What is aram-mayhem-database?

ARAM Mayhem Database is a community project that tracks player statistics for a League of Legends game mode called ARAM Mayhem. Riot Games removed this mode from its public match history API starting around patch 14, so websites like OP.GG and U.GG cannot show any data for it. This project works around that limit by using the local API built into the League of Legends client itself, which can still show your own recent matches and the matches of people you have played against. A small program called a collector runs on a contributor's computer while they are logged into the League client. It follows a chain outward from the contributor, checking their own matches, then friends, then opponents, then opponents of opponents, gathering the champion picks, chosen augments, and win or loss results from each ARAM Mayhem match it finds. That data is saved locally, then exported as a file containing no personal player IDs, which the contributor shares by opening a GitHub Issue and attaching it. Submissions merge automatically without duplicates. So far the project has collected roughly 38,000 matches. The results are published as a free website showing which champions perform best, ranked by a win rate calculation that accounts for how many games were played so lucky streaks do not skew rankings. Clicking a champion shows which in-game augments help or hurt it the most, with descriptions shown in Chinese. Visitors can filter by role and search in Chinese or English, and the site adjusts automatically for phone screens. The site itself is plain HTML, CSS, and JavaScript with no framework and no server, hosted for free through GitHub Pages. The data gathering scripts are written in Python. The project plans to eventually add a machine learning model to predict how well two champions perform when paired together. It is an unofficial fan project and is not affiliated with or endorsed by Riot Games. It is released under the MIT license.