Counting Cheaters - Part 4

Methodology

Because of this blog, I have a record of the name of everyone who kills me, as well as a mostly complete record of the name of everyone I manage to kill.  Every two months this wipe (12.12) I look up all those names in the in-game chat tool.  The presumption being that if a name is no longer in the chat tool then either that person changed their name, or they were banned for cheating.  

It is not a perfect methodology nor do I have a lot of data, but it's far better than just going by feel, which is what most people do. The arguments about cheating on Reddit are mostly silly as people just try to generalize anecdotal experiences. 

My Anecdotal Experience

Since I broached the topic, here is my own anecdotal experience. Before I started doing this I didn't feel like I had encountered many cheaters in Wipe 12.12.  Just a couple.   But since I've started tracking, my view has become more jaundiced and now I have more suspicions. BUT, overall this wipe, I feel like I've only had a handful of encounters (5 or 6) where I really suspected the person might be cheating.  

I have never had anyone call my name out in a raid, nor seen anyone flying through the sky or other obvious obvert cheating behavior. I have had people who shouldn't be able to see me avoid me though, and I may have once encountered an invisible player.  Of course, I have been insta-killed by a PMC many a time, but that could be luck or desync . 

Number Of Players I Have Reported This Wipe

• Zero. 

I have not reported any players for cheating this wipe.  I did submit a couple reports in the previous wipe (12.11) where it was pretty egregious and I was fairly frustrated. But this wipe I have not reported a single player for cheating, even for the handful where I suspected it. You just never know, right? Maybe it was luck, right?

Raids And Other Stats

As of this writing (June 5, 2022):

• 158 Raids Run
• 74 Survivals (47% survival rate)
• 82 KIA 
• 56 deaths to other players ( 68% of all my deaths, and 35% of all my raids)

The Players I Have Killed (and for whom I have tags)

• So far this wipe, I've managed to kill 25 other PMCs but I only have tags for 19 of them. 
• Of those 19 PMC I killed and tagged, 18 are still in the chat system.
• Only one PMC that I have managed to kill is no longer in the chat system.
• That is a ratio of 1/19, or 5% 

Conclusions

Not only am I a terrible player, but I have not managed to kill cheaters, or maybe just one.

The Players Who Have Killed Me

• So far this wipe, I've been killed 56 times by other players.
• Of those 56 killers, 42 are still in the chat system.
• Of those 56 killers, 14 are no longer in the chat system. ( 25% of my death to players, 17% of my deaths overall, 9% of all my raids )
• Those 14 players have possibly changed their name, but more likely have been banned for cheating.

Conclusions

Fourteen PMC are not a lot. Presumably one of them may have simply changed their name. But regardless, that's one out of four killers that has likely been banned for cheating.

Further Conclusions and Remarks

Banning Takes Time

As I've done this three times now, I notice that it seems to take awhile for players to disappear from the chat system (ie, become banned). I had a case were I strongly suspected someone was cheating in one of my raids, and it took almost two weeks for that player to disappear from the system.  It's nice to have my suspicions confirmed. When looking at the data laid out linearly, you can clearly see the missing players regularly occurring through the raids until the most recent span - the last two weeks.  None of the players from those raids have disappeared (yet).  

When I tracked this previously, it was about 1 out of 3 killers who were no longer present. Now it's 1 out of 4. That's good.  BUT, if I exclude the last two weeks, it comes back to 1 out of 3. The problem here is that there simply isn't enough data.  Small changes result in too much perturbation. 

On Suspicions

Of the handful of players from two weeks or later I suspected were cheating, none of them are still in the chat system. Not one. Meaning that the players I suspected of cheating have since been banned. Again, nice to have my suspicions confirmed.

Of the other players that are no longer in the chat system, most of them I did not suspect cheating. I mean, yeah, they killed me. But I suck, so that's to be expected. 

Cheater/Killer Ratio NOT Same As Cheater/Player Ratio

Even if we assume that none of these guys changed their name, the fact that 1 out of 4 of my killers were cheating does NOT equate to 1 out of 4 players were cheating.  I've killed and tagged 19 players and only one of them is no longer in the chat system. This suggests that cheaters have an outsized role. Which is fairly obvious if you think about it. But that means that the ratio of players actually cheating is likely lower than 1/4.  What it is, I'm not sure and am not sure how to calculate.

BattleEye Works

On Reddit people are quick to bad mouth BattleEye the anti-cheat system used by BSG. But AFAICT, it is working.  Sure, it doesn't seem to exactly prevent people from cheating, but given the nature of the cheat arms race perhaps that's not a reasonable goal. But it does seem to be catching folk.  There are always anecdotes that people share (  "I know this guy who is level 1002 and has been using cheats in EFT since 1984 and he hasn't been caught yet" ) and likely there are cheaters who are able to evade for long stretches of time. But I suspect they are exceptional, not the norm. More likely they were caught and banned and just repurchased an account.   After doing these checks and calculations, I'd have to say that the anti-cheat system is working at catching people. 

But once caught does it prevent them from joining and cheating again? Probably not. Could that be improved? I don't know, but I wish they would. 

I Am Not Good At This Game