Price Dynamics and Volatility of Virtual Items in Online Games: Evidence from CS:GO Weapon Balance Updates and Case Releases

Virtual-item prices can react to game updates, but these responses are difficult to separate from trends already under way. Based on Buff.163 daily minimum listing price data from July 2021 to January 2024, this paper examines five CS:GO events—two weapon balance adjustments (M4A1-S magazine capacity reduction, combined AWP and M4A1-S nerf) and three weapon case releases (Dreams & Nightmares, Recoil, Revolution)—using event study methodology, a two-way fixed-effects DID model, and volatility analysis. The results show that treatment-group skins exhibit a significant pre-trend reversal before and after the updates; their outperformance relative to the AK-47 control group disappears immediately and turns sharply negative post-update. M4A4 skin prices exhibit a positive abnormal performance as a substitute weapon. All three cases show negative price dynamics relative to the benchmark over their observable periods. The study demonstrates that virtual asset pricing is highly sensitive to game balance decisions, while also revealing that pre-trends and data gaps impose serious constraints on causal inference.