1. New to the forums? Check out our Mentorship Program!
    Our mentors will answer your questions and help you become a part of the community!
  2. Welcome to Smogon Forums! Please take a minute to read the rules.

"The Unban Everything" Mentality and the need to define a cutoff for OU.

Discussion in 'The Policy Review' started by Tangerine, Feb 21, 2009.

  1. Tangerine

    Tangerine Where the Lights Are
    is a Smogon IRC SOPis a Team Rater Alumnusis a Forum Moderator Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis a Smogon Media Contributor Alumnus

    Joined:
    May 4, 2007
    Messages:
    3,155
    Because I don't want to derail the Latios thread, and I think we should be looking at the tiering ideology.

    There is a mentality going on regarding "unban everything and start from there". In fact, this was the biggest argument that went on in the order of operations. Many people have been criticizing that the suspect test is not the "right way" and we should just unban everything at once.

    There has also been comparisons that "we should have done it like UU".

    So first, I will point out why the nUU testing doesn't mean we can justify the OU testing, and then go from there.

    The point is that I don't think your proposal would have gained much support if there was 5 BL, or even, 15 BL Pokemon. In this case, the "old method" would have worked which was to move BL Pokemon down and "test it". There would be no need to literally "start over" and recreate the tier from scratch. This precisely has to do with the purpose of UU - which is an environment for non OU Pokemon to thrive in. The lowers tiers are environments where Pokemon not seen in standard can thrive.

    I believe the argument was primarily accepted because of the size. The purpose of the lower tiers is to allow Pokemon that are never seen in OU to be used. If there are 50 Fully Evolved Pokemon that aren't seeing the light of day because of a tiering flaw, that is a lot more convincing than "we should start over for the sake of 10 Pokemon". This allows UU to actually serve its purpose.

    This justification does not exist for the OU metagame, since OU's purpose is to be the standard metagame that is "balanced". It isnt' the "first" metagame to be balanced, for obvious reasons since we can unban everything except things that are universally agreed to be broken (Kyogre) and start from there. The metagame in this case will balance eventually, and then we will have "Uber lite" as the first balanced metagame.

    I first find it odd that people who promote this view accept the suspects that we have derived from theorymon - especially since these are people who promote the "we must test and not theorymon" mentality. Supposign that they do unban everything (Except Kyogre), then we do end up with uber-lite as "OU", and that's that. The issue with this is the usage factor - do we really want a metagame that is really just a "balanced" uber tier? "Maybe". Does that mean the current OU will end up being UU, and so on? "Maybe". Is this considerably different from the OU that we constructed? Yes. Is this a problem? Maybe. Will it be smaller? Most likely. But under the purpose of tiers, this is how "it should be done" and anything else is "wrong". Therefore, I challenge this definition of OU being simply a balanced tier - and believe that it should be redefined as a metagame where we start with a set cut off point and go from there. (for example, we start by banning all legends greater than and/or equal BST 600 and go from there). It needs to be redefined, or else, what stops us from saying "this is OU" and going from there? There is a certain stigma attached to OU and I think we should stick to it, whether or not people believe that it is "aribtrary" or not because in the end no matter what, what we ban and do not ban is based on the metagame at hand.

    Does it really matter other than an exercise in philsophical "purity"? To be philosophically pure is to unban everything and starting from there and then constructing the "uberlite" OU metagame, rather than what people may prefer by practice. No matter how pure you might want to be, people will think of OU as the tier with Starmie and Jolteon and Snorlax. This has long been what has accepted to be standard, and believe it or not, this mentality is directly promoted by how the game is set up and the bans Nintendo has set up themselves. Is it wrong to start from Nintendo's 1 vs 1 tiers, other than to "minimize bans"? We minimize bans based on the standard metagame, not the overall bans, much like UU's banlist is simply BL, not every Pokemon in OU ~ Ubers (OU ~ Ubers is included but they're not even considered).

