Metagame shifts: change or improvement?

In general, metagame change is driven largely by a cycle of reacting and adapting to the broken set of the month. Basically, a new threat that's almost very hard to deal with in the current metagame shows up, whether because of Nintendo changing the Pokemon themselves, Smogon changing the tiers to shuffle things around, or somebody figures out a tough-to-crack set that can take on most of the major threats at the time. The first one's pretty obvious, and is usually a huge and permanent change to the metagame.

As for the other two? The metagame immediately reacts as half the community puts that set on their team, teams that get completely destroyed by the new set vanish overnight, and everyone else tries to balance their lineup between countering that one set and doing whatever else they want to do. If an efficient counter is found, that one becomes popular until the set it's meant to counter begins to decline in use, causing the metagame to swing into a new equilibrium which is similar to before but with the newly discovered sets added to the long list of things people must consider when building new teams.

If no efficient counter is found without massive changes to all teams, then there's almost immediate outrage as people feel that the metagame has been broken, and that Pokemon is quickly moved up one tier, resulting in a slight change in the metagame as things that dealt with that Pokemon's older and less effective sets decline in use, and some people attempt to adapt the moveset that owned the game to other, weaker Pokemon.

In general, however, it's really difficult to discuss each change individually in a vacuum (except possibly really major ones stemming from Nintendo massively changing things) as I suspect some part of this thread is trying to look at it. First, because the metagame is a huge, constantly changing creature with thousands of sets in use at any given time. Second, even people using established sets often use customized variations - between personal preference and the demands of the team, it's nearly impossible to track the metagame in terms of single changes as people usually do.

For example, while I don't know modern Ubers at all, you mentioned Wish Blissey gaining popularity lately...you blamed that on countering Kyogre, but I think it's part of WishBliss's overall popularity surge (even mentioned in the Blissey strategy page) due to the offensive shift brought on by especially Platinum, along with the fact that it seems like Wish would be a good idea in general to have on an Ubers team to ensure that your beefy beasts are impossible to whittle down once you've taken out their counters.
 
I wish I had seen this thread earlier, as I think it had the potential for really interesting discussion. But here we are with 50% of the posts not even addressing the OP in any meaningful way. frankly, I think someone should have moderated the living shit out of this thread, but whatever I guess.

Even those posts which actually managed to grasp the basic gist of the thread seemed to run into problems, and honestly I can see how it's difficult not to blur the line between the two a little bit. I feel like Abacus did a good job in his post of explaining the basic idea of the thread, but it's still difficult not to get the feeling that he's sort of contrasting two of the same things here. I mean "optimizational" change is still "optimizing" a Pokemon's options based on other Pokemon, so as far as a Pokemon's moveset changing based on (and only on) its own inherent abilities, that's just not happening unless you're talking about "Ember versus Flamethrower" or things of that nature. So I think it's wrong to use terminology like "inherent" to determine that a metagame shift is the result of optimizational rather than cyclical change--I think the more accurate distinction to make between the two is "how 'much' a Pokemon's changes are dependent on the changes of the rest of the metagame." Basically I think that the way Abacus is framing this thread is probably valid, but it makes things sort of subjective to say that a change which is dependent on only one Pokemon is "cyclical," whereas a change resulting from, say, bulky waters as a whole, is "optimization."

I think the Scarf Kyogre example might be really good because it illustrates, accurately or not, that there was pretty much no reason for Kyogre to use Scarf in the first place. Most actual examples just aren't that simple though, I mean if there's a Pokemon that merits Kyogre using Scarf, and people started using Specs because it was useful against Blissey, that's starting to sound more like a "cyclical" change now. I think this sort of shows another interpretation of the thread, where "cyclical change" versus "optimization" is viewed as a matter of whether the catalyst of the change is, or isn't, itself, changing as well. I don't know how valid of an interpretation that is, or if it's even necessarily opposed to Abacus' interpretation. In fact, I don't even know if I'm being coherent right now so I think this is just a good point to stop.

