Demystifying Stark Mountain

[Tangerine]

If you're like any other user reading and lurking Stark Mountain, you may find many things overwhelming and confusing. Things such as why Smogon does certain things in their policy, to confusing terms, to things regarding how the mechanics of the game affect the players, definitions, how to interpret certain outcomes, and such.

So, this thread aims to answer any questions you may have regarding anything Pokemon related you wish to ask, as long as they're not a question that belongs in the Simple Question, Simple Answer thread or the Research thread. This means that you should only ask questions regarding concepts that you find unclear and thus want a better explanation of. Do not make the questions controversial - I will not answer them. Your questions should also not be too general, but be very specific, or else I can go off in a wild tangent and not answer your question at all. Finally, note that my answers only reflect my opinion and understanding of the processes, and are not necessarily "official" answers, but I believe that they will serve as a good starting point for discussions (especially if you disagree with me!), and I am fairly confident I can answer a lot of questions (if I can't answer them, then I'm expanding my own theory too, so it's good for me!).

So ask questions, and I will do my best to answer. I'm hoping you guys get creative with this!

Do not ask questions like "when will this happen". I am not involved with Smogon Policy, and thus, I cannot answer those questions.

 

[Ramblin Wreck]

I know that reposting locked topics is against the rules, but what if you think a post got pulled off topic or that you could explain it better than the original poster?

 

[cantab]

Why are the mods so lock-happy? I've known no other forum where threads get locked so readily. I won't name specific mods or threads, but sometimes I feel that the mods essentially shut down a discussion based on their personal opinion.

 

[Tangerine]

I know that reposting locked topics is against the rules, but what if you think a post got pulled off topic or that you could explain it better than the original poster?

If you feel as if the thread is so far off topic that you can no longer contribute to the thread, contact a moderator. If the thread is salvageable, they will salvage it, or else, it'll be locked. If you feel as if you can create a new thread with a better OP, contact the same moderator (hopefully at the same time) and see if you are allowed to recreate the thread. It's advisable to have the opening post ready so that the moderator can see it.

However, in general, just contribute to the thread the best you can. If you feel as if you can write a better opening post than the opening poster, then contact a moderator and see if the opening post can be exchanged by your post. If you feel as if a thread is going off topic, then do your best to keep it on topic.

Why are the mods so lock-happy? I've known no other forum where threads get locked so readily. I won't name specific mods or threads, but sometimes I feel that the mods essentially shut down a discussion based on their personal opinion.

The few cases of threads being locked visibly (most threads are locked then deleted) are due to the quality of the opening post. The quality of the opening post is related to the discussion that will come forth - many users are too concentrated on "scoring points" by shutting down the logical inconsistencies or the ignorant points made in opening posts to actually discuss the subject matter brought to light intelligently. So, the poor opening post makes the thread generally difficult to salvage without any significant effort.

The majority of the locked and deleted threads (most of them which you can't see, and believe me, Stark had many of them) are mostly users asking questions that belong in the Simple Questions, Simple Answers thread, or simply other rule breaking violations.

 

[Earthworm]

What exactly do you think is missing from the arguments in the Salamence thread in the Policy Review subforum?

 

[Chickensandwichk]

Cresselia and Porygon-Z aren't OU anymore, and reachzero has decided that they will immediately join UU!

Why exactly are they joining UU right away instead of going through some sort of other process? If they are deemed too strong, when will it be possible to nominate Cress and P-Z as suspects?

 

[Tangerine]

What exactly do you think is missing from the arguments in the Salamence thread in the Policy Review subforum?

Theory.

There's a reason why I pointed out reachzero's arguments were on the line with what I was looking for, while I have criticized others. Many people simply only use their experience and theorymon to back up their arguments, without an overarching theory behind it.

In essence, in order to argue on "this will improve the metagame", you need to demonstrate that you know how the game of Pokemon works, not just on an intuitive level, but on a level where you can express your ideas logically. To an extent, many users forget that part and simply argue only on one dimension of the picture. You need to show "what aspect of the game of Pokemon is this hurting competitively", and you need to make sure your theory behind it is backed up by complete and consistent theory.

