Okay so I've been running some... let's call them uncontrolled tests because honestly my methodology is garbage right now. Ran Mullvad on WireGuard for a week playing the same FPS match daily, switched to no VPN for a week, and then tried Nord on their meshnet thing. The raw ping numbers from the game client say VPN adds 5-15ms on average which should be bad but subjectively I felt more consistent, less spikes. My theory, which is probably wrong, is that it's smoothing out routing inefficiencies between you and the game server, even if the absolute number is higher. But then you hit those servers with VPN IP bans and the whole thing falls apart. Been chatting with a dev friend who runs private servers and he says most game traffic is UDP anyway and VPNs just add encapsulation overhead, so the 'stability' I feel might just be my brain coping with a worse connection. It all comes down to the server location and your base ISP route though, if your ISP routing to the game server is trash then a VPN with a better route might actually help, but that's so variable. I'm gonna set up a proper test with pingplotter logging soon but I have a sneaking suspicion the whole 'VPN for gaming' thing is mostly marketing fluff for most people.