Ranking systems for one-on-one competition have been around for a long time. The most famous - ELO, was developed by Arpad Elo and was the first to have a sound statistical and mathematical basis. It was adapted byWorld Chess Federation in the 1970s. Since then, many ranking systems have branched from it, adding new improvements.
Two of the most popular ones are TrueSkill by Microsoft and Glicko by professor Mark Glickman. They have been used in many online games, including League of Legends, Counter-Strike, Halo, and many others. The secret to keeping players interested is accurate matchmaking. Therefore, these games rely on quick and automatic skill assessment.
On the other hand, professional sports do not have such urgency. It is likely the main reason why ranking methodology in sports often lags behind. In MMA especially, this problem is very evident. In the largest organization - UFC - ranks are decided by media members, which is influenced by human biases and prejudices. Also, the vast majority of MMA fighters around the world do not get ranked at all.
AimRank is a machine learning-based, mixed martial arts skill rating and ranking system. It had been inspired by the vast discrepancy in ranking quality between online computer games and professional sports, especially - MMA. The core engine of AimRank is the Glicko2 skill rating algorithm, developed by Professor Mark Glickman, and applied in computer games such as Counter-Strike, Go, Dota, Guild Wars, and others. It allows for lightning-quick and accurate matchmaking in online games. In professional sports, it offers a superior alternative to human-based ranking systems.
AimRank was developed out of sincere love for the sport and deep admiration for all the athletes stepping inside the ring or cage. We believethat fighters deserve to be recognised and rewarded, for their skill and competitive achievement as fairly as possible. We AIM to create the most accurate and transparent athlete RANKINGS and to continuously improve how it’s done.