[phendrenad2]

Based on those observations, I'm not convinced of your conclusions. There are a lot of unaccounted-for variables (such as game popularity, prize pools, government effecacy in cracking down on black markets, etc.). If someone does this same analysis, and accounts for those other variables in some way, and includes a lot more than one game in the dataset, I might be convinced that there is some evidence that kernel-mode anti-cheat isn't effective, but then, we need to establish that it isn't simply a deficiency in the implementations...