Why exactly are they joining UU right away instead of going through some sort of other process? If they are deemed too strong, when will it be possible to nominate Cress and P-Z as suspects?

The consequence of Smogon's Pokemon rankings based on usage is that, when a Pokemon is no longer OU, it is UU. If the Pokemon is deemed to strong after joining UU, then they will be banned in the UU tier and be moved up to the BL tier. X-Act also made it so that Pokemon that drop down from OU while there's a UU test going on will be temporarily BL until the test is over.

UU does not have a suspect ladder for these Pokemon to be "tested" on, most likely because UU does not have many players, and further dividing playtime into Suspect UU and Normal UU would likely hamper the development of the metagame. Hence, it immediately drops down to UU. This is one quirk of UU, that the metagame fluctuates based on OU developments. UU testing will be done when it is done.

 

[X-Act]

To expand on what Tangerine has just said about Porygon-Z and Cresselia going immediately to UU, it is obviously not ruled out that the guys that take care of the UU process might eventually mark one or both Pokemon as suspects.

 

[JabbaTheGriffin]

Just to clarify: If you do indeed feel that Cresselia is BL, you will be able to nominate it as a suspect 6 weeks from the date UU ladder ratings are reset.

 

[Eternal]

What is your opinion on usage statistics defining a Pokemon's tier? Do you think it's the best way to place a Pokemon in a tier or do you think there are other viable options in choosing a Pokemon's tier?

We know that usage statistics only portray the popularity of Pokemon being used. My opinion is that popularity should not be the only factor in categorizing a Pokemon tier over viability in being in that tier. The reason for this is because Smogon is defining tiers through the correlation of usage and viability as one. For example, Cresselia being moved to UU from OU. The reason it was moved to UU was because of its lack of usage in OU. While OU is completely different from UU. Cresselia might have not recieved much usage in OU due to the rise of Scizor, Tyranitar, Jirachi, and other common threats in accordance to usage.

-----

But, is what we're doing logical? We have moved Cresselia and Porygon-Z to UU due to usage and threw Honchkrow and Gallade to BL due to voting. We have also banned Gallade and Honchkrow from UU. How can we assess that UU would be more stable with Honchkrow, Gallade, Cresslia, and Porygon-Z all being in UU? We can't.

We simply neglected that reason and changed the tiers accordingly to usage and suspect voting at the same time.

 

[Yehom]

But aren't the tiers to seperate the Pokemon who are to powerful for the tier they were before? Garchomp became Uber simply because he was broken. But if for example everyone starts massively using (just an example) Luvdisc, would it then be promovated to UU and later OU maybe because everybody is using it?

 

[MapleSandwich]

I honestly believe the answer to this question would be yes, since Luvdisc has some set giving it the ability to have a niche in the environment if everyone is using it. This example has been brought up as "people playing the tiering system" many times, but there's simply no reason any group of people would ever do it. Tiers are based on how often a Pokemon in question is "used", since logically if a Pokemon is used much more than others, it has a combination of base stats, moves, and typing/ability allowing it to beat many others within the metagame. Determining a Pokemon's tiering directly through how often it is used is much more beneficial IMO than simply looking at base stats and type.

The tiers will never be absolute until after the generation in question is dead and gone, to use the expression, since metagames come and go and shift constantly. It makes things more interesting overall, and reflect what Pokemon are actually "Over used" and which ones are "Under used" as opposed to simply saying a Pokemon is "Used" a certain amount based on how much it was used in a metagame at one point in time, making it more accurate to use this system.

 

[cantab]

I think that usage is the best way. By using usage, we do I feel implicitly assume that usage and power correlate. This assumption is valid as long as people play to win, which they do.

