![]()
You're level 62, that Zoe was level 48. The lowest level player in that game was level 39.
You also didn't have any particularly great score. Sure it was better than Zoe, but you aren't *that* much better.
That's just scratching the surface.
Looking at your match history, even in the games where you are matched with player above or equal to your skill level, you don't do particularly better than usual.
All in all, the level of the players you are matched with doesn't really matter significantly when it comes to your performance. This makes sense seeing as a level doesn't tell you the skill of a particular player. That's what MMR is designed for. You are matched with players that have similar MMR to you, regardless of their level.
And sorry to say it, but you're pretty much the usual 'My teammates are holding me back' type of player. Instead of focusing on how poorly your teammates are doing, focus on improving yourself. (You *really* aren't that much better than the players you get matched with, according to your match history).
|
|||||
![]()
> First of all, Riot games does not have warnings, only punishments, along with pure political zero tolerance policy. This might be the easiest game to get banned at in the entire world, considering that both ToS and summoner's code never draw the line.
At every stage of the punishment ladder, except for permaban, you get a warning with your punishment. The reform card that shows clearly states that future toxicity will lead to harsher punishments. Then it gives you examples of toxicity.
The 14 day ban even states that the next punishment will be a permaban. So you never just get permabanned out of the blue, for toxicity atleast.
> There is no defined line.
Because a clear line is impossible to define. You can only ever show examples of players that have definently crossed this very blurry line, and hope that they can see it for themselves. Which most people can. In fact, most people know not to cross this line in the first place.
> I wonder how many people behave the way that gets people permabanned and never receive a chat restriction.
Only an exceptionally tiny minority. On boards you'll find lots of examples, but that's again just an even tinier vocal minority.
> Riot never told what "toxic" is.
> Actually, Lyte did, but his definition covers only a small number of evil people triggering hundreds of others. The worst of the worst. Not some average joe who called you bad.
I've previously made a post that explains why you can't just 'define' what toxicity is, to an extent. (It was more specifically on why making a list of toxic phrases/things isn't feasible.)
But the same priciple goes. As soon as you make a clear definition of what toxicity, players will just find a way of tip-toeing around that line. Again, you can only hope that most people are rational enough to know what toxicity is, without having to clearly define it.
And as it happens, most people *are* rational enough to do that.
|
|||||
![]()
> Well, decent people find it difficult to process that you might be banned or even permabanned in a videogame for sarcasm or a report call. "Thin ice" policy is the disadvantage in comparison to people who can do those things for years without getting punished.
Decent people would agree that after several warnings, a personal conversation with an employee about behavior and the summoners code teaching you how to be a good teammate, messing up a last chance that you were given out of the ordinary is a monumentally stupid thing to do.
And that 'thin ice' disadvantage you're talking about, is the same disadvantage you get after any punishment.
Any reasonable person would agree that you deserve a harsher punishment if you're continue behaving out of line, right after being told to stop.
Yet 75% of players who recieve a chat restriction, never receive another punishment.
From this, you can of course conclude that the 'disadvantage' isn't as much of a disadvantage compared to these players just not being able to reform, which isn't something Riot imposed on them.
> Basically, you are in the environment where everyone acts in a way that will take your account away forever and you are not informed about it.
You're trying to say that players that were part of the unban experiment didn't know that they'd get permabanned again if they kept being toxic? That's just not true.
|
|||||
![]()
What 'severe disadvantage' do they start at?
Literally the only thing they need to do is not flame, and they'll succeed. They don't need to be nice towards their teammates or anything. Just don't write anything negative.
|
|||||
![]()
The amount of money you spend is 100% irelevant, and mentioning it makes you sound entitled.
You should post your chatlogs, the information you provided certainly isn't enough to go off.
If you want a direct answer from Riot, you'll have to make a ticket with [Support](https://support.riotgames.com/hc/en-us)
|
|||||
![]()
The only way to contact a person, would be to contact support again. When the bot answers, just reply and you'll likely get a human next time.
