Policy Review Policy Review - Focus of the CAP Analyses

[Fuzznip]

Written by Fuzznip, eric the espeon, and zarator. Approved by DougJustDoug.

If you are not an experienced member of the CAP community, it is strongly recommended that you do not post in this thread.

This thread is intended to contain intelligent discussion and commentary by experienced members of the CAP project regarding CAP policy, process, and rules. As such, the content of this thread will be moderated more strictly than other threads on the forum. The posting rules for Policy Review threads are contained here.

I'm sure that some of you remember almost a year ago when we discussed in this thread whether or not CAP analyses should mention other CAPs, be them teammates or counters. We thought that the issue was solved through the addition of a "CAP metagame" section within each analysis, keeping the focus of the CAP Strategydex on the "playtesting metagame" (i.e. the metagame formed by standard OU + the single CAP analyzed). It was a good decision for several reasons, and it helped to underline the fact that CAP is not just a "modded server" but, rather, a community focused on experimentation within the standard, existing metagame. Unfortunately, not all the problems were solved at that time. I'll try to explain how and why below.

We say that the CAP Strategydex sets are focused on the "playtest metagame" - even if not explicitly, and we keep updating the analyses in the several workshop threads, often with important changes (think, for example, of the "Wispy Kit" set, which flourished long after the playtesting was over). However, where are these updates tested in the CAP metagame? Hell, sometimes people (including me) judge the viability of a set through the interaction with other CAP Pokémon (that was the case of a recent discussion around a Fidgit's lead set). But even if we think that such discussions are not correct since we shouldn't take CAP Pokémon into account, making the necessary tests in the full-CAP metagame (and we have little choice here, once playtest is over) is as legitimate as testing an Uber Weavile set in the Standard ladder.

So far, I'm not suggesting any solutions, neither am I taking sides. I just wanted to point out a contradiction that, I feel, we cannot ignore if we want to live up to the Smogon standards of quality. We can't focus the analyses on a metagame if they get tested in another one.

I'll present three possible solutions which should help us find the consistency we have so far failed to achieve. They are just some ideas thrown into the fray, and should be treated as such (i.e. don't discount any "fourth possibility").

Option One
​

We "freeze" the analyses once playtesting is over. We keep updating the "CAP metagame" sections, but the sets stay as they are forever, since any new set could not be tested in the "legitimate metagame". The main flaw of this proposal is that we probably want sets like "Wispy Kit" to get more than a simple mention in the CAP metagame section as if they were "Optional Changes" material. It may be consistent, but certainly not user-friendly. Keep in mind that, once playtesting is over and many people want to learn how to play full-CAP, they will look to our CAP analyses for help in mastering the metagame, so we can't be sloppy about it for the sake of consistency alone.

Option Two
​

We shift the focus of analyses onto the CAP metagame, possibly including a "playtest" section, and write an article about how each Pokemon on it's own fares in and affects the standard metagame. This way, we could freely update the analyses and add new content to reflect what most of the CAP community is playing without any concern for consistency, and the lessons we learn from our creation's interaction with the metagame are spelled out in a direct form. An article is a far more appropriate medium to explain how a new Pokemon fits into the metagame than a standard analysis, which deals directly with "how do I use this to win" not "what does this do to the metagame, what did we learn from it". This article could be included in The Smog, /cap/articles/ or on a separate tab of the analysis.

Option Three
​

We divide the analyses into two tabs, one for the playtest metagame, and one for the CAP metagame. Any changes and/or updates after the end of the playtesting period will affect only the CAP metagame tab. I wonder if it could be possible to split analyses like that, and there is the risk of them looking a bit confusing and repetitive, but at the same time, this solution may be the "best of both worlds" we are looking for.

I hope that, from this discussion, several, more accurate and refined solutions may arise. My only firm purpose was to make clear that "keeping the things as they are" is hardly viable if we want to follow the standards of consistency and clarity Smogon enforces in the other competitive forums.

 

[Deck Knight]

I support Option Three.

CAP is essentially a research firm for the OU metagame. We want the Playtest results to remain as untainted as possible so that our research can be preserved for posterity. While we haven't even considered this, it would not be impossible to have a "re-test" of a CAP and see if we can get any more information from it that would go in the Playtesting tab.

CAP Metagame is a living, breathing thing demanding constant changes. As such it should be its own separate page and change as necessary with each new addition.

It might be more cluttered, but it keeps CAP in focus, gives us a backup for the actual research portion of CAP, and allows us to keep our CAP analysis fresh.