With over 500 Pokemon (counting NFEs and Formes) to tier, tiering 'manually' by discussing and debating how good things are isn't practical. Just look at the length of the argument over Salamence, and imagine that x500. Tiering by discussion may work for games with fewer characters (like most fighting games), but there are just too many Pokemon for it to work.

With the presence of Shoddy, we have the ability to collect usage statistics on a truly immense scale. Last year there were something like 4.8 million battles on the Smogon server. Making the usage tiers is then very easy. (I think we could have a workable tiering system based solely on usage, without the banlists - but they do improve things, and I'm not saying get rid of them.)

The only other workable method I can think of for tiering is algorithmic. Make a formula that takes all the attributes of a Pokemon (stats, moveset, typing, etc), and spits out a rating number. The drawback here is that one could create an endless number of formulae; how do we decide which to use?

 

[Nida]

Arin has stated my question: how do the BL critters affect any newcomers to UU? Isn't it possible that any new drops to UU should be tested against current BL pogeys?

 

[RBG]

This is not meant to be a discussion thread. Tangerine is the only one who will be answering questions in this thread, so their is no need to reply to someone else.

 

[Tangerine]

What is your opinion on usage statistics defining a Pokemon's tier? Do you think it's the best way to place a Pokemon in a tier or do you think there are other viable options in choosing a Pokemon's tier?

But aren't the tiers to seperate the Pokemon who are to powerful for the tier they were before? Garchomp became Uber simply because he was broken. But if for example everyone starts massively using (just an example) Luvdisc, would it then be promovated to UU and later OU maybe because everybody is using it?

It would depend on the underlying assumptions. But in the end, it depends on the purposes of tiers.

Colin has noted in the past that the word "tier" is misleading, and it is. The OU list, is simply, the most overused Pokemon in the standard metagame. That is all Smogon's "tiers" are, a measurement of how often the Pokemon is used. The idea of UU is to be a metagame for Pokemon that are not used often.

So, in the end, Smogon's system isn't a set of tiers - it's just a set of metagames that give every Pokemon a chance to shine. We do not say anything else by dividing them up into supposed tiers, we do not make a statement about how useful they are, how powerful they are, or anything. This is simply based on a simple philosophy.

Now, is a usage based "tiering", ie, ranking of how useful a Pokemon, viable? That depends on the assumptions.

Under the assumption that the game is competitive, then usage based tiering is perfect. The Pokemon that people use are the Pokemon that will give them the highest probability of winning, and usage then becomes a direct metric. Under the competitive assumption, the usage of the Pokemon tells us a lot.

Now you may say "but I use this Pokemon, it's good, but it's ranked low", but the point is, such a system would not care about you, the individual, but the aggregate. What matters is which Pokemon does the group as a whole find it the most efficient, not you as a player. You can be ridiculously successful with a Pokemon that no one uses, but in the end, that is irrelevant if it doesn't catch on (if you do it consistently enough, it will catch on, which is a basic competitive assumption). Consider the "luvdisk" example. If the game is competitive, and you are winning with 5 Pokemon and a Luvdisk, then someone who uses the same 5 Pokemon without the Luvdisk, assuming same player skill, will perform better than you, and have an advantage over you in the game. Hence, under a competitive assumption, we can make this link between usage and how useful it is.

The real issue of course, it's that the competitive assumption does not actually look perfect, due to many factors, such as player biases towards certain Pokemon, or people having different purposes in playing the game, or ladder matches matching players based on ranking. While this may hurt the aggregate measure a bit, it still does not hurt the claim that usage measures what the aggregate whole thinks the Pokemon is good. There are also a lot of playstyles and people play differently, and the aggregate amalgamates all of them into one metric. So in practice, it won't be a perfect measure.

So, I think a usage based tier can be viable, and it provides us with a metric with some measurement errors. But, notice that this is not how we will interpret the metric at Smogon. We hold no such assumptions and we simply say that OU is the Pokemon that are most used, and we create other modes of play that allows Pokemon that aren't used very often a chance to shine.
Obviously, the philosophy behind UU does not care about Pokemon that are strong but are not OU, because it is specifically a tier that allows them to shine. UU will be a constantly changing metagame, not until OU completely stabilizes.

