Policy Review Discussion - CAP Projects for metagames other than OU

[Deck Knight]

This is being moved over from the PR thread, since it's a sufficiently complex topic to merit its own discussion.

I'll re-start the discussion with the relevant quotes.

Well, I guess this is the best place to discuss about it, so I'll just lay out some details.

There's been some talk in the other thread about trying to make a CAP Uber. As other vets should know, we always have been wary about the idea because we feared it could turn the CAP process in a shitstorm, because of how Ubers are inherently broken (but at the same time how they can't be too broken to a certain extent).

So, as I said in that thread, I propose that, should we ever find ourselves making an Uber CAP, when it will come to stats we'll offer two different templates

1) Template 1: 680 BST (no more, no less); no stat below 90 or above 150. This has been the standard for most "cartridge" Ubers past 2nd generations, barring a handful of exceptions (Groudon and Kyogre if I remember correctly, who sit at 670 BST but still respect the 90/150 boundaries)
2) Template 2: 600 BST. No stat restrictions. With "only" 600 points to work on, it is much harder to break a Pokémon on stats alone, no matter how you tweak them (Deoxys is an excellent example).

This way, not only we are more consistent with the in-game precedents, but we also prevents the process from running amok by putting objective and clearly defined limits to the extent of the Uber's power (although it should be noted that particular care should be observed in which ability is given to (1) Ubers).

Any comments on this? If we are to try our hand with a CAP Uber, we should clear out such problems as well.

In general a lot of that is borne out by the process, since we usually select ranks and expect people to keep within them. Generally they're done on a maximum basis, but since Ubers are more flexible we could also set a minimum basis of say, Rank 4 (Below Average) or something. I haven't busted out my BSR calculator in forever, but effectively it means you can't just leave a stat hanging at some abysmal level. The Tank Spread would have to cut 85 points in defenses and push them into SpA o make it 90 to meet such a standard.

That being said, the limit on a 680 model should be 160. Slaking isn't an Uber, but it has Uber stats (670 like the Hoenn Ubers), and its Atk is 160. Its SpD is 65. I'd say having those limits on a 680 mon should be enough, and conveniently enough if it were to make both SpA and SPe and dump stat, it would have Rank 4 strength. Such a system would make the optimized tank stats 160/160/115/65/115/65. If we raised the floor to Rank 5 or more in every category, the optimized tank would be 160/160/109/77/109/65.

Just some food for thought. Picking 6 stats between 90 and 154 is just a tad bit too limiting, and while there is some allure to optimization (me being the case in point), I think using the 65-160 limits would be far more flexible without breaking the spirit of that route.

Exactly what I was trying to get at. Maximums are easy to set, and I knew those usually exist but rarely would there be a need for minimums. Just adding them in makes a huge difference in what kind of spreads are allowable. However, I almost feel that their would need to be some other requirement other than a min if the second template Zarator provided was used, as without one, or with one really low, it allows for optimizing to the extent of brokenness, but with a higher one, there is no reason to ever choose the 600 BST over the 680 BST.

As a more general question about an Uber CAP, if one was to be done, how exactly would it be handled from a design standpoint? While Ubers may have a metagame, it is really made up of cast-outs. No Pokemon there was designed to compete in such an environment, and creating one to do so really would cause us to make something in a way that was never really done before. So the question is, would we be designing a Pokemon to compete in a specific role in Ubers, or designing an Uber type Pokemon in general, and just seeing if and how it fits in? Another interesting thing that could possibly be done if it is the latter, is actually testing it in OU and seeing whether or not what was made is actually Uber material.

Basically, what I would want to know, not just for Ubers, but for any alternate tier CAP (with the possible exception of LC, which is like its own mini version of OU), is how would the process go about designing such a Pokemon, when the way tiers are made does not just allow Pokemon to be placed directly into them?

Technically speaking, no Pokemon in OU is designed for OU either. OU in 5th Gen is arguably a more contrived metagame than Ubers is. The point of the CAP process is to have a concept that, when built around, creates a Pokemon with a specific goal in mind for whatever metagame it's being put into. The only real difference between OU and Ubers is that the caliber of Pokemon in Ubers exceed even the power creep that has become endemic to OU. The only thing a project on an alternate tier would have to do is answer questions relevant to that concept. Ubers might even be a better environment in some respects, as its now a much purer metagame. It doesn't have complex bans like Swift Swim + Drizzle, and the tier itself is more stable.

