Tracer: A Forensic Framework for Detecting Fraudulent Speedruns from Game Replays

The Tracer framework establishes a rigorous forensic methodology for verifying the authenticity of speedruns, addressing the systemic vulnerabilities inherent in traditional trust-based moderation. By categorizing gameplay into three distinct interaction spaces—Physical, Human-Computer Interface, and Cyberspace—the framework provides a structured heuristic for identifying manipulation. This approach shifts verification from subjective community consensus to objective analysis, focusing on physics coherence, media continuity, and internal game logic to detect sophisticated deceptions such as footage splicing, input automation, and the illicit manipulation of pseudo-random number generation.

The necessity for such a framework is underscored by a surge in high-profile fraud cases between 2020 and 2025, where financial incentives and community prestige have driven increasingly complex methods of deception. These incidents, ranging from manipulated pre-recorded content to compromised live-streamed performances, demonstrate that current verification systems are ill-equipped to handle modern technical exploits. By documenting these controversies, the analysis highlights that fraudulent activity is not limited to a single platform or genre, but rather represents a pervasive challenge across the global speedrunning ecosystem.

While current implementation remains a labor-intensive manual process, the long-term objective involves transitioning toward AI-assisted tools and standardized logging protocols. Enhancing the scalability of these forensic techniques is essential for maintaining the integrity of competitive gaming as the industry continues to professionalize. Ultimately, the adoption of a standardized, multi-dimensional forensic framework serves as a critical defense against bad actors, ensuring that the competitive landscape remains grounded in verifiable human performance rather than technical artifice.