Norwegian student proposes identifying CS2 players by mouse and keyboard behavior
The research project links accounts through control patterns in an attempt to make ban evasion harder.
SHOT editorial team1 min read
A student at Norway’s NTNU has developed a behavioral identification method for Counter-Strike 2 players. The system analyzes mouse movement and keyboard input to create a user fingerprint.
The method was produced as a thesis project by the student u/Magga_. It processes match recordings and compares how players move their mouse and use the keyboard across different accounts.
The test set contained more than 1,000 players. Mouse movement alone distinguished every known participant in the sample, keyboard data reached 98% accuracy, and the combined signals separated all known players and accounts. The system also uncovered several previously unknown smurf accounts.
The method has not yet been tested against a player base the size of CS2. Large-scale trials by Valve or major matchmaking platforms would first require safeguards for privacy, false matches and wrongful penalties.
Share this news
Send this story to a friend or your team chat
