Research Impact: This article explored which in-game metrics predict in-game performance in the popular esport Rocket League. Rocket League is a largely popular esport, with more than 90 million monthly players and ~15million USD earnt as prize money through playing it. We used data from more than 20,000 Rocket League matches to build Random Forest models that predict performance and in-game rank, a measure of expertise. By looking at variable performance in these models, we uncovered key performance and rank indicators in 1v1 Rocket League. Our results could inform training and strategy for Rocket League players and coaches at all ability levels.
Smithies, T.D., Campbell, M.J., Ramsbottom, N., Toth, A.J. A Random Forest approach to identify metrics that best predict match outcome and player ranking in the esport Rocket League. Sci Rep 11, 19285 (2021). https://doi.org/10.1038/s41598-021-98879-9