The Metagame - Centralization, Overcentralization, Diversity, and related topics.

[gorm]

wow it's the best topic in the history of stark =)

I just want to say firstly i agree with most everything in this thread,
and secondly we should start establishing a parameter of "maximum number of threatening strategies a team can "handle"." to get an "upper bound" on the size of ou.

this would not really be easy but we have stuff in the works, namely x-acts "how much does this set do to 'the metagame' " applet. were we to combine it with a "how much damage does this team take from 'the metagame' " we could get a decent idea of what makes a team "successful" and what doesnt.

unfortunately there' a zillion more factors than "on average how much does this move do based on weighted usage" but i think it's start if we want to identify "ideal diversity"

 

[cim]

"Counter Mentality" is a close second but that's gone :(

Just posting to agree with the jist of the OP. We don't know what overcentralized is, everything centralizes to some extent, etc. should be ingrained in people's minds.

 

[Hipmonlee]

When you say we have no idea what overcentralised is, who is the we? Do you mean we havent agreed, or do you mean that none of us individually have a concept of what it means to them. Because I have a concept of what it means to me.

I mean, I played advance, and I know that it was never as centred around one pokemon as DP was around Garchomp. When you read this thread, pretty much everyone agrees Garchomp was the best pokemon in OU. Is there such clarity now? Does anyone really think that Scizor is as dominating as Garchomp was? Or should that be Heatran or Salamence? Maybe Gengar, which was forever the runner up in useage to Garchomp but has suddenly slipped to #10? Do you think that Garchomp would have plummetted in useage with the addition of Platinum tutors?

The concept of "maximum number of threatening strategies a team can "handle"." seems bizarre to me, I think you need to explain this more, but I seriously doubt it is anything legitimate. Also why would it give us an upper bound to the size of OU? It would give us an upper bound to the number of threatening strategies.

Also I dont like MoP's post. Firstly it is very hard to follow, but it really seems to suggest that if Rayquaza had been OU for some reason he would support it remaining OU (I know MoP pretty much has said he would agree with that). I always played offense in DP, and I promise the list of pokemon that can revenge kill Garchomp is very small. You can sacrifice all day (well 6 times) against Scarfchomp, but actually killing it is really fucking hard.

Have a nice day.

 

[gorm]

When you say we have no idea what overcentralised is, who is the we? Do you mean we havent agreed, or do you mean that none of us individually have a concept of what it means to them. Because I have a concept of what it means to me.

probably the former. we all do have our own idea of what we want, and it seems the more we talk about it the more we agree, we like a certain amount of "diversity" in our metagame.

The concept of "maximum number of threatening strategies a team can "handle"." seems bizarre to me, I think you need to explain this more, but I seriously doubt it is anything legitimate. Also why would it give us an upper bound to the size of OU? It would give us an upper bound to the number of threatening strategies.

well mathematically when we all agree that "more diversity is usually good" it seems like a good idea to define "but how much is too much" when we're looking for a "good size" for ou.

in my head, this would be a measure of "how much stuff we can stop with a single team". i.e. if we can measure out on average how many pokemon/strategies are "dealt with" by a team, we can correlate that to how many threats people "like dealing with" on average.

overall i'd say there's a pretty huge amount of strategies (compared with the max amount of "strategies" within a single team) that a team should be able to hold up against, or else it's a weak team. there isn't a useful way to measure "how well a team counters strategies" but if we can measure "how much damage it does/takes" that's a pretty good first step in measuring "maximum team potency" or whatever.

next we would use some ladder score weighted ("avg damage done" over "avg damage taken") function, a sort of "average team potency". if we get significant enough numbers, we can measure "how many threats can a decent team take on" and go "let's base the size of ou on that, and ban things that lower the average potency of a team".

i.e. in garchomp times a ton of anti chomp measures had to be taken to not get completely owned, and in deoe times it pretty much beat every lead and you had to go pretty unconventional (less potent, or more gimmicky as tangerine defines it") to beat it.

