Enhancing fairness and player satisfaction in electronic games: an analysis of matchmaking algorithms and player feedback in MOBA games

Abstract

This research focuses on addressing the challenges identified in current matchmaking systems to enhance matching accuracy, fairness, and player satisfaction in Multiplayer Online Battle Arena (MOBA) games through the development and implementation of a novel matchmaking algorithm, CentroMatch. Leveraging foundational elements from established skill rating systems such as Elo and TrueSkill 2, and innovative graph-based algorithms, CentroMatch aims to create a more accurate and balanced matchmaking process. A comprehensive literature review was conducted to examine existing skill rating systems mentioned previously and emphasize the importance of player feedback and fairness in maintaining player engagement in MOBA games. The Elo system's estimation of winning probabilities and score updates, TrueSkill 2's incorporation of additional variables, and graph-based models' visualization of skill gaps form the fundamental structure of CentroMatch. Surveys and interviews were administered to collect quantitative and qualitative data that would be integrated within the fundamental structure of CentroMatch to apply to the CentroMatch algorithm. The MOBA game of interest, League of Legends ™ (LoL), was selected to serve as the empirical testing ground for the new algorithm. From the gameplay results, CentroMatch creates more balanced matches and reduced the possibility of toxic behaviour by constructing a model structure based on the feedback from questionnaires and interviews. The improvements contribute to the broader field of electronic game design and development, offering insights for future advancements in matchmaking algorithms.