Or to summarize: We introduce CAPs into a metagame to answer questions about that metagame, and we have done OU in the past because it was the most balanced metagame with the largest following, and therefore the best suited to answering questions about competitive Pokemon.

In this generation, OU is significantly more complex and fairly volatile since complex bans have been put in place. It is no longer the de facto stable, go-to metagame. If Ubers is a better environment to answer questions for because of its lower volatility, then creating a CAP for it would not be totally out of line.

With that being a solid basis for discussion, I want to focus specifically on finding the strengths and weaknesses of various testing environments. Our next CAP will in all likelihood be for OU to try out our new implementation system, but getting the ball rolling discussing the potential merits of building for other tiers seems warranted.

Remember that we want each CAP to answer questions, and the metagame we select should be conducive to answering those questions. It must also have enough players to attract discussions and excite the forum. In theory, we could even build a CAP for an older period of 5th Gen OU, now that we have them documented.

 

[a fairy]

Sorry if this is out of place, but " In theory, we could even build a CAP for an older period of 5th Gen OU, now that we have them documented. "

Why not previous generations? ADV, DPP(?) and others?

 

[Deck Knight]

Sorry if this is out of place, but " In theory, we could even build a CAP for an older period of 5th Gen OU, now that we have them documented. "

Why not previous generations? ADV, DPP(?) and others?

CAP is about getting information about competitive Pokemon. Sadly, older generations aren't supported on a widespread basis, and even if they were, you would not get the levels of participation that would justify spending 3 months building something for it.

This means essentially we'd have to build around 5th Gen metagames, be they Uber, OU, or VGC - the basic requirements being that the metagame we build for is a) best suited for answering specific questions in the concept and b) has sufficient following and participation to get good data.

Technically speaking, VGC is the worldwide official metagame. But it isn't competitive Pokemon in its entirety under Smogon's understanding of it.

 

[Aerodactyl Legend]

Before the thread turns into Ubers completely (which it looks like it will most likely be)--I've been pondering about it and what I think is that if any other tiers are done in CAP, then we should only explore Ubers. LC can work I suppose, and VGC can be done by doing OU and focusing on a concept based in working in Doubles (a concept proposed on CAP1 actually). UU, RU, NU, etc. however, brings too many restrictions. It's much easier to make a powerful Pokemon than making one specifically "weak" enough to fit in a lower tier. And it would be pointless in making a weak Pokemon. Not to mention that it would be hard to devise thoughtful questions which are to be answered by the time playtesting ends.

Restricting stats probably won't help in making one that would fit comfortably in UU or below. For example, Breloom is OU despite the fact that its BST is only 460, which is the lowest in the tier. And then look at Slaking, as Deck Knight mentioned, which has a BST of 670 and is in RU. With that in mind, we would move forward and then just give the Pokemon "bad" abilities, moves, and typing which then limits the creative process and makes it less about competitive play.

If anyone has any ideas on how we could overcome the difficulties, then bring it up, as I think it's also an important issue when discussing the exploration of other tiers. Otherwise, Ubers looks like the way to go.

As for Ubers itself, I agree with Deck Knight in not being hung up on the tier we build the CAP for during the early steps. We just have the tier in the back of our minds and roll with it. Concepts shouldn't use the words "OU" or "Uber" in them, just try to establish a subject to learn about. Characteristics like stats can be ironed out during the appropriate steps, and then, it's up to the community on keeping the project in check.

Since stats are definitely not the only deciding factor in a Pokemon's tier placement, I'm wondering if instead of going on a maximum/minimum outline for individual stats, focusing on maximum/minimum outline for BST. This gives us more freedom during the the proposal of concepts. Of course, a maximum for an individual stat should come up (154), but a minimum is probably less necessary.

Any thoughts?

 

[Metal Bagon]

You guys do pre-evo's as a side-project at the conclusion of CAP, don't you? Then why do LC when you could use the pokemon alreayd made.

 

[Aerodactyl Legend]

You guys do pre-evo's as a side-project at the conclusion of CAP, don't you? Then why do LC when you could use the pokemon alreayd made.

