Statistical Analysis of Tribunal Cases

Comment below rating threshold, click here to show it.

gnfnrf

Senior Member

08-28-2012

Quote:
Originally Posted by Mokkun View Post
Things that I'm somewhat curious about.

% of different types of reports, and how often the types get punished for.
(maybe which champs garner which reports more often)
(maybe which type the perma-bans tend to come from, vs initial warnings)

Summoner spell breakdown.
I would be curious about how often "troll summoners" get reported/punished, but that's messy. Promote and surge seem to have more use on Dominion, and low level games have a number of revives on brand new players. Maybe just checked for ranked?

You already mentioned items.
Reform cards don't have individual report types available.

I think there is a (downloaded, but not rendered) "Most common report type", which I could use, but I'll have to check to see how it works and if it's useful.

I can do a summoner search, I think. On the other hand, it will make me sad, since my best character (in Normals, I don't play Ranked) is a Revive/Surge sacrifice Karthus.


Comment below rating threshold, click here to show it.

Xander756

Member

08-28-2012

Wow this data is truly irrefutable factual evidence of the flaws in the Tribunal system. 61.2% decide to punish? That is way too high. And of that, over half of the punishments end up being bans? LOL COME ON.


Comment below rating threshold, click here to show it.

Mokkun

Senior Member

08-28-2012

ah ya, I derped, they don't show report reasons, too used to trib cases.

Oh, and karthus... is special, and you can make him special.


Comment below rating threshold, click here to show it.

Methelod

Senior Member

08-28-2012

Quote:
Originally Posted by Xander756 View Post
Wow this data is truly irrefutable factual evidence of the flaws in the Tribunal system. 61.2% decide to punish? That is way too high. And of that, over half of the punishments end up being bans? LOL COME ON.
Alright. I'll take on this discussion.

For it to be irrefutable means that there would have to be an objective standard to compare to. The one you've created is a subjective opinion, meaning that it's not irrefutable. I'd think that if you compared the amount of cases deserving of punishment to the amount of cases punished that you could create a better case but at present, the goal would be to have a 100% punishment rate because then it only catches people deserving of punishment. To my knowledge, the US criminal system has a higher conviction rate.


Comment below rating threshold, click here to show it.

Gonzodamus

Senior Member

08-28-2012

These statistics are satisfying my geeky need for crunched numbers. Very nice work. And may I just say that it's really nice to see a constructive thread about the Tribunal that uses numbers instead of anecdotal evidence and wild theories. As usual, the answer to a complicated issue like this is that, well, it's complicated.

I'd love to see Punish/Pardon rates tied in to how long each case was read, but I'm guessing that data doesn't exist. I bet we'd be able to learn something about Tribunal members from that.

Can't wait to see Part 4 and, of course, the conversation that follows.


Comment below rating threshold, click here to show it.

gnfnrf

Senior Member

08-31-2012

Part 4, the final part, is in the fourth post.


Comment below rating threshold, click here to show it.

theshim

Senior Member

08-31-2012

This was truly fascinating. Thanks for doing this, it's been incredibly enlightening!

I kinda wish you could do it again, like once a month, just to see if trends change or normalize over time, but I appreciate that this was a hell of a lot of work and it might not be what you want to do over and over. Still, really cool. Thanks again!


Comment below rating threshold, click here to show it.

iatre

This user has referred a friend to League of Legends, click for more information

Senior Member

08-31-2012

Yay, more statistics.

I see how badly the "intentional feeding" reason is being abused. I virtually never report under intentional feeding, because I virtually never see anybody doing it. I guess that's one of the driving forces behind banning false reporters.


Comment below rating threshold, click here to show it.

Gonzodamus

Senior Member

08-31-2012

/swoon

Thanks for finishing this up, I'm really fascinated by some of what's been revealed. Like iatre said, I had suspicions that Intentional Feeding was the report type most often used and likely the one most often abused.

Also? Go Trundle, for statistically saying more than just "kthxbye"


Comment below rating threshold, click here to show it.

onihero

This user has referred a friend to League of Legends, click for more information

Senior Member

08-31-2012

I didnt see this, might have missed it.

Whats the % of reports from a loss vs the % of reports from a win?