But I can already almost guarantee you that it wont help you, other than to understand *why* you were punished.
In any case, you should've semi-recently have had a 14 days ban. On the reform card it states that the next punishment would a permaban, so it's not really without warning.
> I might deseve some form of punishment, maybe, but I do not deserve to lose an account I have spent so much money on over the past decade.
Sorry, but you don't deserve special treatment just because you've spent a lot of money. That's would be outrageously unfair towards those who can't afford to spend like you do.
|
|||||
![]()
So looking through your match history, if found the 'infamous' player.
My only question is: How do you define inting, deterministically. (Meaning, you can't just say: "Inting is intentionally losing"). What steps would one have to take, to verify that a player is inting?
You could look at the score? But he isn't doing paticularly horrible KDA-wise. Sure he has a lot of deaths, but he also has alot of assists and kills.
The damage? He consistently does an expected amount of damage for a support Alistar.
Doing role related things? He also often places a shit ton of vision.
The build? Nothing is inherently wrong with his build. You shouldn't get punished for not building the exact same items anyone else does.
And infact, over the last 110 games, he had a positive winrate?
So why do you say he's inting? Unless I have the totally wrong player (I'm talking about the 'Ohmwrecker Alistar'), there isn't any conclusive proof that he is actually trying to lose, right?
And if it's not possible to determine whether or not someone is inting, then how is *Riot* supposed to do so?
Soft inting is per definition not something you can detect. It's inting, that isn't detectable. So how do you expect 'soft inting' to be adressed in any way?
|
|||||
![]()
> [{quoted}](name=Jikker,realm=NA,application-id=yrc23zHg,discussion-id=tjjLFMgb,comment-id=0000,timestamp=2019-02-12T14:40:54.170+0000)
>
> "This page isn't working"
>
> Maybe try uploading to imgur?
https://media.discordapp.net/attachments/372610819856531456/542678949981978624/received_2280911765527753.png
It was just the arguments in the URL that messed it up :)
|
|||||
![]()
Except in real life, self defence works as a detterence.
In game, "self defence" only works to make the game even MORE toxic.
You aren't going to make anyone flame you less, by flaming them back. In real life people will stop attacking you if you hit them hard enough.
|
|||||
![]()
> Only the client dies funky things like the 100% load bug that many people get.
Ah, I agree in cases where bugs with the game is at fault, then there shouldn't be a LeaverBuster.
> Once you get leavers, you get it more often, by no fault of your own
Care to explain? It's not like the bug will happen more often to you, if you get it once right?
I'm genuinely asking, as I've never personally experienced it.
|
|||||
![]()
Sorry-
LeaverBuster only targets consistent AFKing, so that players that get accidentally disconnected once or twice wont get punished.
But disconnecting still means the game is pretty much ruined for the 4 other players on your team. If you often have technical issues, you should solve them before trying to play the game. Also ensure that you have enough to time play out a full game.
In short, a LeaverBuster for going AFK due to technical issues is 100% intended. You shouldn't play when you're not sure you can keep a stable connection to the servers.
|
|||||
![]()
I mean...
Aren't you also being overly sensitive if you're set off like this because of a 'snowflake' calling you toxic?
|
|||||
![]()
> I dont have a reform card
Then what does it say when you're trying to log in?
|
|||||
![]()
Shaking my head my head
|
|||||
![]()
> The english language has less than 200,000 words.
Sure, but for each of those words there are a lot of ways to intentionally misspell them. But hether or not a list of these misspellings is *actually* infinite isn't really important. It's just too large to be of any use.
> There's probably an infinite number of compounds that can be created. Doesn't stop us from making a table with the base elements.
But we don't want to ban players for just using the base words. We want to ban the for using them in specific ways.
If we're talking hatespeech and encouragement to suicide then we might always want to ban players for using those terms in 99% of cases. And maybe a list of those things could actually be made. (But it shouldn't be released, because players will undoubtedly just use that to their advantage to flame others without getting punished)
But when we're talking about something more nuanced like regular toxicity. Something where we don't necessarily want to just ban players for the words that they use. We have to make sure that they are actually toxic, and not just saying words that a toxic player might use.
