Matchmaking systems are one of the core features of experience in online gaming. Matchmaking systems influence player satisfaction, engagement, and churn risk. The paper looks into the current state of the theoretical and practical implementation of such systems in the mobile gaming industry. We propose a basic classification of matchmaking systems into random and quasi-random, skill-based, role-based, technical factor-based, and engagement based. We also offer an analysis of matchmaking systems in 16 leading mobile eSport games. The dominant industry solution is skill and rank based systems with a different level of skill depth measurement. In the further part of the paper, we present a theoretical model of engagement and a time-optimized model.