Currently, pre-evos are done closer to the end of the actual CAP. It's different than actually creating a Little Cup Pokemon to begin with because: 1) The process of creating them is nowhere near the depth and detail employed during the actual CAP project (generally less focus on competitiveness and more on being the pre-evo;, and 2) Since it's a pre-evo of an existing creation, we already have limitations set of what can and cannot be done right away.

Producing a Pokemon for LC is limiting in its own right directly out of the gate, which is why I'm not too keen about it.

 

[zarator]

About the replies to my stat limits proposal:

600 BST with no limits still allow for brokenness.

Very true, but I think that, in order to be broken even by Uber standards, a 600 BST spread should be stretched to the limit it only takes the common sense of the TL to ban it. The fact we say "There are no limits" doesn't mean we accept everything. There have never been limits on the stats of OU CAPs so far, yet many TL refused arguably "ridiculous" stat spread submissions.

Limits for 680 BST should be different.

The 154 upper limit seems hardly important to me. Only the Ubers from generations 1 and 2 had 154 stats, and we all know Pokémon's most important standards about legends (like the trio cartridge groups) have been established from gen 3 onwards. Plus, it would only cause us to produce only 154 stat legends, which would look a bit "exploiting", and far from what the modern Ubers look like.

I'm instead totally contrary to the 65/160 limits. Slaking and Regigigas are very specific Pokémon, which don't fit at all within the Uber category. Instead, they're on a category of their own. Probably, if GF were to release another Pokémon with a very crippling ability (i.e. not defeatist), it would use again a 670 BST spread with those boundaries. But Ubers follow other standards.
Moreover, the lower limit in particular is way too low, and allows for many ridiculous spreads. For example, since most Ubers have 90+ speed, having 65 or 89 speed rarely matters. Hence, you can drop Speed and one attacking stat to 65 while turning the Pokémon in a fearsome tank (ex: 160/65/110/160/110/65).

In short, look at Dialga, Palkia, Giratina, Reshiram, Zekrom, Rayquaza. Those are the ubers whose model we should follow closer. You say they don't allow for creative spreads? I completely disagree. For example, I still have to see a 150 Speed 680 BST uber. Plus, Pokémon are not only about stats. A Bug/Ground uber, an uber with Intimidate, an uber with Rapid Spin... there are a LOT of things which could make our Uber very unique, which are absolutely not limited to the extent of its stat spread.

 

[jas61292]

I would like to clarify a bit what I was trying to say in the post quoted in the OP. Well, actually, I think Aerodactyl Legend did a pretty good job of saying what I meant. While I had reference Ubers, that tier would probably be of least concern, as it can be seen as an untouched metagame; Pokemon in its purest form. While becoming Uber is not natural, being allowed in is.

The bigger concern would be making CAPs for lower tiers. As much as I love the sound of making a CAP for my favorite tier (RU), it just does not seem right. Pokemon are not given a tier based on how good they are, they are given a tier based on how they are used. Even if we create a Pokemon, there is no guarantee that if it were just added on as a new Pokemon that it would even be allowed in the tier we chose. Doing a lower tier CAP really wouldn't be any different than taking an OU Pokemon and allowing it in UU, or some similar change. As appealing as the concept is, I really don't think a lower tier CAP would do much for us.

That being said, I fully support the idea of alternate tier CAPs. I just believe that outside of OU, Ubers and Little Cup are the only other tiers where creating a CAP would teach us things that we have no other way of learning.

 

[Rising_Dusk]

This was actually discussed in extensive detail a very long time ago among Doug, myself, and a lot of CAP regulars on IRC. It received unanimous support, actually. There were a lot more interesting mechanics proposed for that one (double CAP), but I think we should start it slow if we start it at all. There's really not much to discuss either, to be perfectly fair. It would operate identically to OU's CAP creation process, it would just require a TL and ATL attuned to that different metagame. Uber is the most obviously cool one, IMO, since we already kind of make LC with CAP Pre-Evo and UU/RU is up in the air right now.

That's really all I can say; we don't need to change anything except the way we think for an Uber CAP. All of these limits that others are proposing shouldn't exist, but should be things that the TL keeps in the back of his/her head during the process. They shouldn't be hard-coded into the process by any rights.