these are just examples and i'm not really claiming to be able to take into account all metagame trends, they just "fit" into my model (unlike something like punishability index or whatever which measures stuff based on the amount of turns it takes o_o)

tldr; i'm pushing for team size to be used as a parameter in our definition of overcentralized since teambuilding is the reason there are useage stats and a metagame, and effective teams/strats are popular teams/strats until they are tided over, or banned. we have relevant statistics so we can see if this kind of theorizing is conclusive.


again this is pure hypothesis/statistical theorizing, i just seriously think that measuring average team potency gives us a good shot at figuring out "what we collectively want"

 

[Tangerine]

When you say we have no idea what overcentralised is, who is the we? Do you mean we havent agreed, or do you mean that none of us individually have a concept of what it means to them. Because I have a concept of what it means to me.

we as a community. The point is that we all have a different concept of it so the definition is hazy. I mean I stated my definition a million times and either no one gave a damn or people disagreed. (FYI my definition of overcentralizing is that a trend is overcentralizing if the presence of large amount of counter trends does decrease the usage of the trend or at least "keep it in check" relative to the other Pokemon)

The concept of "maximum number of threatening strategies a team can "handle"." seems bizarre to me, I think you need to explain this more, but I seriously doubt it is anything legitimate. Also why would it give us an upper bound to the size of OU? It would give us an upper bound to the number of threatening strategies.

The analogy I have is with the forum mafia games you guys play. Of course, it's not a perfect analogy but I'll try to make the point anyway. Now, every player has a specific win condition and in Mafia it is possible to win together with a group of neutrals. Now, imagine less mafia/villagers/whatever third team and more neutrals, except neutrals still have their win conditions and also have a "you win only if no one else wins" condition. Winning would literally come down to luck and it'd be a pretty dumb game to play.

Similarly with Pokemon. If there are too many viable strategies then the game becomes rock papers scissors since if you're unprepared for team X you will lose to it a good proportion of the time you face it. That will be the situation with "too many viable strategies", although now I think such a condition may be impossible considering the limited mechanics of the game (meaning I'm not sure if this hypothetical situation actually exists in this game). But the point is that "maximizing number of strategies" isn't our goal, neither is "decentralization" but to keep a "reasonable number of strategies".

Finally, I linked to MoP's post because I think his attitude within the game is an approach every competitive player should consider when playing the game instead of immediately yelling out "UBER".

 

[X-Act]

Funny how we now have a good measure of diversity.

But having a good measure of diversity is one thing. Commenting on changes in diversity from one month to the next is another!

 

[Veteran In Love]

I dunno about everyone else, but isn't anyone else getting tired of discussing redundant topics like these over and over again

 

[umbarsc]

This is my favorite line in the thread.

Now Colin goes on about the impracticality of looking for the "ideal" metagame, so that we should give up, settle, and leave at be. I'm perfectly fine with this idea-- because I think playing the game (and having fun) is a lot more important than bitching over these types of details.

I definitely agree with ChouToshio here (if I'm interpreting him right), ban enough so that we have "enough" decentralization and settle at that, then we can actually play the game and not have to care about what's a good metagame or centralization or anything. Just lay out an iron set of bans and let the players make it fun for themselves instead of stirring up new bans/tests to make it interesting for them.

 

[Imran]

I see what you mean Umbarsc, but doesn't that go against the entire point of this thread? The majority of the discussion in the OP is not "what we want in a metagame?" but "where do we draw the line?"

Saying we need "enough" decentralisation just doesn't cut it simply because we cannot quantativley or qualativley explain what that means. X-Acts diversity topic is the "closest" we have got to that, but even if we do have "values" for decentralisation people will never agree on what is "enough." Sure someone can say "right, that is it, get on with it" but who really has the authority to say that?

 

[Tangerine]

I see what you mean Umbarsc, but doesn't that go against the entire point of this thread? The majority of the discussion in the OP is not "what we want in a metagame?" but "where do we draw the line?"