If someone says 'Wow, this Annie is trash', then what part of the sentence do we detect? Do we just see that they used the word trash?
But if someone else then says "Wow, this Blue Buff is trash", and all we're doing is looking for the word 'trash', then we'd punish an innocent player.
So now the list of base elements is useless, because it would just result in us punishing players for something they didn't do.
|
|||||
![]()
Then we'd be back to square one :)
How would you enforce that second rule?
Manual reviews wouldn't be possible, it would just require too many resources.
So what do you do? You can't just make a list that filters through the variations of "I like to eat Pizza". You'd just end up with the same problems.
It's a loop :)
|
|||||
![]()
Well a lot of players believe gaming to be something more than just gameplay. Part of it is also the community, and the having fun with your peers. A positive game with no toxicity is A LOT more enjoyable than a game with much toxicity, atleast in my opinion.
I think having part of your company focus on Player Behaviour is a great thing. It's not like throwing more people at the gameplay or art division will magically make the game prettier or better balanced. So it's not like Riot would gain anything from removing the Player Behavior team, but they'd still lose the people that are fighting for the possibility of playing games where everyone can just act nicely to each other.
|
|||||
![]()
> We can teach a computer to process, but not to feel.
That's a quite controversial opinion. Especially since defining what a 'feeling' actually is, is quite a hard task.
Nevertheless, I agree with you. Atleast to the extent that Riots system doesn't actually 'feel' anything.
But a discussion like this is better left to the philosophers :)
> No reasonable detection system for an online video game is going to find this, but a person reading it would reasonably be expected to easily interpret the meaning of the message. A specialized supercomputer like IBM's Watson might be able to figure it out after extended deep-learning training, but it would hardly be reasonable to expect a gaming company to develop a supercomputer and deep-learning algorithm to detect this. Human reviews will remain crucial in identifying anything subjective in online interaction for the foreseeable future.
This is exactly my biggest problem with the system. It already works like an AI, but due to the complexity of toxicity it requires too much data and processing power to suffienctly train the algorithm.
Using an AI is a really great idea, and it's such a shame that we aren't better at using the technology.
|
|||||
![]()
There's a saying that anyone who regularly works with Regex, can tell you why you shouldn't work with Regex :)
But let's say that we decide to do so. You *might* be able to use an expression, that expression would again end up being infinitely large. Maybe not for simple words, but certainly for sentences.
So you'd either end up with an expression that is too narrow to detect toxicity, or too broad so that it matches too many things that *aren't* toxic.
Regex is really great, but it's a trap. You can't use it for things like this. You need something that can dynamically assert whether or not a chatlog contains toxicity, which is what Riot currently uses.
|
|||||
![]()
> Also, you are giving a very detailed explanation on how to avoid the IFS, you should be banned for this from the Boards.
Riot is smarter than you might think :)
This isn't an explanation of how to avoid the the IFS. This is an explanation of why you CAN'T avoid the IFS.
After the part where I explain that you can't make a list, I explain what Riot's system *actually* does. It doesn't care about weird characters, or alternate ways to phrase things. It can actually dynamically decide whether or not something is toxic, even if it has never seen it before.
|
|||||
![]()
Then we end up at square one :)
To make it a bannable offense to circumvent the filter, would require yet another filter to catch the different ways of circumventing the original filter.
For various reasons, it's not feasible to have humans check every chat log for players that are circumventing any filter you make.
Even if you only had to review 5% of all games played, and each review on average takes about 10 minutes, we are still talking about years of manhours of work, just to meet the quota for a single day worth of games.
|
|||||
![]()
It's clear that my post didn't properly communicate my point, sorry about that :(
My point was actually that making a list (or creating a filter) would be impossible, because you can always find a way to avoid a filter. (Changing up letters and such.)
Which toxicity it becomes ever more difficult to make a filter, because toxicity often isn't based on any words, or sentence fragments that you could even begin to put into the filter.
