The problem is obviously the same as in many other industries - how do you distinguish honest legitimate players who swear they haven't cheated from people who will say anything to get you to unban them. I don't work in that department personally, but I've seen reports shared internally where the player literally went to local news station to say how unfairly they are treated and how we banned him without any info or any reason and how it's affecting his mental health and his family and he basically made a huge stink around it, and then we pull up the ban report for his account and we clearly see a screenshot from his machine where he's running cheat engine with cheats for our game enabled. Some people will just lie through their teeth to get their way. So you have to rely on what you know with absolute certainty - you detected something that is absolutely indicative of cheating? You ban them. Anything else is a no no. At least where I used to work no one used any kind of algorithm for automatic bans, those were only used for manually reviewed cases where someone would actually watch a replay of your game before issuing a ban.
Does that mean the system is foolproof? No, of course not. But banning honest paying users is a huge risk to any business - so obviously no one wants to do that, every system like this errs on the side of caution by default for that reason alone.
And obvious disclaimer - I can only comment on my own experiences, I have no idea what every company out there is doing.
> how do you distinguish honest legitimate players who swear they haven't cheated from people who will say anything to get you to unban them.
It's mostly not about the appeals process. You want to avoid the false positive accusations to begin with.
> and then we pull up the ban report for his account and we clearly see a screenshot from his machine where he's running cheat engine with cheats for our game enabled.
Hypothetically things like this can happen where someone is reusing passwords that end up in a data breach and then some script kiddie gets their hands on it and wants to dip their toes into some cheating without risking their own account. Then you have the original account holder screaming at you because they know they didn't cheat.
Or they could just be cheaters who doth protest too much.
But there are ways you can at least try to distinguish these things, e.g. did the cheating happen on the same PC or IP address the account normally uses?
> Does that mean the system is foolproof? No, of course not. But banning honest paying users is a huge risk to any business - so obviously no one wants to do that, every system like this errs on the side of caution by default for that reason alone.
It's apparently failing enough that this thread has multiple people saying they've experienced false positives, and it doesn't seem like they're interested in getting their accounts back.
You can't just say that though, you have to actually do that, which is apparently not what's happening.