Oh well, either way I thought this thread had a lot of potential so I really wanted to just (hopefully) "get it going." Although a "reset" would also do it wonders, seriously, someone needs to delete like two thirds of these posts lol
 
Overall I agree with what Blame Game says (even about the "moderation" thing^^). I think that cycling change is still cicyling change whether the influencial factor is a Pokémon (for ex: Suicune) or a group of Pokémon (for ex: Bulky Waters). Optimization is true optimization only when, like Blame Game said perfectly, there was no reason to use the old set in place on the newer one. For example, Specs Kyogre was overall better than Scarf Kyogre from the very beginning.

Which brings me to another point. If we do not take into account "new entries" like Platinum move addiction, Latias and so on, I cannot remember of a single, clear "optimization shift" which dates more recently than D/P era. Does this mean that we explored everything worth discovering in the OU metagame? That every further change is a result of "trends" not of - how can I say - "knowledge"?
 
It's difficult to say. I'm kind of hesitant to say that all recent developments in the OU metagame can be classified as "trends," I just think it's a complicated thing to try and classify at all.

It's kind of interesting to think about how other metagames (I heard about a year ago that Lanturn was OU in Japan and Heracross was banned or something?) compare to ours and how some of the differences might be chalked up to "trends" versus "oh well they're just not as good as us."
 
Blame Game said:
It's difficult to say. I'm kind of hesitant to say that all recent developments in the OU metagame can be classified as "trends," I just think it's a complicated thing to try and classify at all.

It's kind of interesting to think about how other metagames (I heard about a year ago that Lanturn was OU in Japan and Heracross was banned or something?) compare to ours and how some of the differences might be chalked up to "trends" versus "oh well they're just not as good as us."
Click to expand...

Well, assuming Japan used the same ruleset and banlist of us (which I don't think so, I also heard you can only use 1 out of a list of Pokémon like Salamence, Dugtrio, Tyranitar, Garchomp etc), if we have discovered "better things" than Japanese people, it would theoretically possible for someone from Smogon to (ab)use this knowledge and take said metagame by storm (to a certain extent). If, for example, we think Lanturn is not OU because it is bad for some reason, it is not that difficult to prove it in practice.
 
Cleaned out this thread.

Anyway, I think you're really making a leap really without hammering down the basics.

The thing about Pokemon is that each Pokemon suits variety of roles. You pick a Pokemon that satisfies all the roles you need.

Optimization is extremely rare in terms of individual Pokemon. When you look at a Pokemon, you have a good idea on what kind of roles it can do. You can, however, find a set of roles that work well relative to the metagame at hand. You can only optimize relative to the metagame at hand. Of course, the biggest thing is that it depends on a lot of factors - particularly your team. This means that there are a lot of usable sets for a Pokemon in a given metagame, all because it is all six Pokemon.

From what I can tell, you call such changes "cyclical". All changes, other than like a small improvement from using Ember to Flamethrower (okay, you can make a really really stretched argument that "it has more PP so it has a different purpose" but that's irrelevant), are cyclical in nature.

I will argue that every change is cyclical, affected by the trend that is going on. There are no objective "optimal" changes that are made often in a metagame, simply people adjusting to new strategies and new trends.

If this is true, then there isn't really a huge advantage for plat teams built after people discovered versus plat teams today - or rather, there's no reason to expect a clear team advantage unless someone has found many new sets to abuse in the meanwhile that the other player cannot deal with - which is very, very, rare.

For example - look at what Theorymon does - he pushes the metagame forward by finding new Pokemon that can fulfill certain roles. No one's going to argue that Delibird is an "optimized" change in Ubers - Delibird just happens to fulfill a very specific role.

In the end, Pokemon has far too much variety to the degree that nearly every change is going to be 'cyclical'. You're not going to find the "most efficient way to use a Pokemon" or "the most efficient Pokemon to use", you're going to look for roles in your team that you need fulfilled because of how other players act, and find a Pokemon best suited for it. It's all relative to the metagame.

If you think of all changes as "trends", you can explain the differences between Japan's metagame and ours (if those statement about japan's metagame are even true), since depending on the starting point they go towards different directions.
 
