Sony Working on Matchmaking System Based on Game Session Lengths

Sony has filed a patent for a matchmaking system that will match players according to session play time. The system aims to partner players with similar session lengths and will consider other gaming session parameters.

Skill-based matchmaking is commonly used for competitive games like Modern Warfare 3, pitting players with similar experience, match wins, or kill/death ratios against one another, while player location is another factor considered in matchmaking. But, as well as facing the ire of casual gamers in non-ranked Call of Duty lobbies, skill-based matchmaking isn’t as relevant in co-operative games, where players are looking for teammates that stay the course.

To better serve co-op players, Sony is currently working on a matchmaking system that will consider game session length when pairing players. A patent filed with the World Intellectual Property Organization (WIPO) details a system that compares typical session length and looks at factors like the frequency and even the length of breaks that players take. The system then makes matchmaking recommendations based on these factors and, although no specific games are mentioned in the filing, it could be used for casual and campaign-based titles like Bungie's Destiny 2. It isn't clear how far into the development of the system Sony is, or even if work has started on it at all.

This isn’t the first time Sony has attempted to overhaul or improve the matchmaking process. It was reported last year that it was working on a Cross Skill Competition ranking system that groups players into “houses” with other players of a similar skill level. Several houses would be grouped together to create a “neighborhood” with multiple “neighborhoods” forming a «village.» Cross Skill Competition wouldn’t just be used to match players of similar abilities, as players would receive invitations to join appropriate houses, and tournaments would take place for players at different levels. In 2021, Sony filed a patent for an AI-driven