It's a series of words, that can be put in a infinite amount of different arrangements. You can't make a infinitely long filter. No computer in the world could process it, and you could never even finish it.
|
|||||
![]()
Sorry about that :)
My post is a bit messy, because at first it was just a response to players saying: "Well Riot should just make a list over things that are toxic".
But it kind of turned into a rant about how many player who frequent Player Behavior are actually really condescending when explaining things.
---
It basically boils down to me trying to be the change I want to see.
I explained the concept of a list in a way that is painfully obvious to most players. But I did it because there are also a lot of players that *don't* know why making a list over toxic elements, isn't possible.
So without being condescending, I explained it in every detail.
I then ended by explaining how the system currently works.
---
But because I knew my audience would mostly be players who frequented Boards often, I tried to give a speech on how we could all gain from not being rude, even when explaining obvious things.
I hope it clears it up a little.
|
|||||
![]()
Actually, the tribunal suffers from much more severe problems than a computer does.
It was painfully slow and wildly inconsistent. Trolls would just farm IP by not even reading the case, and then clicking the 'punish' button.
There was too many problems, all of which a single uniform system could mitigate.
---
Now, we're just suffering from some new problems, mainly that we don't know how the system actually works. Depending on the complexity and exact algorithm Riot uses, not even they may know exactly how it works, because neural networks processes things in a way that humans have a hard time visualising.
But it's been demonstrably better than the tribunal ever was! Getting instant feedback means that players respond much more. Since this system was implemented, verbal abuse actually dropped a significant amount (According to Riot, around 40% as compared to before the system was in place).
And because the instant feed back is instant, players get a much better idea of what behavior isn't acceptable. And reform rates are much higher than they were with the tribunal.
|
|||||
![]()
Absolutely.
If we can reasonably verify someones intent when banning a teammates champion, then they should be punished. I don't know how such a system would even work (it would require a ton of planning), but I'm all for punishing trolls.
In your specific case, I'd agree that a punishment should be in place. Sadly, I don't make the rules :(
|
|||||
![]()
Just a heads up, we aren't at all affiliated with Riot. Rujitra has the specialist tag because they are very knowledgable about the system, but they aren't working for Riot in any capacity.
This is just a public discussion forum. Quite often people will post here asking for help, and help is often given by volunteers.
But we can't access any information about you, that you haven't given to us. And we can't contact Riot for you, nor contact them in any way differently from what you're able to.
> I'd like a real support representative as the original comments suggests.
But that just isn't withing our capabilites, sorry. We are just random players, just like you. If we had the powers to summon a Rioter, we would.
But I understand that you would still like to contact Riot. As others have said, the only way to do so, is through [Riot Support](https://support.riotgames.com/hc/en-us/requests/new).
But please don't begin to get rude towards us because we can't resolve your issue automatically. We are trying to help you :)
|
|||||
![]()
I'll say the same as the others:
It's a *reeeaaaallly* dick move, but it's not punishable to ban your teammates champ.
|
|||||
![]()
> Also wondering why Mods haven't removed this for "encouraging abuse of the IFS".
I'm not encouraging abuse of it. The post actually explains how the IFS *can't* be abused by typing irregular letters. It will still catch people for toxicity because it's smarter than a simple list! :)
|
|||||
Rioter Comments | new Comments | Views | |||
![]()
Doesn't the feedback specifically state that it was a report that triggered a punishment?
|
|||||
![]()
I don't think it would be abuseable.
I mean, if you see someone with honor level 0 or 1, you can assume that they are more toxic than the average player and possibly encourage them to be toxic. But then again, you'd just be shooting yourself in the foot at that point.
|
|||||
![]()
> Smurfing is bannable the hell are you talking about? Intentionally grieving the game? Thats like rule 1 for a ban.
It's not. There's not a single rule against smurfing. Perhaps you're confused on what smurfing is? Show me a single piece of evidence. A rule, an official statement from Riot. An example of someone banned for smurfing (an example I can verify as being truthful).
|
|||||
![]()
IMHO, you were the most toxic player in those chatlogs.