 

[Umbreon Dan]

Yeah, it's got to be option three imo. The "CAP metagame" sections aren't working out, and that problem is going to beome more prominent as we add more CAP Pokemon. The playtest metagame tab would give us one static analysis and one CAP analysis that gets regular updates.

I also agree that going back and retesting the CAP Pokemon in Standard once in a while wouldn't be such a bad idea. It takes more than two weeks to learn everything there is about a fresh Pokemon in the metagame.

I wonder if it could be possible to split analyses like that, and there is the risk of them looking a bit confusing and repetitive, but at the same time, this solution may be the "best of both worlds" we are looking for.

Do you mean from a programming standpoint? I've been looking at the CAP website's coding lately and adding another tab would be pretty simple. As far as I know, Doug (or chaos) is the only one who can modify cap_template.js, but from what I can see, adding a single line would create the new tab we need.

from cap_template.js

function composeDexPage(cap) {
	
	var lowerName = cap.pokemonName.toLowerCase();
	
	var txt = "";
	txt += '<ul class="tabs"> \n';
	txt += '<li><strong>CAP</strong></li> \n';
	txt += '<li class="tabspacer"><strong>Dex</strong></li> \n';
	txt += '<li><a href="/cap/pokemon/moves/' + lowerName + '">Moves</a></li> \n';
	txt += '<li><a href="/cap/pokemon/strategies/' + lowerName + '">Strategy</a></li> \n';
	txt += '</ul> \n';

We will have to consider renaming the original Strategy tab, but don't worry about the technicality of adding new tabs.

While we're at it, we should consider giving some of our CAP Pokemon ubers analyses ;)

 

[Rising_Dusk]

I support option three of the ones listed without a doubt. Option 1 and option 2 both don't make me happy, as I am actually quite the fan of the way we currently do it. I like having a CAP metagame section more or less, but am not averse to change if it doesn't complicate the display too much. Multiple tabs keeps the thing manageable, but still lets us do some adjustments here and there.

I have one big issue with what is being proposed, though, as we do not have "playtest analyses" for all of the CAPs prior to Krilowatt. What would we do, dig them out of archives? (If we can do that, then that sounds like a plan) If we cannot do that, we can't just make up some stuff as if we're familiar with the playtest metagame when we're not.

I also have another issue with this idea. Many sets are discovered post-playtest metagame, but would work excellently in the playtest metagame. However, people consider these sets "CAP Sets" insofar as we've never tested them in the playtest metagame. This is a contradiction and something I disapprove of greatly. I can say with absolute certainty that All-Out Attacker Stratagem would wreck the playtest metagame if it were to ever get used there, but since it was found and made and tuned in the CAP metagame, it gets no chance to prove itself. I have always preferred and not really minded a bit of theorymon on the matter of sets and "projecting them" back into the playtest metagame without having used them there. If I were to be able to pick anything, I would do the following:

• Keep the analyses exactly as they are right now
• Remove the "CAP Metagame" section for each analysis
• Create a new "CAP Metagame" tab for each CAP Pokemon that explains their interactions with the other CAPs, with pretty pictures and stuff. (Much like our current threat list, just with only the other CAPs and none of the stats stuff)

As I don't expect my idea to garner much support, I will be OK with this PR if the conclusion is that Option 3 be made the official policy.

 

[Deck Knight]

Considering we are implementing a revamp of Pokemon movepools, I don't think it's absurd to have another playtest period for each CAP.

Yeah, I know it's tough to balance those and the last time we tried it, it really peteredout by the end of it. But it does give us a decent level of housekeeping while we create CAP 11, which we could probably use as a breaking point after retesting Stratagem (CAP 5) or Arghonaut (CAP6) depending on pace.

 

[eric the espeon]

I'm not sure if an analysis is the best way to present the lessons learned in playtest, and the interactions with the rest of the metagame, which means I prefer option 2. If it's as simple as UD suggests, we could add the article as a tab. The actual analysis would be geared to the CaP metagame, current, and constantly updated, this is what a vast majority of players will be wanting once playtest is over. The article would be written once and left, and would be the conclusion. How effective the Pokemon was, what we learned, some example sets, how it applies to non-CaP (maybe a new interesting moveset for a normal Pokemon was found?).

 

[DougJustDoug]

As UD mentioned, adding tabs to CAP pages would not be too hard.

I really like the suggestion of making a Playtesting tab that contains a summary of the playtesting results of the pokemon in the standard OU metagame. It might look like a cross between an article and an analysis -- since the playtesting will invariably cover some sets that were most effective, and perhaps even list some sets that were expected to be good, but were not. But there will be other commentary from playtesting that looks more like a traditional article.