I think calling "Ember -> Flamethrower" an example of optimization forces us to ask why we can't say the same thing about Brick Break versus Close Combat on Heracross, or something a little more relevant like the bulky water example I remember from earlier (though I don't remember the actual example anymore). All three of these examples are identical in that they're "technically cyclical," but are dependent on factors that are so deeply rooted in the metagame and unlikely to change that I'd argue that each of them has a very strong case for "optimization."

I think you're overcomplicating things by bringing up team roles and whatnot, though. The Delibird example could be (for obvious reasons I'm not going to argue that it definitively "is") optimization if Delibird were better at its specific role than whatever Pokemon used to fill that role on that particular team.

I also think there's some merit to comparing two "possible evolutions" of a metagame. I don't have a hard time believing that it's possible to choose which one might be better, although by "better" I don't necessarily mean "a team from branch A would beat a team from branch B," so much as, "branch A has progressed such that we can consider it more 'advanced' than branch B." I'd like to think that, pretending for a moment that our initial rulesets were identical, we could theoretically determine whether Smogon's metagame or Community X's metagame were "more advanced."

I'm sure there are ways to elaborate on all of these points in a coherent way, but my last attempt at a post got deleted and it's late so I'm sort of left with these half-ideas. whatever, thanks for cleaning the thread up and hopefully more discussion goes on as this is the only Stark thread that interests me at the moment.
 
http://www.smogon.com/forums/showthread.php?t=36192

Whenever we have these discussions/debates, I like to bring up this thread to show how far we have come in understanding the metagame. If Obi so cavalierly brought up maybe having Moltres on a team today (without spin support, either), we'd ask what he had been huffing. Even if you believe that most metagame changes are cyclical, which is probably true, the growth of Stealth Rock is undeniably an optimal change. As a consequence of this, metagame changes derived from the proliferation of Stealth Rock are also optimizational changes. For example, the growth of a Steel-based metagame and the fall of Pokemon such as Weavile from OU play.
 
All three of these examples are identical in that they're "technically cyclical," but are dependent on factors that are so deeply rooted in the metagame and unlikely to change that I'd argue that each of them has a very strong case for "optimization."
Click to expand...
How do you decide what's optimal? In what way? That's my big "gripe" about calling things optimization since everything has a different purpose and different trade off. I don't think we can call any changes optimizational.

the growth of Stealth Rock is undeniably an optimal change
Click to expand...
Sure, that's because Stealth Rock is pretty much the best move in the game - but remember, it does have a "cost", just that what you can get for that one move is ridiculous to the degree that it's "worth it" to have it on every team.

I think you're overcomplicating things by bringing up team roles and whatnot, though. The Delibird example could be (for obvious reasons I'm not going to argue that it definitively "is") optimization if Delibird were better at its specific role than whatever Pokemon used to fill that role on that particular team.
Click to expand...
How can it be optimization when it's the *only* thing Delibird can do? You're not optimizing since Delibird is useless other than the specific role while other Pokemon, while being less effective, aren't complete dead weight past that one specific role.

I also think there's some merit to comparing two "possible evolutions" of a metagame. I don't have a hard time believing that it's possible to choose which one might be better, although by "better" I don't necessarily mean "a team from branch A would beat a team from branch B," so much as, "branch A has progressed such that we can consider it more 'advanced' than branch B." I'd like to think that, pretending for a moment that our initial rulesets were identical, we could theoretically determine whether Smogon's metagame or Community X's metagame were "more advanced."
Click to expand...
How would you measure that? If they're COMPLETELY different branches you can't really compare it - you can take a a half way evolved metagame and then compare it, but you're assuming that they're going to reach one equilibrium in the end (which is not true)
 
Tangerine said:
Sure, that's because Stealth Rock is pretty much the best move in the game - but remember, it does have a "cost", just that what you can get for that one move is ridiculous to the degree that it's "worth it" to have it on every team.
Click to expand...

Well, yes. Would you agree that changes resulting from the ubiquity of Stealth Rock are also optimizational?

~Weavile's fall from grace.
~General loss of Flying-type Pokemon from OU bar high-powered sweepers.
~Proliferation of Steel-types and consequent strengthening of Fire/Fighting/Ground attacks.
~etc.