 

[smallvizier]

It would just require a TL and ATL attuned to that different metagame.

If we do decide to run a CaP based on Ubers, Doubles or any other 'non-standard' metagame, it might be fun to run a couple of events beforehand. Nothing major; just a post in the CaP forum (and maybe one in the relevant metagame's forum) telling people that our next CaP will be in Ubers, starting in a week, and encouraging lots of people to get online on (say) Wednesday evening and Saturday morning, to get re-aquainted with the metagame and to chat with some of the other CaP people.

There could even be a mini-tournament, if somebody wanted to run one. Or a Ubers+CaP tournament.

Whatever the format, a low-key, promotional event could drum up interest, increase people's knowledge of the metagame, and get people thinking about potential concepts in advance of the poll.

 

[zarator]

All of these limits that others are proposing shouldn't exist, but should be things that the TL keeps in the back of his/her head during the process. They shouldn't be hard-coded into the process by any rights.

That's pretty much what I meant to say. Nowhere should be written something like that, but I think that a wise TL would better respect them. That's my opinion.

 

[HD]

There could even be a mini-tournament, if somebody wanted to run one. Or a Ubers+CaP tournament.

Whatever the format, a low-key, promotional event could drum up interest, increase people's knowledge of the metagame, and get people thinking about potential concepts in advance of the poll.

This (CaP mini-tournies) has been attempted before by CaP during generation 4. It did not get much attention and was given up.

Otherwise, I'm in support of making CaPs for other tiers. The point of CaP at the moment is to learn about the current OU metagame. No reason not to extend that definition to other metagames.

 

[shinyskarmory]

IMO building pokemon for metagames other than OU should have been explored a long time ago. By building multiple CAPs at once (say, 1 OU and 1 fill in other metagame here), we keep the CAP metagame OU centered but also keep more people involved. With more involvement, we'll gain more interest, which might allow us to attract enough people for a third CAP mon. It's a snowball effect with literally no drawbacks. There's no reason not to increase the number of simultaneous CAPs unless the quality begins to decline.

 

[tennisace]

The only drawback to that is if there isn't enough interest in a CAP project to keep it going consistently (i.e. if an OU one stalls), there 100% won't be enough participation for two (let alone three). Also, do you realize how much work the mods (and by extention TL/ATL) put in? Having two or more CAPs at once is completely ridiculous.

On topic, I know next to nothing about the Uber metagame at hand. However I completely agree with Zarator's last post. While yes, we could do something stupid like 1/200/1/200/1/197 with wonder guard, I have faith in the TL (and mod staff) to reject or outright delete obviously "broken" stat spreads like that. Because we force every CAP to have a concept, submitters must prove that their stat spread fits the concept. I cannot see blatantly unbalanced stat spreads fitting any concept aside from a concept titled "Blatantly unbalanced Pokemon" (which would never be accepted lol).

TL;DR I trust the TL and my fellow mods to quash stupidity when it arises. We dont really need many limits aside from like a BST 680 limit. Any other limit would be a "soft" limit (i.e. if you come up with balanced stats with one stat over 160 I would consider it).

 

[Paradox]

I think an important thing to lay out for this process is when exactly we would decide on the metagame we're working with. I definitely think that it is something that should be voted on (as in we vote on what metagame we're in, directly or indirectly, at some point) to make sure there is suitable support for it. I personally think that it's something that should be dealt with at the concept stage, as it was when somebody (Destiny Warrior maybe?) submitted a double's concept for CAP 1. If people like an alternate-metagame concept enough to vote for it then they should support it throughout the project.

So it would go as such: somebody could submit a concept like "Uber Rapid Spinner", where they specify that the concept is for the Uber metagame. If the concept wins we have an Uber CAP that was voted for by the majority of the community, and otherwise we don't worry about it. The obvious problem with this is that we won't always have, as Dusk said, a TL and ATL "attuned" to the relevant metagame, but we could always have the TL choose an advisor or something if they didn't feel comfortable enough in the relevant metagame. The only other way to really deal with that issue is to vote on the metagame before actually deciding on a TL and ATL, which could get a little messy, but could also work.

 

[capefeather]

RD said what I was going to say about stat limits...

