from abstract derivations to results on real data is very cool.