M.E.T.A., Trolling, and the difference.

The Ecdysiast··1 votes·423 views
This is to inform you of Riot's stance on the topic. This is not my opinion. This is the fact of what counts as trolling.

#OFF-META IS NOT REPORTABLE, BUT TROLLING IS NOT OFF-META

For those not aware, the term meta refers only to what is currently (read "during this patch") the strongest option. It refers to both champions and items, and can very seldom refer to roles.


#Meta

For example, currently according to op.gg Kog'maw is the best ADC at Diamond level, with a win rate of 53%. And his meta build is on-hit Runaan's + Guinsoo. Kog'maw currently meta, and he most effective with that build.

Conversely, the worst ADC this patch is Jhin. This means Jhin is currently off-meta, but his most effective build is Youmuu's>Duskblade>IE. Which means that your best bet is to build that way, but even when you build that way your chances are lower than Kog'maw's because that build is only going to put you on par to achieve his 46% winrate odds against a team of equal skill.


#Trolling Now, what does this have to do with trolling? Well, some people who troll pick like to defend it as "off-meta", which is simply ignorant of the term. As I explained with Jhin, off-meta only refers to a champion that is not currently the best option, with their highest-performing build and role performing worse than the others choices.

It does not refer to champions being picked for roles and/or with builds that their stats and scaling are not designed for.

For example, Gangplank is currently the best performing top laner with his standard Trinity build. However, just because Gangplank is the best when built correctly does not mean he retains that performance when built full AP (or 6 boots) in the bottom lane. Doing so ventures out of the realm of off-meta and into trolling. Of course, the order of your items and your finished build will change depending on how the match is going, but building an AD-only champion like, say, Riven with full AP items is inexcusable for anyone past level 20.


In conclusion, off-meta does not mean the same thing as a troll pick. Playing to win, if it isn't as blatantly obvious as I know it is, includes champion select and your build. It's perfectly fine to pick an off-meta champion if you feel you're better with or simply want to practice to get better with that champion because their style suites you. What's not okay, and I mean objectively not okay based on the rules presented in the Summoner's Code, is to troll by doing something like picking Jhin for ADC but building full AP.

If you're curious as to how that's trolling, I'll refer you to two other cases that set the precedent for why this goes against the Code. In both cases, and in any similar case of troll-picking/building, the issue is not the build itself. It's in the lack of communication that you intend to play a different role than what the champion is best at. In these examples, both involving the "Support" not communicating to their team that they were instead going to bring smite and take the enemy jungle rather than stay in lane and actually be a support-- which is akin to stealing the role of the jungler. And now with the new champion select, such things simply look like you're angry you were autofilled. https://boards.na.leagueoflegends.com/en/c/player-behavior-moderation/79aGu3wR-support-singed-isnt-banable-but-thats-not-the-point https://boards.na.leagueoflegends.com/en/c/player-behavior-moderation/FrILlWp7-14-day-ban-because-playing-nunu-support-with-smite-is-stealing-other-peoples-roles For my examples, AP Gangplank or Riven isn't just role-stealing. It's closer to leaving, since there aren't roles that Riven or Gangplank fulfill when they're built full AP.


Edit: I'd like to point out one particular that a comment mentioned, and refute it.

He claimed that meta is popularity based, rather than effectiveness based. I argue that that's flipping around the cause and effect. A champion does not become meta after being popular. A champion will however become more popular if it's found that new changes have made them meta.

LeeSin For an example of a champion who's almost always both popular and meta we have Lee Sin. He has been consistently strong and fun to play ever since his release, especially considering his mobility given the state of mobility in League at the time. And as we'd expect with the meta/popularity model that I propose, his popularity declined when he wasn't meta--- most notably during the lifespan of Sated Devourer.

Nunu Then for a current example of a champion becoming meta and gaining popularity, we have this patch's Nunu, who jumped up from last place in jungle pick rate to 13th because his new changes made him the number one in effectiveness. So good, in fact, that his "counters" are still losing to him more often than winning, while champions he counters are losing to him 64% of the time.

AurelionSol And finally, as an example of a champion who is meta but not popular, we have this patch's Aurelion Sol. He may be the best performing, but his kit is still unfun, leaving him less popular despite his strength.

