[POLL] Cheat-o-meter for nexus

[weavii]

I want to know your opinion about a stats page with integrated cheat-o-meter for nexus (currently only rome).
This is the version working with EA backend: http://www.team-des-fra.fr/CoM/

It grabs the the stats of a player, calculates a few more and compares it to the stats of other players on a suitable distribution. I think some stats should be weighted differently on nexus than on EA regarding the possibilities of modding. (Score/time for instance)

This will result in a score from 0-0.9(period) that shows "how" suspicious the player behaved. Also we thought about setting up a plugin for procon, that takes action if a player with a certain "suspiciousness level" joins the server. (Notification for instance)

Ofc this is not the holy grail of cheat detection, but when saw BFBC2 EA communities using it, I had the impression of it to be a solid tool that can help admins. I only heard good things of it until now, this is why I especially would like to get the answers of people who think this is a bad/useless tool.
Since the API is being updated we can't really do anything right now and the 800 player stats we scraped so far are not enough to get reliable scores.

On the other hand it could lead to more accusations and "blind bans" without even watching him.

Yes or no and why?

[RobbingHood]

Yes or no and why?

Yes, because it can be an additional factor to consider when "judging" about a player.
Emphasis on additional, however; Using it as the only reason to kick/ban should only be considered for extreme values (like 90% headshots).
I also like the idea of a plugin... Maybe allow to only votekick players with a critical "suspiciousness level"?

[weavii]

Yes ofc this should never be the only reason for a ban, just a little help for admins.

[mouz]

No, because in order to "detect" a cheat basically the planets must align. The player must have been using the aimbot constantly with his soldier, otherwise stats from clean gameplay could mask the cheat. I a player uses a wallhack it wouldn't necessarily affect his stats. The percentage of cheaters that would fall into this category might be too small to bother with this. A player could even use these stats as a proof that he isn't cheating, even though he might have been.

Yes, because it would give us another tool to deal with cheaters. It would also give more detailed player stats, and maybe even a banner generator? It should be relatively easy to port/implement into VU as well.

[Rodney]

Maybe you could ask the french guys for the source code ?

[Kerrigan]

I was laughing at this because it's typically French. Looking for reasons to ban someone who is just better.

Anyway: It might be funny to see how it does work out. It could be a valuable addition when server admins would use their brains to filter out bad guys manually. Keep it going.

It's a "yes"from me.

I will add a poll.
Reasons could still be added by commenting.

[blueman24]

Yes, but saying someone a cheater by only using this tool will not be a good ideal.

[Bruce Lee]

Sorry, but there is no substitute for a cheater demo.

I like to fly the helis in BC2 because it takes some real player skill and also its just plain fun.
There were plenty of matches where I scored 40/4, 50/10, etc.
I thought I had just spent a lot of time learning to fly but maybe cheat-o-meter stats thinks I am haxing, lol.

I understand the desire to weed out cheaters but using blind stats to ban cheaters is like using an axe to swat a fly.
You may get lucky and kill the fly but you are going to do a lot of collateral damage in the process.

[weavii]

Maybe you could ask the french guys for the source code ?

Great idea. We already asked bfbcs.com for the source of the API to dev API converter, but they denied.
Maybe after some minor changes we could use c-o-m right ahead.