But it is difficult to pin down such changes as optimizational, because Platinum and the Suspect tests really change things around.
 
If something is optimal, then this is *the best*, no contest about it, use of resources. How can you argue ANY of those changes are optimal?

Pokemon is such a varied game with many different ways to approach it that calling something optimal is very, very, hard to pin down. It's silly to think about it in a huge majority of the cases.

I would 100% disagree the changes resulting from the ubiquity of SR are optimal, but rather, cyclical. The list of changes you put down is a result of OTHER factors and not just Stealth Rock, really.
 
Tangerine said:
How can it be optimization when it's the *only* thing Delibird can do? You're not optimizing since Delibird is useless other than the specific role while other Pokemon, while being less effective, aren't complete dead weight past that one specific role.
Click to expand...
You're not optimizing in the sense that "independent of other factors, people who use Delibird are more likely to succeed than people who do not use Delibird," but you're optimizing teams which operate more successfully with Delibird than without, and you're optimizing Delibird itself.

Your gripe about everything having a different trade-off comes into play here. It's extremely difficult to say that Delibird's introduction into the Ubers metagame was "optimizational" when you look at it from the standpoint of "Delibird is otherwise useless anyway; I'll just use some other Pokemon/team and be just as successful if not moreso." But I can look at a metagame that uses Delibird, and even if it's only used in a very particular team setup, if it's always better to use Delibird in that setup, that's optimization. It might not necessarily be optimal to use that team, but if "if I'm using that team, I'm using Delibird" is true, then that's "a kind of" optimization.

That's why I don't think the idea of "measuring two completely different branches" is that unreasonable. Obviously a ton of differences between the two metagames are going to be completely cyclical no matter how you slice it, so you can't just have branch A players play branch B players and declare some sort of "winner." But sometimes there are forces that just aren't going to differ, like "Stealth Rock" or "bulky waters"-- chances are, those things are going to matter a lot regardless of how the metagame evolves. And if some of the differences between the two branches are entirely contingent on these relatively static factors, I don't think it would be tough at all to call one of them "better" than the other. I mean, if there's still "essentially no reason not to use Delibird on team X," and branch B still uses team X with Deoxys or whatever, that's a (small) example of optimization in my book.

Obviously from a practical standpoint it's kind of a joke to try and pin any specific examples down though. But I think it's theoretically possible to look at little examples like that and say that one metagame advanced "better" than another, or even maybe "further."
 
You're not optimizing in the sense that "independent of other factors, people who use Delibird are more likely to succeed than people who do not use Delibird," but you're optimizing teams which operate more successfully with Delibird than without, and you're optimizing Delibird itself.
Click to expand...
Except to use Delibird like that, the role it fills must be significant enough - which is 100% cyclical, making it a cyclical change.

But I can look at a metagame that uses Delibird, and even if it's only used in a very particular team setup, if it's always better to use Delibird in that setup, that's optimization. It might not necessarily be optimal to use that team, but if "if I'm using that team, I'm using Delibird" is true, then that's "a kind of" optimization.
Click to expand...
The reason I have my gripe is that, it's impossible to say it's ALWAYS better to use X in anything due to the sheer number of variety in Pokemon. There are many ways to accomplish the same thing and thus the only reason why you'd pick one over the other is because of the team. If you're going to say "Delibird is optimal with these 5 other Pokemon" then there's no point of you using the word optimal since you're comparing it with literally nothing and does not tell you much about the entire metagame.

Like I said, until you can find me a more concrete way of how these "changes" mean and how exactly you're planning on measuring optimality... it's sort of futile and it's simply better to think of them as different.

Of course, the only case is an underdeveloped metagame, where you can argue that one metagame is "further" than the other. Other than that, we just have a different set of Pokemon that are mainly used. You can test out how much a team in one metagame will do in another, but in the end, that doesn't tell you that much because pokemon is really too varied. I'm not going to say "Bulky water" is an optimal change, it's just one way of using Pokemon, and different dominant play styles will lead to a different metagame. The point is that depending on the starting point, you're going to end up with different equilibria, with no real way to "compare" between the two.
 
