Counter-Strike 2 player magga_ completed a master's thesis at the Norwegian University of Science and Technology, researching how to identify different accounts operated by the same person based on gameplay habits. In a test sample of over one thousand players, this method successfully identified 13 sets of alternate accounts known to the researcher.
This study addresses a limitation of account bans: cheaters can simply register new accounts or purchase existing ones to return to the game after losing their current account. The research used standard match records saved by the game to extract individual behavioral features from mouse movements and keyboard inputs, including repetitive actions such as quick flicks, counter-strafing, recoil control, and grenade throws.
In the sample tests described in the paper, mouse movement analysis achieved a 100% accuracy rate in identifying users, while keyboard input patterns reached 98% accuracy. By combining both metrics, the system identified 13 sets of alternate accounts known to the researchers. When the authors verified some matches initially considered false positives, they even discovered additional backup accounts.
These results apply only to the current test conditions, and the method has not yet been validated across Valve's large player base. The researchers also noted that multiple family members sharing the same computer could increase the difficulty of identification. He suggested using behavioral characteristics to flag accounts for manual review rather than as direct grounds for bans, and expressed hope that Valve or FACEIT would assist in conducting tests in real-world environments.