Continuing on from yesterday, the only problem I have with CS is the cheater issue. Yes there are some systems in place like trust factor and stuff, which works very well in theory. But this isn't a fool proof way of anti-cheat and could seriously get some legit players in trouble if anyone suspects them of cheating, and if they're good at the game they almost certainly will. Apparently the team are working on VAC improvements, but I think it's fundamentally flawed. Their late detection system, allowing accounts to play for months without a ban even after detection is, again, good in theory, but in practice it just lets cheaters play loads of matches without consequences. I understand that VAC only ever bans players when its 100% certain, as you usually never receive a false VAC ban, which is good, again, in theory. I hate the game, but Valorants kernel level anti cheat is actually good. I've never encountered a cheater in that game and it's because if you want to cheat in Valorant then you need some serious cheats (I think, I'm not 100% confident on this). With CS, you can literally go and pay £10 for a cheat and play for months, if not years! It's ridiculous, and if you're smart enough you can just write your own cheat and almost never be detected because VAC works from comparing data sets. For example, if VAC detects an unusual .exe file, it will log it in their system. If it detects multiple users with the same .exe running the same code then that can trigger a ban wave. But if you're the only one with your cheat, it can be harder to detect. I understand that kernel level anti-cheats are very intrusive, but I'd honestly take it just to play properly. I take it seriously and don't mind sacrificing some ambiguous privacy.
So, my solution would be to make VAC kernel level. I'd even go as far as to verify home addresses or something crazy like that. That might raise privacy issues, but make it optional? Or require you to have a phone number linked to your account, I don't know. Secondly, I would add overwatch back in two forms. The first form would be classic overwatch where you can watch demos and determine if the suspect is cheating. The second form is reserved for very trusted players and it allows you to actually watch the match live and determine if that person is cheating. The suspected cheater wouldn't know that someone is watching them, meaning every time ANYONE plays a match, they could be being watched and if they slip up then that can result in a ban. Then again, I'm not sure if this second overwatch method would allow for instant bans or if they'd just put a verdict at the end and then the account would be actioned a few days later.
There are obviously trade offs to kernel anti-cheat, as well as adding even more layers of protection like overwatch and this new live overwatch that I suggested. It's like making a sandwich. You can keep adding different fillers to make it taste nicer, but at some point it'll just become a messy meal. I mean ESEA and FaceIt both have excellent anti-cheats that outperform VAC in terms of speed and deterrence, meaning you encounter far less cheaters on those platforms. Why don't Valve just take inspiration? I am not sure, and I don't think I will. Valve make strange decisions that don't always seem to make sense, but maybe one day they will.