They say the percent rate of people that do get banned or permanently banned is 0.06% i don't know if that's based of the entire population of players, or based of the region you are playing from. This number is most indefinitely wrong, a lot of players get banned for little silly things that do not deserve a ban or permanent for that matter.
And this is based on.... what exactly? It's really contradictory when you say you don't know how they pull the numbers, but it's wrong.
Let's remember this is an "ONLINE" competitive game, of course their will be "TOXIC" players and "TOXIC" behaviors by players.
Not sure why online is in quotes, but that's neither here nor there. Just because there will be toxic players, doesn't mean that Riot just wants to accept them. There are going to be people robbing stores, does that mean that we should just sit there and do nothing? You also have to take into account League's reputation of being and extremely toxic game. Having a banning system and banning toxic players aims to fix that.
The term itself is irrelevant to banning of players, if you google the definition it means 1, a poisonous substance, 2 denoting or relating to debt that has a high risk of default.
Context is important here. Specifically the third defination of the word toxic as provided by Marrim-Webster's Dictionary:
**extremely harsh, malicious, or harmful **
The bans on most people are unreasonable speaking relatively, The AI system bans on key phrases or key words in correlation of the term "TOXIC" what does this mean?
Relative to what? You can't just say it's relatively unreasonable with nothing to compare it to.
From what I understand, the AI is a learning system. Each region has it's own specific AI that does the banning, it looks for common phrases that a majority of players deem toxic. That's why you can get away (generally) with more toxicity on EUW as opposed to the NA server. Not only that, Riot has done multiple spot checks to make sure that the AI is working, and they do manually review cases from time to time. But having someone read every single report, that's not realistic.
In my opinion it first of all means, it's looking for evidence in correlation to a proper ban, which searches for a large and vast variety of words in it's database that it has collected over the months, years it has been running.
More realistically it looks for phrases more so than individual words.
In a sense you could say all vocabulary words can be "toxic" if interpreted wrong, or grouped with other words that form a toxic sentence.
This is where "words" and "phases" come into play. Again, context is important, in an ARAM you can say, "Lux suicide to tower, don't give them gold". I've been playing ARAM when you had to play it in a custom game, I've said that phrase multiple times, I have yet to get punished.
This can easily, be misread, miscalculated, misjudged.
While it isn't perfect, it's a lot better than the old Tribunal that we had, the old Tribunal, people would often press punish without reading the logs and they were right 80% of the time. There is no such thing as a perfect system. I suppose you could say that if millions of people reported people for saying "gg" at the end of the game, sure you'd be correct. However the likely hood of that actually happening, not realistic at all.
Delete enter, and play the game and you probably won't get banned if you can't talk, assuming that is.
You won't get punished for not talking in game, that's what the ping system is for. You don't have to talk at all. I've been playing this game for 6 years this December, I almost always talk in chat, I have yet to get so much as a warning. It's not hard, don't be a dick and you won't get punished.