13 Comments

TrulyBland2 votes

I don't disagree with the point you are making in principle, but your choice of words really doesn't help people who like to pick off-meta. Specifically your definition of "meta" is the problem.

Meta is not an acronym and even ignoring the actual definition of metagame and instead going by what people commonly refer to as the meta in LoL it isn't anywhere near "the best option". The meta refers to the most popular strategy, and while most of the time that strategy will at least be among the best, there have been plenty of cases in the history of LoL where the meta only prevailed because people were unaware of a better strategy, or because other strategies (even if more effective in niche cases) were not as good as "default" strategies. There were even a few cases where very strong, sometimes even overpowered, champions simply didn't feel strong enough for people to actually pick them up.

tl;dr: Picking off-meta is neither trolling nor "not going with the best option". It is simply not the most popular strategy, nothing more and nothing less.

The Ecdysiast1 votes

Popularity has nothing to do with it. In fact, the meta as it is actually defined can change the popularity of a particular champion or strategy. Not the other way around.

As for evidence of this point, I'll give you this patch's Aurelion Sol. He is, according to op.gg, the best mid laner. He is currently meta. However, contrary to your loose definition, he is still not a popular pick. And it's because as you said, he doesn't "feel" strong due to his kit's style.

Refer to the links I gave, and the Rioter comments in them to find the evidence of my main point.

TrulyBland3 votes

the meta as it is actually defined

I'm sorry... but the moment you wrote out meta as an abbreviation you essentially proved that you have no idea what it means. But I have better things to do than to argue with somebody who believes circular logic is "evidence".

Feel free to use your meaningless definition of "off-meta" that literally excludes everything but the best strategy, best picks and best builds, and for the same reason is impossible to define seeing as there have been plenty of times when the community didn't even know what the best choices were to begin with.

MasterDClone2 votes

I have a few issues with what you are saying because i play "off-meta" a lot but you're version of it is very narrow. Off-meta seems to be very opinion based because i see it as anything that isn't the current meta as long as you are still tying to win in game.

  1. playing Ap- jhin works i would take my bros ap jhin over an adc main anyday of the week he consistently goes over 20 kills per game while still playing the adc role would that be trolling when in every second ranked game my adc is 0/10 playing kog maws and vaynes and nothing i can do because they are doing nothing wrong.
  2. There is no rule saying you have to communicate with your team too and usually when you play things that others consider trolling talking in chat works against you
  3. Just because sombody is playing something you dont like it doesnt make it trolling
The Ecdysiast1 votes

Your points are moot. I gave you the objective facts, since this is not opinion based.

And I gave you the examples to prove this point. It doesn't matter if you don't think it's trolling. Riot Support does. It's kind of like how Riot support doesn't care if your feelings are actually hurt or you're abusing the report system, a hate speech report with a word they deem valid will still end in a ban.

I didn't make this thread to make people who like to troll salty. I did it so they don't get punished for breaking the rules unknowingly.

MasterDClone1 votes

Your points are moot. I gave you the objective facts, since this is not opinion based.

But it is opinion based since everyone has a different opinion on what is considered trolling

And I gave you the examples to prove this point. It doesn't matter if you don't think it's trolling. Riot Support does.

where does it say that riot support thinks that, riot encourages us to experiment new builds and ideas as long as it doesn't ruin the game experience for others

I didn't make this thread to make people who like to troll salty. I did it so they don't get punished for breaking the rules unknowingly. That was the point i was saying, is that those who troll are those who play to lose the game not those who play off-meta you can still pick a meta-pick and build but be a troll, you're telling me that you don't want people to troll to get punish but the peolpe you are reffering to aren't trolling they are just playing a different style than the current meta and we aren't breaking any rules either

Umbral Regent1 votes

I'm not sure how much I agree with you - I wanna say it's in the upper range of 60-70%, though. I'd love to discuss the concepts of meta you brought up, but, I believe that's better suited for Gameplay, so, I'll withhold it.

But, I will agree with some notes - namely, blatant troll builds like AP Riven, and the issue of communication regarding off-meta strategy. It's one thing to take an off-meta pick and communicate the barest of it with your team - it's another thing altogether to pull an off-meta strategy that hinges on communication without saying anything.