In the end though, I think there are a lot of different rulesets that are viable metagames. However, at this point, Smogon only considers the current system at hand. But I'll be the first one to tell you that every ruleset will create a different games, and there are many, many, possible metagames that may be "better" in some sense than the current Smogon metagames.

EDIT: I fixed up the post quite a bit, so I recommend that you guys reread it.

 

[Tangerine]

I edited the previous response, so I recommend that people re-read it - I wrote that response in a hurry so I apologize for the confusion it may have caused.

But, is what we're doing logical? We have moved Cresselia and Porygon-Z to UU due to usage and threw Honchkrow and Gallade to BL due to voting. We have also banned Gallade and Honchkrow from UU. How can we assess that UU would be more stable with Honchkrow, Gallade, Cresslia, and Porygon-Z all being in UU? We can't.

This was of course the biggest objection we had when the UU tier began testing - that it was a relatively unstable metagame because OU is unstable. But there was enough demand for it, and so the process started. It is up to Jabba and reachzero to decide what they will do with it next, and perhaps they can answer the question regarding how they plan on dealing with it.

Make a formula that takes all the attributes of a Pokemon (stats, moveset, typing, etc), and spits out a rating number. The drawback here is that one could create an endless number of formulae; how do we decide which to use?

My answer would be based on theory. When you want to create a metagame, you will want to decide the purpose behind creating the metagame, ie, the traits that will define it. Depending on what kind of metagame you wish to create, you will consider different formulas. All of this, however, requires analysis that is quite advanced for anyone to bother with (ie, most people who can do that level of analysis would rather do other things).

This issue occurs of course, with any sort of function that tries to convert N dimensions to a single dimension so that you can order it (in order to apply ordering on properties of vectors or sets of data, you need to define such an ordering). The definition of the ordering will depend on what kind of ordering you want, and what kind of game you wish to create. Smogon, does not have any official theory behind the metagame as of yet.

The absence of such a theory and the need for change in the metagame is why we had the suspect tests - rather than having people argue without theory, it was found to be more efficient to let the people draw the lines themselves (this is precisely the reasons the characteristics are vague - yes, it was done on purpose) and let the people decide what kind of metagame they wanted. However, this did not mean that we would test everything, but only things that people felt were prematurely banned, or things that people would feel that hurts the competitive metagame.

 

[WigglyWigglytuff]

Why are we not allowed to use certain moves with other moves? It dosent seem fair since the reason you cant is because the move is fro ma different game, but isnt that what Pal Park is for anyways?

 

[Tangerine]

Why are we not allowed to use certain moves with other moves? It dosent seem fair since the reason you cant is because the move is fro ma different game, but isnt that what Pal Park is for anyways?

Because the game mechanics do not allow you to get certain moves while having another move.

RBG Edit: Read this for more info.

When will NU be recognized as an official metagame? Will it be after OU Suspect testing finishes, and UU stablizes?

http://www.smogon.com/forums/showthread.php?t=63430

Anything related to the NU tier is available in this thread.

Please refrain from asking specific policy questions such as these - I am not involved in Smogon policy and thus, I will not be able to answer them.

 

[Xaqwais]

When will NU be recognized as an official metagame? Will it be after OU Suspect testing finishes, and UU stablizes?

 

[Felix91]

1. Are there any "official" non-VGC doubles rules?

2. Perhaps this is a bad question, but is this best way to improve your battling skills just practice in the tier, or are there any specific resources I should read/know to have similar benefits.

 

[Stylish Interval]

Theory.

There's a reason why I pointed out reachzero's arguments were on the line with what I was looking for, while I have criticized others. Many people simply only use their experience and theorymon to back up their arguments, without an overarching theory behind it.

