Ran and Experiment on Report System. Results not Surprising, System is Biased and has Flaws.

TL:DR Riots report system only punishes based off chat logs. It claims to be fair and asks players never to retaliate but only to mute people and report them, however it does not hand out punishments equally. Actions in game can physically ruin the game and you cannot prevent that loss, chat can tilt people but CAN be fixed by muting, however riot punishes chat but does not punish trolls. Something doesn't add up. I present a couple of options to mitigate this.

I do care about the issue and would love for the community to improve, as this is probably the only game where I've seen the trolls actually have the upper hand and are able to inflict punishment on good players with little to no consequence. It also promotes a "lets report so and so" environment just like how everyone in the real world tries to sue each other.

So, as someone who is/was utterly disgusted with Riots "chat scanning auto report bot" because it hardly gets rid of trolls and punishes good people. I understand it benefits riot not to have a tribunal where players decide whats best for the community and riot moderates it but rather replace a human interaction with a piece of code which they hardly have to oversee, but I'm not sure how that's best for or even helps the players. If an equal amount of resources from the cost savings was devoted to player issues Riot please feel free to let me know, in which case the return on your investment has not broken even yet.

However, despite all of this I'm a person of reason so even though I was pissed, when talking to some friends they wanted to try some impartial experiments I figured it might be a good idea Note: I did not participate in any of the games only looked at the data after. So no use in bashing me for ruining "normal" for random people I didn't order, execute or encourage it, just examined the results. Feel free to point out any flaws in the design or defend Riot if you want.

Three accounts were created

Account A: Lets call this the FLAMER - This account always played exceptionally well in game but said a lot of nasty things in chat. Account B: Troll -This account never said anything negative in chat, although did occasionally claim to be trolling or not care but would follow up by later saying they are not trolling Account C: Neutral Party

During each of the games the above scenarios would take place, Account B would run it down mid, build troll builds, intentionally feed or go to the wrong lanes, but was silent. Account A would get mad at account B. At the end of the game--not every game because it would diminish the report effectiveness account C would report one or both of the players-equal amount of reports over the total duration, and would encourage people to actively report the troll. Note that this was a biased hypothesis against the troll because I really wanted to stress test the system. Now I'm happy to provide all of the data for each game etc...but, basically what happened was by the end of 50 games,

Account A: received multiple reports and account bans, finally was perma-banned Account B: Absolutely NOTHING. Account C: Nothing-as expected.

There is a plausibility Account B would go straight to perma-ban or ban without warnings because a chat ban wouldn't have harmed them, but if that scenario exists 50 games seems a long time to not have a punishment. Now, during the games while account A had toxic chat, any player could have muted account A, the overall impact would be small but would negatively affect the player experience by prohibiting communication which can be useful in-game to coordinate strategy if used effectively, however Riot has probably determined this is negligible in random solo-q type games because players rarely can develop a good enough relationship to effectively use it, which is why I suspect they allow players to play with chat bans, which I'd hope they wouldn't do if the data showed that players with chat ban substantially lose more often from lack of communication. So Account A could have been muted and would not have affected the game. Player B did not say anything in chat, but there was no way to mitigate or "mute" the actions of a troll. There is no way to make it so their deaths don't negatively affect your team. There are likely three options. A) Ignore the troll- this may work if they get tired of trolling but it has no tangible effect and leaves it entirely up to the other player. Maybe you playing will inspire them to try, maybe not. You will lose LP for the loss but can report the troll hoping to clean up the community B) Try to reason with the troll via chat, be mindful that anything you say can and will be used against you to the fullest extent of the chat bot which scans logs. There can be a very fine line of what riot deems is aggressive towards an individual, and showing anger to someone wasting your time will likely get you reported for escalating the situation

Here I don't see a winning situation, if the report system did anything then option A would be a consolation to player C and A who were negatively affected by player B, but since it doesn't they lose in both scenarios.

Now in conclusion I FULLY understand why it's VERY VERY hard to report trolls and have any sort of CODE detect it. So I present 2 solutions to be pragmatic and practical.

Option A) Its nifty that Riot has a record feature. Now I know they cant possible save every single game that's out there, but if a game is flagged for multiple reports of trolling why not capture it? Why not put the decision BACK in the hands of the community or have a moderator examine it. In this case hand out a VERY HARSH punishment to demonstrate this behavior will NOT be tolerated, and the punishment can be harsher because it required riots human resources to enforce it instead of a system. Due to the new found stigma of reports for actions and trolling ACTUALLY having an impact it will quickly deter players, as currently I've seen many trolls say go ahead report me I've said nothing wrong and the system can't tell I'm a troll (or something similar). The only thing the system does detect is idling or AFK but it doesn't work if the person moves around in base every 30 seconds to crack the simple idle code.

Option B) Adjust the LP +/- according to the performance of the player! Now again I understand sometimes you have a case where someone truely has bad stats and they are a team player etc...well that's lovely but chances are its a fluke if they are ranking up solely by that game after game, Have a +/- 5 LP for performance. If I lose and I'm flawless on the game, give me only -10 LP and a troll who gives up, give them -20. Make them feel the wrath of their decisions more than those you've stuck them in the pot with. I understand riot wants there to be a shared risk/reward, and there is. Losing 10 LP still sucks, so theres still plenty of incentive to work together and not alone. Alternatively, if someone wins and is the carry give them +25 instead of +20 and give the troll they carried +15 instead of +20. This system can scale with your ELO standard LP awarded and also can scale with how bad you troll... Say you were only slightly better than everyone, maybe you only get +1 added to your score. If you were slightly worse? only -1 from your base. This would also work better than having to win multiple games in order to see a change in your ELO.

Bonus Option C) Stop using the stupid chat scanner to report people from games where its a human response to be angry because your system is ineffective and you give them no solution or outlet. Bring back the tribunal where i know many people went years with no reprimand, yet were banned through the scan bot.