As I see it, there are two issues to deal with for an Uber CAP (well, I suppose these also apply to other metagames, too) as opposed to an OU CAP:

Firstly, the flavour aspect is going to be radically different and less defined. Are we going to make an endgame Uber legendary (e.g. Mewtwo), or are we going to make a Pokémon that fits the concept but is not a legendary (e.g. Scizor, Forretress, Ferrothorn), or something in between (e.g. Heatran)? Are we going to entertain the idea of putting an Uber-centric CAP in CAP OU? If we make a legendary, are we going to give it a signature move or even a signature ability? Such questions are going to come to mind as we start doing stat submissions and art submissions for an Uber CAP.

Secondly, Ubers is not a "main" metagame; i.e. it runs solely on its own popularity and the interests of its players, and no one user is "required" to play it to be "in the know" on Smogon (in quotes because of people like Theorymon, lol). So we're probably going to have users who have never played Ubers piping in (and VOTING) without knowing what they're talking about (at least they actually play OU...), and a lot of the potential CAP population may butt out out of a lack of desire to play Ubers. This may cause an Uber CAP to miss the point entirely. I'm not saying that this is a reason not to do an Uber CAP, just that it's a major potential problem that we should keep in mind if we're going through with it.

 

[sax king]

CAPs player base is small enough, I'd rather not split it into five different tiers. Yes, it'd be nice to have five different tiers, but realistically, no one would play any of them. There are about 10 people online in the LC tier at a time. Imagine how small the CAP LC tier would be. We should just focus on OU.

 

[zarator]

Ubers is not a "main" metagame; i.e. it runs solely on its own popularity and the interests of its players, and no one user is "required" to play it to be "in the know" on Smogon (in quotes because of people like Theorymon, lol). So we're probably going to have users who have never played Ubers piping in (and VOTING) without knowing what they're talking about (at least they actually play OU...), and a lot of the potential CAP population may butt out out of a lack of desire to play Ubers. This may cause an Uber CAP to miss the point entirely. I'm not saying that this is a reason not to do an Uber CAP, just that it's a major potential problem that we should keep in mind if we're going through with it.

That's a good point, actually, one to which I don't know how to reply. Although tbh I don't expect many more "clueless" votes than in OU. Keep in mind CAP is experimental by its own definition. If someone is good enough to "metagame" about OU, I don't expect him to say anything excessively idiotic about Ubers. And don't think there wouldn't be "clueless" voters for an OU project, either.

But on the other hand, I'm worried about the potential of an Uber CAP to alienate some potential contributors, since admittendly, Uber is not as popular as OU. After such a long hiatus, maybe an OU CAP would be better than an Uber CAP to garner initial support. Then, once we're back in business, we could even try our hand with an Uber CAP sooner or later.

I'd like to point out that this proposal comes from someone who really enjoys Ubers, and who would really like an Uber CAP to see daylight in the first place.

 

[Focus]

Just posting to say that I would make an effort to be involved in an Ubers CAP if there ever is one. That metagame would be interesting to experiment with because the characteristic centralization of Ubers can raise a bunch of new metagame-related questions to try to answer, partly because the issue of a pokemon single-handedly centralizing the metagame around it is not as big of an issue. Ubers is a relatively easy tier to get into for those who have little experience with it. Unfortunately the stigma that Ubers is somehow "broken" may deter people at first from the project, but the issue of Ubers being foreign to people (leading to most voters making bad decisions) will be ironed out over time, as people get more acquainted with the tier. Perhaps the biggest question that an Ubers CAP would answer is "Can the CAP process function properly in a starkly anti-OU environment?"

 

[tennisace]

Firstly, the flavour aspect is going to be radically different and less defined. Are we going to make an endgame Uber legendary (e.g. Mewtwo), or are we going to make a Pokémon that fits the concept but is not a legendary (e.g. Scizor, Forretress, Ferrothorn), or something in between (e.g. Heatran)? Are we going to entertain the idea of putting an Uber-centric CAP in CAP OU? If we make a legendary, are we going to give it a signature move or even a signature ability? Such questions are going to come to mind as we start doing stat submissions and art submissions for an Uber CAP.

I can honestly say we will not be making a "legendary" Pokemon as to limit flavor overtones from the process (in addition to the technical limitations we currently have, I think we're probably going to ban custom moves/abilities alltogether from now on. There are so many new moves/abilities this generation that if we have to make up some, we aren't thinking outside of the box enough.) Think Garchomp for this process, not Lugia.