In essence, in order to argue on "this will improve the metagame", you need to demonstrate that you know how the game of Pokemon works, not just on an intuitive level, but on a level where you can express your ideas logically. To an extent, many users forget that part and simply argue only on one dimension of the picture. You need to show "what aspect of the game of Pokemon is this hurting competitively", and you need to make sure your theory behind it is backed up by complete and consistent theory.

From what I understand from a lot of people's perspectives, experience and theorymon is all you need. Would it be possible for you to expand on how someone would come up with a theory, or what that theory would be like? I understand the showing process, but how does one figure out how a Pokemon contributes or takes away from a metagame? How can you make sure it's complete before sharing?

 

[Tangerine]

1. Are there any "official" non-VGC doubles rules?

No. Smogon will work on creating a competitive 2 vs 2 tier when Shoddy Battle 2 is released.

2. Perhaps this is a bad question, but is this best way to improve your battling skills just practice in the tier, or are there any specific resources I should read/know to have similar benefits.

Practice in general is the best way, but what matters the most is how you practice. I'm not aware of any guides that will help you think through how to approach competitive Pokemon, but my general advice is that, before you make your decision, write down everything you know about your opponent's team. Consider the situation you're in, what your opponent knows about you, what you know about your opponent, and based on the information, make your decision. The most important part for new players is to be able to manage the information they gain from the battle, and so writing them down while playing a lot, I believe, is the best way to practice.

From what I understand from a lot of people's perspectives, experience and theorymon is all you need. Would it be possible for you to expand on how someone would come up with a theory, or what that theory would be like? I understand the showing process, but how does one figure out how a Pokemon contributes or takes away from a metagame? How can you make sure it's complete before sharing?

It doesn't have to be complete. I'm not asking for a book.

Experience affects how well you can theorymon, and both of them give a perspective (which is really a very shallow theory).

In order to figure out how a Pokemon contributes or take away from a metagame, you need a way to analyze the game. What do you look for in the game? What is competitive Pokemon? What is the purpose of the game? You need to be able to answer these questions before you can actually answer if something takes away from it or not - without it you're just guessing. Your experience will be able to help you answer these questions, and with these questions, you'll be able to show how a Pokemon contributes or takes away from the metagame you have defined. Without this, any tiering argument is automatically incomplete - you're not arguing for anything without that, and you'll have to rely on other people some how mysteriously knowing what you're talking about in order to get any point across. (The latter isn't actually difficult to do, but it's still bad practice)

 

[kd24]

Why is a "buffer zone" theory in uu a bad idea? By adding it, you can keep the ever changing metagame a bit more stable, (if a Pokemon was not broken in UU and was moved to OU right near the line, it could continue being used in UU).

You obviously want Stark and PR to have the highest quality of posting. What are top three things you think many posters leave out or completely miss, that if added correctly could make every post much more coherent or at least up to par? (I have read your very helpful thread on "how to start discussions or argue", but a lot of posters (myself included), still feel we are missing things that could really help us out when discussing).

 

[Tangerine]

Why is a "buffer zone" theory in uu a bad idea? By adding it, you can keep the ever changing metagame a bit more stable, (if a Pokemon was not broken in UU and was moved to OU right near the line, it could continue being used in UU).

I don't think anyone has ever called it a bad idea. It is something that Smogon at this point does not do, but that may change in the future.

You obviously want Stark and PR to have the highest quality of posting. What are top three things you think many posters leave out or completely miss, that if added correctly could make every post much more coherent or at least up to par?

Many people ignore the other person's point and purpose behind the argument and simply concentrate on barraging the errors that may be present in the argument that do not affect the validity of the point the person was trying to make. Hence, reading comprehension may definitely help.

The counterpart to that is that many people do not know how to put together coherent arguments, and often their points are accompanied by irrelevant facts, logical inconsistencies, and just other aspects that show that the user has not thought about this for more than ten minutes.

The last part, ignorance, is something I'm trying to fix right now :p

I will answer the next batch of questions Tomorrow evening.