So the server is coded very poorly and they're too lazy to fix it or add a mechanism to cut the connection when players start trying to abuse known faults in the software. Just head straight for DRM. Par for the course.
The lag bots are gone at least for now because Valve in its eternal wisdom... disabled voice and text communication for all new players. (The little I read about those lag bots, they used the game's audio voice system to spam the server so that it became laggy to human players, so Valve "fixed" it by disabling it altogether for new players.)
Not sure why they also disabled text chatting, probably because those cheater bots usually also spammed the text chat and used expletives like "ALL N-WORDS MUST DIE!!!" etc., trying to provoke people. Valve probably felt they will be sued by some SJW organization for allowing hate speech in their game, if they don't prevent it from happening at all.
Of course that "solution" now sucks for new TF2 players as they can't communicate at all with other players, not with audio nor text.
However, when it comes to actual cheating like an aimbot doing the aiming for you or at least "clicking" the mouse button at the exact right time that you get 100% hits... I think the problem is that it is quite hard to reliably determine whether someone is cheating or not. Even for a human player it is quite hard to tell if the player is just exceptionally good with long-range headshots, or cheating.
The system can't simply determine "if the player gets 20 headshots without misses, then it must be a cheat" because then the cheat program would simply be modified to miss occasionally on purpose. And, if there is some damn good player who gets 20 headshots in a row, against newbies who are easy targets running in line?
Unfortunately the best way to catch cheaters (and others who are apparently there trying to "break" the game) is that human admins can oversee what is happening in the game, and even watch the alleged cheater playing. When an admin has determined that it must be cheating, at that point the question becomes, how can the admin ban the gamer from the game for good?
As I explained before, e.g. the cat-bots that infested the game at least earlier (and I think they are still there, the name has just changed), the whole cheater program was apparently an open-source software that you'd run in a Linux machine, and it would automatically create new free Steam accounts which the cheater bots can use. So it didn't really matter if some automated system or a human admin would ban a cat-bot, soon afterwards ten fresh ones would pop up.
So there are two separate problems:
- How to detect cheaters (reliably).
- How to ban detected cheaters for good.
DRM doesn't help with the first one (unless it is those anti-cheat systems that have high system priviĺeges so it can see if the player is running any known cheater programs), but it does with the second one. But not alone, there has to be also something else like the the game costing money (in which case the banned cheater would lose money because his purchase doesn't work anymore), or some sort of semi-reliable identification that can't be easily automated by the cheater bots.
As said, there is a reason why the competitive mode of Team Fortress 2 seems to be free of cheaters (at least I don't recall seeing any), while the free casual mode is full of them, even blatant cheater bots that don't even try to hide the fact they are cheating or trying to break the game.