The Ecdysiast1 votes

We can discuss it here. It is central to my point, after all.

Umbral Regent1 votes

Well, TrulyBland put it pretty well, actually.

Meta (to me) is partially about popularity and also primarily about what works - winrate has hardly anything to do with it.

Like, as an example, let's use bottom lane - you say Jhin isn't a meta bottom laner, but that isn't true. The bottom lane meta is primarily Marksmen - with some few Artillery Mages (hi Ezreal, you count for this). But, there are Champions who fit this meta depiction who don't actually fall into the bottom lane meta - like Kindred. They're Marksmen, but their short range and their Mark of the Kindred make them predisposed to the Jungle more than the Bottom Lane.

Besides, if you're using winrate as the sole, defining metric for meta, especially the winrate of high tier players (Diamond and up), you're effectively calling quite a number of Champions (basically anyone who doesn't see play in Diamond or higher) non-meta picks - and that just doesn't make any sense.

Here's a real example of non-meta; Yasuo Bottom Lane. He's not a ranged Champion, so he has clearly decreased effectiveness on the bottom lane, but his ability to farm and reliably deal with Marksmen and other Ranged Champions (even some melee) make him a viable pick on Bottom lane, provided you know what you're doing.

I personally view meta as "common practice". Some things can't be questioned or done differently (like buying Triforce on Gangplank), and others are generally malleable (you could rush Triforce or Titanic Hydra on Yorick and both would work), and others are just differences in build (Crit Jhin, Lethality Jhin - both are meta, just one is more popular.)

But things like Rengar Top Lane, Bard Jungle, Yasuo Bottom, or Garen Mid, those are non-meta. They're viable in their own respects, but they're not common practice.

The Ecdysiast1 votes

Ah, I was afraid you'd specifically agree with that. And as I said to another comment, this is not a matter of opinion. It's a matter if ignorance of the terms. I'm not using "my" definition. If you click the links in the OP, you'll see that I'm using what is considered trolling by Riot Support. This thread is not to convince people of an idea, it's to inform them of a truth.

Refer to my edit of the OP for my points on popularity.

But as for your other points, a pick like Yasuo bot or MF support is just as meta as a specific counter-pick. If the enemy picks Darius, then the most effective choice is Wukong regardless of the fact Gangplank is the best top laner. However, for a choice like Yasuo bot or MF support you need to be certain that you're actually in a situation where you've countered the enemy. Otherwise when you feed, you've trolled just as if you'd picked crit Vel'koz as your ADC.

Jhin is not currently in the meta. He's the bottom tier ADC. But non-meta is not reportable. I was pointing out that non-meta is not trolling, but people like to pretend trolling is non-meta.

As for your final bit about some random picks being non-meta. That's true. I never once said that non-meta means something isn't viable. And for such cases it is exactly the same as before; if it's a counter pick, it's meta in that situation.

But Riot's stance on, say, Nunu "support" with smite, is that it's trolling. Not specifically because it isn't meta, but because it unfairly forces the rest of your team to play around your whims. As does something like Bard jungle-- most likely forcing your team to play extremely safe due to the enemy jungler's better clearing times.

Now, this is all different from my opinion on the matter. If I do not feel I have been negatively impacted by an ally's decisions (i.e. we won regardless of picks), then I will not report. What I wanted people to take away from this thread is that just because player opinions differ on when something is worth a report, your punishment is still justified if they do.

sleepysithlord1 votes

I dont see the difference between (in ranked) Jhin adc and AP riven. I'm either goint to tilt off the face of the earth, open and ff at 15, or dodge if I see either of those picks in champ select. Actually, no, the jhin adc is less troll because I can dodge that. I can't dodge the AP riven or GP becasue I won't know they are going AP until the game starts

The Ecdysiast1 votes

I kind of outlined the difference in the OP, but I'll make it clearer:

Jhin ADC has roughly a 46% chance of winning.

AP Riven is akin to, if not worse than, having an AFK because an AFK player can't feed. According to a 2015 Reddit post I found, that's about a 5% chance of winning. Though considering the fact the game's only gotten more snowball-oriented, I'd wager it's closer to 1%.

sleepysithlord1 votes

At the end of the day, both picks are dead weight

More from Player Behavior

More by The Ecdysiast