Do reports have a time span?

[jamepig1]

Like, do you need to be reported for a certain amount of times in a given time for Blizzard to monitor your account? For example, if you're reported 5 times in 2 weeks, will they start investigating? and if the 2 weeks have passed will the number of report reset to 0?

[DvASystems]

Doesn't matter, when they look at your stats after report, they will decide whether to ban or not to ban.

[jamepig1]

Do they usually look at the stats after a certain number of reports has been met? There was one guy who claimed I was hacking so I assumed he reported me. I'm just wondering if that will have an effect on my safety or not. Thanks for the fast reply Dva!

[DvASystems]

Do they usually look at the stats after a certain number of reports has been met? There was one guy who claimed I was hacking so I assumed he reported me. I'm just wondering if that will have an effect on my safety or not. Thanks for the fast reply Dva!

They look at accuracy and your past stats. If there's a 300% skill increase over a short period of time they'll know you are hacking.
1 report won't do much, you need a lot of them as it's an automated system. The ban reviews are manual.

[Babawanga]

i played over 300h on soldier 76 in season 2 using aim assist soldier 82, thats around 1400 games and been told and probably reported hundreds of times. Been reported by group of 6 and 5. Someone even told me that hes recording my aimbotting, I didnt turn it off. I am still in game. Been only silenced/muted in game once for 1 week for flaming on team mates. Thats all

[ownedscrub]

They look at accuracy and your past stats. If there's a 300% skill increase over a short period of time they'll know you are hacking.
1 report won't do much, you need a lot of them as it's an automated system. The ban reviews are manual.

Lol no, they dont look at your stats. That was hilarous tho.

[DvASystems]

Lol no, they dont look at your stats. That was hilarous tho.

How else would they detect AHK cheating?
Please, share your wisdom.

[atmos]

How else would they detect AHK cheating?
Please, share your wisdom.

My guess would be that they just check the mouse/kb input of the replay file and determine whether it's human or not (iirc they use similar replay system as dota 2).
There's been shit tons of whitepapers (well at least a few) for differentiating between the 2 even with low sample rates and considering blizzard's approach to cheating in OW this far I think it'd be reasonable to expect a system similar to that.
While they probably could make the process actively scan for evil bots and flag accounts, I think they are only concerned of 'cheaters' who disrupt the equilibrium, i.e., get reported enough.
Then they just need to confirm it really is cheating to prevent PR fails.

Also pretty sure they scan for some public .exes every now and then.

Edit: just guesses, but given what the korean(?)/chinese 'study' said about their anti-decompile/anti-cheat stuff, it's fairly safe to assume that whatever they use to determine probable cheating (after the fact) is stored on their servers and that would be the location/input/replay stuff that already is stored for at least x amount of time.
..and the human/bot differentiation from input data would make sense due to the low amount of false positives (I've seen on) OW due to the insanely low rate of false positives that those algorithms achieve.

[DvASystems]

My guess would be that they just check the mouse/kb input of the replay file and determine whether it's human or not (iirc they use similar replay system as dota 2).
There's been shit tons of whitepapers (well at least a few) for differentiating between the 2 even with low sample rates and considering blizzard's approach to cheating in OW this far I think it'd be reasonable to expect a system similar to that.
While they probably could make the process actively scan for evil bots and flag accounts, I think they are only concerned of 'cheaters' who disrupt the equilibrium, i.e., get reported enough.
Then they just need to confirm it really is cheating to prevent PR fails.

Also pretty sure they scan for some public .exes every now and then.

Edit: just guesses, but given what the korean(?)/chinese 'study' said about their anti-decompile/anti-cheat stuff, it's fairly safe to assume that whatever they use to determine probable cheating (after the fact) is stored on their servers and that would be the location/input/replay stuff that already is stored for at least x amount of time.
..and the human/bot differentiation from input data would make sense due to the low amount of false positives (I've seen on) OW due to the insanely low rate of false positives that those algorithms achieve.

Well your theory seems to match the intrusion problems their anti-cheat had recently when they got flagged as Malware. Presumably due to the extreme checks they impemented recently.
Banwave 2.0 soon?

[jamepig1]

So how did most people get banned then? Was it because the cheat was memory based which interferes with the game? I'm using Stinkyjoint right now, is quite safe?

[atmos]

Well your theory seems to match the intrusion problems their anti-cheat had recently when they got flagged as Malware. Presumably due to the extreme checks they impemented recently.
Banwave 2.0 soon?

I'm about 76.03% sure they've been doing those deep(ish) checks/scans with each update/patch/(every time it'd be 'ok' for the patcher to go apeshit on your storage) from the get go except they probably just ****ed up something this time or possibly they just got ballzier as they've already established solid enough playerbase.
If my (comparable to conspiracy) theory is even mildly correct and they only ban those who are obvious enough, then the results from the scans would be used as just additional evidence against you.
When you think about it, aimbotter who is really subtle about it, is just a good player to others. The ill-feelings he creates in others with cheating are equivalent as those caused by a legitimately good player (seagull e.g.). By effectively eliminating memory aimbots with the witch hunts, anti-decompile system and rudimentary scans, they've 'roundaboutedly' capped the capacity of the aimbots as well. So a successful cheater isn't Seagull * 20, just 0.45.

Another funky part is the fairly low amounts of people getting banned. Out of the 10 million copies initially sold, only what 1.7k or so were banned in the first wave according to some article.
When you think about the traffic even this forum was getting at the time the 1.7k doesn't even come close to the amount of people who had (probably) cheated. Add in nulled, etc etc and that 1.7k becomes pebbles suggesting their anti-cheat is either totally garbage or they focus on the flamboyant ones leaving the rest to do as they please as long as they are coy about it.

I'm probably way over estimating Blizzard's swagger when it comes to this, but once every thousand years, someone does come up with awesome solutions to ever persistent issues.

[DvASystems]

You are correct.
Judging by this article
Overwatch has more than 20 million players - Polygon
Even if there were 100,000 cheaters, we'd sit at %0.005 playerbase cheating.
And judging from numbers I've seen in other sites, I'd believe the real number is around 10k cheaters. meaning 0.0005% of the playerbase.
Obviously those 20 million players are XBOX, PLAYSTATION, PC combined, but regardless even if we were generous and skewed the numbers to 10 million PC players we'd still sit at 0.01% userbase cheating (assuming 100k cheaters, otherwise 10k = 0.001%)

I think Blizzards PR tactics scared off any cheater. 20 million semi-active playerbase yet all forums related to Overwatch have less activity than WoW 10x.
Blizzard can easily enforce a lax anti-cheat environment as hunting that 0.001% isn't worth it.

[ownedscrub]

How else would they detect AHK cheating?
Please, share your wisdom.

Common sense.
No one does that. Do you really think they got enough people to watch every single player profile to find weird stats? Lol. They must receive 1000s of reports everyday.
Also what if I share my account with my brother who is a pro player? My accuracy would skyrocket and bam, Im banned?
They scann your pc for known hacks/hooks/memory shit. If you get reported they scann your machine a little differently (more aggressively, more frequently, or sooner).

[DvASystems]

I dunno, Atmos theory was quite resonable in that regard.

[ownedscrub]

Also 100,000 is the 5% of 20,000,000 not the 0.005%.