Tangerine said:
Except to use Delibird like that, the role it fills must be significant enough - which is 100% cyclical, making it a cyclical change.
Click to expand...
If that role were always there, and were always significant, then that's an optimal change though, right?

Tangerine said:
If you're going to say "Delibird is optimal with these 5 other Pokemon" then there's no point of you using the word optimal since you're comparing it with literally nothing and does not tell you much about the entire metagame.
Click to expand...
Well you'd be comparing it to a metagame in which Delibird is not used with those 5 other Pokemon. I totally agree that it "doesn't say much about the metagame," but I think that it does say something and that there are enough examples of these sorts of optimizational changes that we could say that we're "smarter" at the game than we were a few months ago. I know that that's an oversimple characterization of the game, but it's not like it couldn't possibly extend to more complex examples--just, we wouldn't really be able to analyze it, because as you said, the game has so much variety. Actually, I think we would be able to analyze it, just not in a really precise or "concrete" way, which may or may not be useful.
 
If that role were always there, and were always significant, then that's an optimal change though, right?
Click to expand...

It's always there, but it's not always significant. The metagame dictates what is significant or not. If you're implying that the discovery that Delibirid could be used for that role is optimal - then I think you're just throwing the word around to the point where it might as well be cyclical.

Well you'd be comparing it to a metagame in which Delibird is not used with those 5 other Pokemon. I totally agree that it "doesn't say much about the metagame," but I think that it does say something and that there are enough examples of these sorts of optimizational changes that we could say that we're "smarter" at the game than we were a few months ago. I know that that's an oversimple characterization of the game, but it's not like it couldn't possibly extend to more complex examples--just, we wouldn't really be able to analyze it, because as you said, the game has so much variety. Actually, I think we would be able to analyze it, just not in a really precise or "concrete" way, which may or may not be useful.
Click to expand...

In which case, what's the point of studying these changes then? Will you look at a metagame where darkrai isn't used very much for a lot of other factors and say "hey our metagame is better because they don't use delibird"? Sure, if darkrai was used a lot and people were looking for something that can spin and deal with lead scarfrai, then that's not a "completely different metagame", just a metagame that hasn't developed like ubers have. I don't know what's the point of just calling those changes optimal really - when we say optimal, we should make a statement about the metagame - we say Stealth Rock is optimal, and for a good reason.
 
My point is that if one set of players understands that Shed Shell is optimal on Skarmory whenever people start using a lot of Magnezone, while another group just flounders around and uses Whirlwind more often or something, that first group is doing something "right," not "different." Sure, Magnezone might hardly ever reach the amount of usage in metagame B that it does in metagame A, but that doesn't necessarily mean it's not still smart to use Shed Shell, it just means the metagame B players might not have been observant enough to notice that it was, at least sometimes, optimal.

I don't know how clear my position is here because I know I'm sort of using an obvious example of cyclical change to illustrate an example of optimization. What I'm basically saying is that metagame A's switch between Leftovers and Shed Shell is obviously cyclical. However, the fact that metagame B never reaches a point at which Shed Shell is used over Leftovers either shows that there's no need for them to ever switch to Shed Shell ("the metagames are too different"), or that the people playing in metagame B just don't realize that Shed Shell is better than Leftovers in some cases. The fact that they don't realize this would mean that they're doing something "wrong," which means that metagame A is doing something "right," which is why I'm calling it "optimization." The question is whether this really is an issue of observance, or if we can just chalk it up to cyclical change (maybe metagame B has "evolved" some other way of reliably dealing with Magnezone). This obviously is going to be super complicated 99% of the time (for reasons you've stated) so the only reasonable way of verifying whether a difference is optimizational or cyclical would be having the two groups actually communicate or play with eachother.