Smurfing isn't bannable, because they don't 'ruin the ladder'. The chances of you *actually* playing with a smurf is quite low.
Also because tracking smurfs is almost 100% impossible.
And even when you do play against a smurf, if you can't beat them, then what do you think will happen when you climb? If you climb, you'll still be playing against more difficult players.
In that chatlog you were just being an asshole towards your teammates and you were blaming them for losing the game. Meanwhile, you were 0/8. Arguably, you did the worst on your team and yet you're *still* blaming your teammates for your loss.
I'm just stating my honest opinion here.
|
|||||
![]()
I think what Periscope *meant* was that they didn't have to worry about getting punished. In turn, they then shouldn't worry about the reports either.
|
|||||
![]()
You make a ticket with [Riot Support](https://support.riotgames.com/hc/en-us)
However, they'll investigate, but they wont tell you the outcome of the investigation.
|
|||||
![]()
Ya' know, it's comments like that that makes us want to actively kill the napkin industry.
Yeah. We do it out of spite for you guys, that apparently just love napkins.
|
|||||
![]()
Kind of a mixed answer.
I'll start by saying that you should both learn to jungle anyways. Even if you don't like jungling, you'll learn a lot that you will gain from when playing any other lane aswell.
Secondly, if this scenario ever occurs, definently negotiate with your teammates. Explain that you both really suck at jungling, and that you'd much rather switch with one of your teammates. Often, someone will switch their roles.
But whether or not it's bannable is tough to answer. Stealing someone elses designated role is definently bannable, but doing something completely different than what your designated role usually does, isn't.
For example, there was an infamous case of a player going support Singed with smite. Instead of actually helping out his ADC, he just went around the enemy jungle, stealing camps. Now, he legitimately went this role. He wasn't inting in any capacity, and his strategy actually worked. But he still got punished. Riot stated that his punishment was because he didn't communicate this strategy with his teams. He never said anything, he just left his ADC to 2v1 in his bot-lane. This means that the ADC goes into a game completely unprepared and unknowing, which is a terrible experience to have.
The Singed player agreed to explain his strategy and sometimes not play it, if the game really didn't call for it. In turn, Riot wont punish him for playing his off-meta role.
In short, the scenario you proposed would be VERY rare. One of you should at the very least take Smite, as you'll be at a significant disadvantage without it. Explain to your team what's happening, and if they don't agree to switch, then by all means, play duo top lane.
It's just important that you atleast prepare your teammates for the unorthodox game ahead, but going off-meta, even like this, isn't bannable in and of itself.
|
|||||
![]()
My point is that lying about something to this would just be illogical.
> League player base is pretty massive and we both know many players abuse the report system to vent their frustrations.
And preventing this optimally is the exact purpose of this system. You cannot abuse a system that performs a full review on a player once reported, even if only once.
I don't know about any streamers not getting banned, but it sure isn't on topic.
There's nothing really to gain from the discussion. I won't change my mind because I choose to say that the official word is better than anything we can try to imagine ourselves. You wont change your mind because you believe the opposite. Nothing is fundamentally wrong in either train of thought, but we just wont gain anything :)
|
|||||
![]()
> But then again when the official has no answer the unofficial does become a bit more reliable.
But the official *does* have an answer. The answer is that multiple reports don't stack.
Assuming Riot has lied about his, is assuming that Riot doesn't want to treat players fairly. Both report weights and report stacking are unfair mechanism when it only takes the system 15 minutes to review a report.
|
|||||
![]()
> Unofficially my friends and I notice reports seem to have more weight when done by multiple players not in a premade, and by players who rarely use the report function.
Riot has specifically stated that those exact things aren't real.
Report weight *used* to be a thing, but it's not anymore. Multiple reports done by multiple players don't increase the likelyhood of a punishment. That's because a single report triggers a full review. Having additional reports trigger the exact same review doesn't make any sense, as the outcome would just be the same.
|
|||||
![]()
> It has reddit and LOL boards with non-stop posts about what the customer base is looking for.