I know a normal analysis page is broken into sets and other sections -- but I think we should strive to intentionally differentiate the Playtesting tab from the Strategy tab. But, I also would like the tabs to all follow a similar theme (like the dex and strategy tab do today). So, we'd just need to be creative and come up with something different, but the same -- makes sense, right?

 

[Zystral]

The only major problems I have with any analysis keeping tabs on the CAP Metagame as a whole is the fact that they will constantly need updating every CAP, which could prove tiresome. I'm aware CAP has many hard-working individuals ready and willing to constantly do so, but it may be a bit long.

However, jumping off Fuzznip's idea of a fourth way, and expanding on what eric says, I think we should go about each section differently.
An analysis for the CAP Metagame is more beneficial, since new players are more likely to jump in during a Full CAP time, so by reading analyses they will be able to fit into the metagame much more easily.
This of course requires a lot of work, but as I said, we should have more than enough people willing to participate.

As for OU + CAP and for "What we learn", rather than an analysis, I think something along the lines of an article or a written report will be more effective. People who are honestly interested in finding out about such things would be prepared to read through the large article to absorb the information they need. An article focuses on the once-only metagame that CAP existed in, so it would not need to be adjusted and such.
That, and the fact that an OU analysis for that CAPmon is redudant, in my opinion, since you're never really going to use that set in that metagame ever again, and what may be effective in OU isn't always effective in CAP, as we learnt.
Of course, the barrier to this solution is that we'd need truly experienced people, those who really know what they're talking about to write said articles. I personally don't think that's a problem since we require the same standard from those who would write the analyses anyway.

 

[Rising_Dusk]

As a player, I'd like it if the "Playtest" tab be in analysis form. We could add a new section instead of the "Overview" that is more like a "Report" on how it succeeded and so forth. Everything else, though, I think should be the same as for a normal analysis as we'd currently write it. I'd really want counters and team options and to know those sorts of things in retrospect about our CAPs and how they fared in playtest. Just a wall of text report like some people are suggesting would be terrible and no one would read it.

 

[Zystral]

But the thing is, nobody else reads and seriously takes the Playtest Analysis anyway, since they're like "I want to know how to use this CAPmon" and then they don't get anything about CAP apart from 1 paragraph and then they ignore it. If someone was really interested they'd be willing to go through the wall of text.

 

[Fuzznip]

I personally like the idea of splitting the analyses for CAP Pokemon into two tabs. One for the CAP metagame, which would be constructed as an actual analysis with the Overview, Team Options, etc, and one for the Playtest metagame, which should be constructed a little differently. I don't want the Playtest tab to be only done as an analysis, I kind of want it like an article too. Like Doug said, it could be a cross between an analysis and an article (I kind of did this with Dan with the Krilowatt article). If we're going to go by this, though, we'd need to come up with a standard format.

I'm not necessarily worried about CAP analyses needing to get updated after every CAP, as we have many willing contributors up for the job. Furthermore, it gives users an excellent chance to contribute and get involved within the CAP community, something which I feel is very beneficial. We probably don't even need to update every CAP analysis with each new CAP. If one of our CAPs are not affected by the newly-made CAP, they don't need to get an analysis update. We only need to update the analyses that are hindered or improved by a new CAP. This is what I think, anyway.

Thoughts?

 

[Elevator Music]

Keeping this somewhat short, but it's always been my opinion that it's a huge waste of effort to make an analysis for a metagame that is never played. I like the idea of making the analyses geared towards the CAP metagame, and making an article or something (or not making anything at all!) for the "playtest metagame". I guess they could go in The Smog or on the CAP website, I don't know, but at least they aren't as useless as an analysis.

 

[Rising_Dusk]

I'm cool with the idea of having a playtest tab organized in some later-to-be-determined way for explaining how the CAP interacted with the OU metagame. Then, we can keep the strategy tab inclined toward the CAP metagame, as EM and everyone else has suggested. I really can't think of much else to say in this regard, as it seems that everyone agrees. In retrospect, I also don't mind the playtest thing being in report-form instead of analysis-form.

 

[Fuzznip]

It's nice that we came to a relatively fast conclusion. This is what we will be doing.

Conclusion
​

• Have a CAP and Playtesting tab for every CAP.
• Allow the mention of CAP Pokemon in CAP analyses.
• Have an article/report/analysis-type format for the Playtesting analyses.