 

[HD]

I think for CaP Ubers we should do (at the end of the cap obviously) a suspect test thing. We make the CaP and everything, and then test the mon in OU. At the end of the testing period for that CaP we hold a vote to decide whether or not we made a broken mon. It would probably be 2/3s of the vote to decide. If it isn't Uber (as if, this is CaP) the tl would make another thread to discuss how we royally screwed up.

The numbers (and proposal) are up for debate (obviously). Thoughts?

 

[tennisace]

First of all, OU is getting rid of suspect tests. That should tell you something. Anyway, this isn't going to work. The ubers metagame is a completely separate metagame from OU, just like 5th gen OU is separate from 4th gen OU. There are different niches to be filled. So if we build a Pokemon for the Ubers metagame, and it isn't broken in OU, that doesn't mean its even useful in OU. I could make a pokemon that counters Kyogre and have it be totally useless in OU due to other factors (think 4th gen Shedinja: good in ubers where Rain/Sun are dominant, shitty in OU where Tyranitar/Hippowdon were everywhere). We should focus on building Pokemon for whatever metagame the concept calls for and not worry at all about their impact on OU.

As an aside what would finding out if they were broken in OU accomplish anyway? It doesn't seem to me like it would provide any benefit to CAP OU aside from being able to use them in the CAP metagame, which in and of itself is only a byproduct of the different CAPs and not something we actually "care" about during the process.

 

[Elevator Music]

Echoing sentiments that others have brought up in that I don't really think it's a great idea to do a CAP for other metagames that aren't OU. Honestly the biggest reason I see for this is that it really limits the number of people who can participate in the discussion. OU is the most popular tier by far, and if we to say, make an LC CAP, there would only be a select few number of people that would actually know what they're talking about (small number of people that play the tier and thus a smaller number of people that play the tier and participate in CAP). I don't want to have argue with a bunch of people that have no idea what they are talking about when I know I'm right. Likewise, if we do UU, I wouldn't be able to participate without seeming like a complete moron (I don't play UU)... and I don't think anyone who does play UU would want to put up with me. "Just don't participate" doesn't really feel like something the CAP project should be saying, so.......

 

[Monstermongler]

I know I'm new here, so I'll get used to being wrong, but I think the great strength of UU, NU etc is the ammount of pokémon already available that can be used there. While there is only a handfull of pokémon really viable for use in the uber category there are tons of good NU pokémon.

For that reason I think it's more usefull to develop pokémon for the higher tiers. It's a lot easier to create a pokémon that really adds something to the game in OU or Uber than in UU or NU, and I say this being an UU fan myself.

So yes, making an uber might be a good idea (I don't like ubers, so I'm not going to try and make that decision), but making pokémon fo any of the lower tiers is probably less fruitfull than making them for OU (although of course I wouldn't mind pre-evo's, but that's barely a side-effect).

 

[Rising_Dusk]

I'm the kind of person to want to try something before saying that it is doomed to fail. Cape raises some good points when he talks about how people might just not get it when it comes to making an Uber CAP, but gauging how people think before you ask them to think about it is a dangerous play. I think we should try a CAP Uber project to see how it goes, and then in the aftermath decide whether it was a terrible idea or not, and whether we should do another afterward or not.

I think people who don't know the Uber metagame will try to participate in the CAP. I recognize that means that a lot of users won't know what they're talking about, but I don't think that's a bad thing. Remember why we have CAP in the first place, to learn. I have always been terrible at Ubers, though it's something I've always wanted to explore. Making an Uber CAP would be a tremendous help to a lot of users, myself included, who want to work with some of the more experienced players to create and then understand an Uber Pokemon and their workings. As a learning experience, I think exploring new metagames in CAP is unparalleled.

I think an important thing to lay out for this process is when exactly we would decide on the metagame we're working with.

This is an important point that I think isn't getting enough attention. My question is whether it should even be up to a vote at all? What if Uber ends up more popular somehow and we never make an OU CAP again? I don't like that. I think it should be decided by the moderators, or better yet, the PRC. I always imagined that an Uber CAP after every 3 OU CAPs would be a fair spread, but that's just how I saw it way back when.