Sorry to say it, but that's just not true. They have Boards and Reddit to say what Boards and Reddit wants for the game.
For example, if Riot listened to Boards and Reddit, we would never have had the different Jungle plants, but they've been a great addition.
Boards and Reddit represent a vocal minority, and you should under no circumstances blindly use them for gauging what the playerbase wants.
Another problem with this is that while people might say that they want something, but they really don't. For example a couple of years ago, the vocal part of the gaming community desperately screamed for more diverse games. But in reality, sequels and things like that sells MUCH better than a completely new concept.
Riot has MUCH better tools for these things. Through surveys at ends of game and statistical analysis they can get a much better idea of what people want.
---
In my opinion, the real problem was that Riot has almost exclusively catered to existing players, whilst ignoring new players.
But while that might keep old players playing, it wont keep them paying. When an old player already has all the skins, champions and rune-pages they want, they are just going to stop spending. Because of the sheer amount of content in League, it has taken a long time for this to become a problem. But now they have to really pump out new content constantly, just to keep up.
---
Lastly:
> Without customers playing your game, HOW ARE YOU GOING TO FIND SPONSORS FOR YOUR E-SPORT!?!?!
Just like there are many people who don't actually play football, but still watch it, A LOT of players don't play League, but still watches the pro-scene. Sponsors don't care about the number of players playing the game, they care about how many people will watch their ads.
|
|||||
![]()
> Most of my chat wasnt negative
I guess that's where we disagree.
At this point, nothing I say can convince you that you were being negative. But seeing the responses to the thread, and the fact that you got the punishment must surely mean that you can atleast admit that others also found you negative, right?
|
|||||
![]()
> What did the support tickets complain about? That's interesting.
Whilst reading up on it, I actually came to an interesting discovery
https://old.reddit.com/r/leagueoflegends/comments/369a4k/reporting_feels_so_useless/crbz8bh/
> If more tiers of punishments improved player behavior, we would have done it. We see that about 75% of players tend to improve their behaviors after an initial verbal warning, and the rates of improvement just plummet after that. We never saw a benefit of using 1 day, 3 day, 7 day, 14 day bans and chat restrictions and ranked restrictions, so that's why you don't see as many shorter bans--they don't improve anything beyond verbal warnings while longer bans do.
That's a comment from a Rioter (probably Riot Lyte, a previous head of the player behaviour team at Riot)
So, even if they got spammed with tickets, they shouldn't have removed the warnings, unless there was other reasons.
I found the original article where I read about the warnings:
https://en.volu-eu.org/pbkb_general_user_view.php?art=3
(By the way, this website is great for knowledge on the system. It's not always completely up to date, but most of the things *I* know, come from there)
|
|||||
![]()
I'm not sure what you're replying to?
Perhaps
> They said they'd never be toxic again. That their chat could just be muted.
This was a bit confusing?
I didn't mean that they got perma-chat muted.
They don't do that because in the past, when players got really long chat restrictions, they were increasingly more likely to just begin intentionally feeding instead.
Seeing as intentionally feeding is both more severe and a lot harder to catch than chat-related toxicity, it's not worth the risk to give players really long (or permanent) chat restrictions.
|
|||||
![]()
> I know flamming isnt good and thats the reason I got perma banned, but I think everyone desirves a 2nd chance since people waste time and money to get just banned instead of getting some kind of approach to help them not to flame.
I totally agree, but you have to realise that you've already HAD your second chance.
And Riot has tried giving even more chances in the past, and it has never gone well..
The punishment ladder goes like this:
10 game chat restrict
25 game chat restrict
14 day ban
Permanent ban
You can skip levels if you're extremely toxic, but the intention is that you'll never have recieve a permaban without first receiving a 14-day ban.
When you recieve the 14 day ban, it states that the next punishment will be a permanent suspension. It warns you, but you don't heed the warning. So in reality, you've already wasted your second chance.