This might be totally off-track from what you're trying to get through to me right now but whatever. It seems like I might be talking about a "different kind of optimization" but basically I'm just gravitating towards whatever interpretation I can find that seems meaningful, so bear with me >_>

I think this sort of thing would be interesting to test. Have two teams of like 5-20 players and have them independently play with the same arbitrary banlist, and try to "evolve" their own respective metagames until having to face eachother in a team tournament or something. I don't know how much it would tell us, but either way it at least sounds fun
 
My point is that if one set of players understands that Shed Shell is optimal on Skarmory whenever people start using a lot of Magnezone, while another group just flounders around and uses Whirlwind more often or something, that first group is doing something "right," not "different." Sure, Magnezone might hardly ever reach the amount of usage in metagame B that it does in metagame A, but that doesn't necessarily mean it's not still smart to use Shed Shell, it just means the metagame B players might not have been observant enough to notice that it was, at least sometimes, optimal.
I don't know how clear my position is here because I know I'm sort of using an obvious example of cyclical change to illustrate an example of optimization. What I'm basically saying is that metagame A's switch between Leftovers and Shed Shell is obviously cyclical. However, the fact that metagame B never reaches a point at which Shed Shell is used over Leftovers either shows that there's no need for them to ever switch to Shed Shell ("the metagames are too different"), or that the people playing in metagame B just don't realize that Shed Shell is better than Leftovers in some cases. The fact that they don't realize this would mean that they're doing something "wrong," which means that metagame A is doing something "right," which is why I'm calling it "optimization." The question is whether this really is an issue of observance, or if we can just chalk it up to cyclical change (maybe metagame B has "evolved" some other way of reliably dealing with Magnezone). This obviously is going to be super complicated 99% of the time (for reasons you've stated) so the only reasonable way of verifying whether a difference is optimizational or cyclical would be having the two groups actually communicate or play with eachother.
Click to expand...

Ignoring the flaw within the example (you lose a lot by switching from leftovers -> shed shell, many people are fine just using whirlwind and taking that risk rather than lose lefties recovery so it's definitely not necessarily optimal) I think the point you are trying to make is that one metagame is more competitive than the other, and thus, the one that is more competitive is optimal. The answer here is "obviously", but you can't assume that "being competitive" will lead to one, specific, metagame - it'll lead to an "efficient" metagame. But what happens if there's another metagame, that started from a different starting point, and ended up with a completely different metagame at the end? It is also efficient, and thus "optimal" according to you.
 
@mtr12 - SR does have some risks setting up. You risk your Azelf or Areodactyl to die to flinching from Jirachi or a Pursuit from a Tyranitar? Is letting your lead die worth setting up SR when your opponent has six 'mons left and you only have five?

As far as Stealth Rock usage goes though I admit people have been using it more intelligently since the advent of Platinum... Look at Infernape, Jirachi, Swampert, and Metagross, they are pretty safe leads to set up SR and pose the risk of taking down something else/doing heavy damage/Tricking/scouting via phazing.

This game is all about risk and reward, but playing and planning the smartest tends to yield decent (I didn't say best due to hax :x) results. I wouldn't bother playing with Skarm without Shed Shell, its not required, but being able to switch out of Maggy is very valuable. Sure you're going to miss having lefties recovery every time you're forced to switch it in on a Scizor's U-Turn, but its worth it in my opinion. Some teams will require Skarm to have lefties though due to it being one of their only walls.

No offense, Blame Game, but some of your posts just seem confusing...

But yeah a couple players teams to develop their own metagame would be fun and interesting. If you were to adopt a format like the world cup tournament format and made a tournament to do it I think that'd be fun. I'd love to see the results of that.

Admittedly by now I would've thought Mence would've taken the no. 1 spot from Scizor (due to Dragon move/EQ/Fire move hitting everything hard, Intimidate, great resistances, etc.) and Cresselia to not be at the bottom of the barrel in OU, but the metagame has progressed somewhat over time and I guess that's all you can really ask for...

Tangerine, your avatar makes me want to stare at it again.
 
@mtr12 - SR does have some risks setting up. You risk your Azelf or Areodactyl to die to flinching from Jirachi or a Pursuit from a Tyranitar? Is letting your lead die worth setting up SR when your opponent has six 'mons left and you only have five?
Click to expand...

I realize this. Therefore, I would be inclined to argue that the wax and wane of suicide leads would be cyclical, but the general use of Stealth Rock (the best move in the game hands down) would not be.
 
You must log in or register to reply here.
Back
Top