No, the point of the OP was

In the end I feel as if everything just boils down to this - that overcentralization is a very arbitrary definition that has the potential to mean everything to absolutely nothing - and the overcentralization is defined based on what people want in the metagame - people who are fine with it and people who want "more variety".

"what we want in a metagame" *is* "where we draw the line".

Saying we need "enough" decentralisation just doesn't cut it simply because we cannot quantativley or qualativley explain what that means. X-Acts diversity topic is the "closest" we have got to that, but even if we do have "values" for decentralisation people will never agree on what is "enough." Sure someone can say "right, that is it, get on with it" but who really has the authority to say that?

"The community", obviously. This is why we have the voting scheme.

 

[Imran]

Yes, the community as a whole makes the descision but that doesn't mean that some individual people will disagree with certain additions / changes. I think in essence you summed up what I wanted to say by mentioning your first point, I failed to see it that way for some reason, but saying just draw the line and get on with it is something that this thread isn't trying to achieve, as a community we must decide where to draw the line, and that is done through the voting system and discussion in threads like this. A thread like this is trying to understand what "the community" wants, and "just drawing the line" is not something we are looking to do, simply because that will not be a community wide descision.

 

[umbarsc]

I see what you mean Umbarsc, but doesn't that go against the entire point of this thread? The majority of the discussion in the OP is not "what we want in a metagame?" but "where do we draw the line?"

Saying we need "enough" decentralisation just doesn't cut it simply because we cannot quantativley or qualativley explain what that means. X-Acts diversity topic is the "closest" we have got to that, but even if we do have "values" for decentralisation people will never agree on what is "enough." Sure someone can say "right, that is it, get on with it" but who really has the authority to say that?

That's why I feel that discussion about this sort of thing is sort of pointless. No matter what we do, it will be decided arbitrarily and subjectively to some degree, and we can't please everyone obviously, so I think we should just find a point that is arbitrarily and subjectively "good enough" and keep it at that.

 

[Imran]

Should we not strive to do this "as well as possible?" I mean sure there comes a point when we have to stop fiddling around with every little thing in the metagame, but there has still been a lot of fallout from the Skymin vote especially, and it seems important to determine where we draw the line.It did, after all, draw some quite passionate opinions. I mean imagine simply turning around to essentially half the community and telling them that Skymin is definately OU or definately Uber, a significant proportion of people on either side will be unhappy with the descision.

At the very least this thread allows people to have their input on issues like that so even if we dont apply a set quantative value on decentralisation, at least we have given those who are unhappy about descisions the chance to give their opinion and explain to people why they believe what they believe. They have had the chance to sway opinions by giving coherant arguments, I feel that is something we should strive for.

 

[Sebastien Loeb]

The battler (particularly in the metagame DPP) it is a being that tries to be always and however more cunning than the others, in that sense? Simple when banned a Pokémon comes that is too much main point for the proper team, the battler it has two 2 possibilities:
1) to get off a team, and to refer completely afresh it
2) to find another Pokémon that takes the place of the Pokémon banned.
In 60% of the cases the second hypothesis happens, thing volgio to say? Do I immediately explain you him, banned Garchomp? To his place the battler inserts SDance-Lucario, banned Skymin? To his place the battler inserts NastyPilot-Togekiss, done result? The metagame changes few or nothing.

 

[gorm]

it isn't arbitrary and subjective when you have data behind it. would you say election results are arbitrary and subjective because it's just "what this amount of people want"? in the same way usage stats/team stats can help us refine our definition of "what is the ideal amount of diversity"


if a certain amount of people arent happy with what the majority "like dealing with" as far as a diversity level (or a range of diversity levels) uh who cares? people voted dsds ou so it's not like they're infallible. we're looking for what level of diversity the majority wants that we can determine acceptably. that's where we "draw the line"

as far as skymin, ipl voted ou, and frankly i really don't give a crap if ipl wants to ban skymin, he doesn't like dealing with it so it's completely up to him. who am i to tell ipl that he should be able to comfortably deal with skymin? i myself dont mind skymin that much but yeah the community has yet to come to a decision.. but yeah more numbers are always good.