That being said, Riot has experimented with giving players their accounts back. Sorry to say it, but when they initially did it, 95% of players got banned again. They did an experiment recently, but we don't have the exact stats of it though. News are that it didn't go well, sorry.
And these players were the exact same as you. They said they'd never be toxic again. That their chat could just be muted. They'd do anything. And when they got what they wanted, they just got banned again.
---
So you might understand now why you probably wont be recieving your account back :/
I'm not saying that you'll definently just get banned again, but you're statistically likely to.
|
|||||
![]()
And then when you recieve the punishment, you know you've been toxic.
You shouldn't recieve any more than a chat restriction for the first punishment, unless it's extremely severe.
As you said, we're big kids. We shouldn't need warnings to not be toxic.
---
That said, Riot actually HAD warnings in place once, but removed them because they were confusing players. They were being flooded with support tickets.
|
|||||
![]()
> Here's my question: Why does the millions upon millions of dollars Riot makes every single month not provide REAL people handling reports and complaints?
I calculated this a while ago. It was just taking in statistics for trolling though. If we combine it with having to analyze chat the numbers will ofcourse be different. There will be more cases to analyze, but each case may only take half the time.
> But then, how many people would you actually need to do that? LeagueOfGraphs.com has analysed 6 million matches in the last two days. So ~3 million matches per day. If we assume that in around 10% of these games, someone is reported for trolling that would make 300k matches each day.
> Now how long does analysing, judging and dealing punishment take? Generously, atleast 20 minutes on average per case. Some matches it will be a quick decision, others not so much. So now we're up to 6 million minutes of work, each day.
> That's:
> 100.000 hours.
> 4.166 days.
> 595 weeks
> 134 months
> 11 years.
> 11 years of work, all to be done in a single work day.
That's too much. Even if you *could* pay for that many employees, they'll all still need training. They'll need exact guidelines to be semi-consistent.
But they'll still not be consistent. They'll all have slightly different opinions of what should be punished, and what shouldn't.
---
It doesn't matter who started it and what not. If someone is flaming you, you mute and report them.
In real life, 'defending' yourself may work because it'll deter whoever is attacking you. But in League, not so much. Defending yourself will 100% of the time just lead to MORE toxicity.
> Why is this like elementary school? where it doesn't matter who starts it fighting is fighting?
Because it's only elementary school kids that think saying: "they started it!", is a valid excuse for their actions.
|
|||||
![]()
I know you apparently want 'exact lines' that are toxic, but that's just not how it works.
If I had to do that, I would have to copy paste almost the entire chat.
You were constantly spreading negativity and whining about your team. Constantly blaming others.
Examples would be when you complain that someone isn't helping, or saying 'Enjoy the reports'.
That's just examples though, as almost the entire chatlog is filled to the brim with negative comments.
---
All in all, I'd say that this is a great example of what is worthy of a chat-restriction. You're being overly annoying and whining the entire game.
|
|||||
![]()
> why is Riot NEVER doing anything about chat unless it's "homophobic" or telling someone to go commit suicide anymore?
They are, but you might not be seeing the results directly.
The popup that comes when someone's been punished because of your reports, only shows rarely. Even more rarely when it's not severe toxicity such as the things you mentioned like hatespeech.
A single instance of flaming and blaming just rarely gets anyone punished. The system is lenient enough that people can have bad days, and you should also consider that. People most likely aren't flaming you because of personal reasons, but because they themselves are having a hard time.
Of course this doesn't excuse their behaviour. If someone is consistently toxic over multiple games, then they'll get hit with punishments. Punishments that escalate untill an eventual permaban.
|
|||||
![]()
I'm not sure how two different accounts found this thread on the same day, but I'll tell you the same thing as I told the other guy.
Necroing threads is against the rules, you should start a new one if you wish to continue discussion, and I advise that you do. What you've discovered is very interesting. It's not 100% definitive proof that Riot is directly tracking smurfs, but it's absolutely an indicator that something more is going on.
As I said, I won't continue discussion about this topic though.
|