    Now, we still want to minimize bans. That's why we have suspect testing. Suspect are Pokemon from the banlist that we want to test. But first, note how we decided on the suspects to begin with. Suspects are Pokemon that we assume would be perfectly viable and not broken in the current OU metagame. Suspects are Ubers that we theorize that it may be viable in OU.

    There's an issue again, by unbanning all at once. It's not the idea of a "clusterfuck", but the possibility of it that we should be worried about.

    There are pros and cons to the argument of "unban all suspects" now. The Pros are, while we may be finished with the test faster most of the time, there is a chance that we end up with a tangle of mess that we can't quite derive the cause of, due to the number of suspects affect the metagame. This, believe it or not, is a significant problem since if we end up with the latter, we lose the purpose behind the suspect test and the bans end up being more arbitrary than it needs to be.

    Hence, the suspect tests. We test each suspect in the established baseline and see how each Pokemon can perform each each stage, which gives us a much better idea on what is causing what. This minimizes the chance that we as a community can't figure out exactly what is wrong at stage 2, or stage 3. The con, of course, is that the process is a lot slower than the above method of unbanning all suspects at once.

    I disagree with anyone claiming that the suspect test is flawed for that specific reason that we didnt start by unbanning everything, because in the end we will end up with a much solid reason, even though in the end everything is "subjective" anyway. It is because it is arbitrary we need to make a more systematic approach that concentrates on the performance of each Pokemon. The suspect test gives us much more control and clarity, a reason enough to "prefer" it.

    Hence, this is why

    1) the reasoning for the nUU testing does not apply to OU, unless we restructure everything again.
    2) we need to reconsider the initial cutoff of OU
    3) Why unbanning all suspects and the suspect test ends up in the same result anyway - except the suspect test is slower and it gives clearer results.

    Discuss. I don't want to hear "This is exactly what people who preferred the old UU argued" because, uh, it doesn't apply.

    EDIT: There is also a need to develop a philosophy for competitive Pokemon specifically - rather than just accepting the systems applied for other games. Pokemon is a much different game than most other games due to the variety and the versatility and I find it silly that we're just mindlessly accepting certain things as "true".
  2. Hipmonlee

    Hipmonlee Have a rice day
    is a Smogon IRC AOp Alumnusis a Super Moderator Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnusis a Past WCoP Winner

    Joined:
    Dec 19, 2004
    Messages:
    7,336
    One other thing I hear, is that the ruleset should be as simple as possible, or have as few rules as possible. Therefore we should ban as few pokemon as possible. This is not something I find convincing at all. The rule is the fact that there are restrictions on what pokemon can be used. Having made the decision to have such a rule (for whatever reason), the restrictions on what pokemon should be banned should be based on the reason for having a restriction in the first place, not based on a desire to have as simple a ruleset as possible when you have already made the decision to introduce the rule.

    I cant really see any important difference in terms of simplicity between "Dont use this pokemon" or "Dont use these pokemon" .

    Have a nice day.
  3. Kristoph

    Kristoph

    Joined:
    Jun 10, 2005
    Messages:
    1,725
    There's way too much to think about in this thread

    We do need to come up with a solid philosophy, I agree completely, but I'm still just not seeing how the "standard competitive philosophy" can't more or less apply to Pokemon, and I'm especially confused as to how "variety" in particular is something that you highlighted. I was thinking about this yesterday and I really believe the only thing Colin really "missed" was a proper identification of his core userbase (in that he judged bans virtually entirely on his statistics, which means many decisions were made based on the actions of bad players). And if you'll look at the complaints about the Bold Voting process, or the Skymin test, or even the Lati@s tests, many of them directly pertain to that issue. Do we care about the best players? The smartest posters? People who just play the most? Obviously this is getting into specifics which might not be entirely appropriate for this thread, but I never did see why the "standard" couldn't work for Smogon as long as we just applied it properly based on how community operates in general.

    For example, the fact that players are allowed to actually make direct changes to the metagame by voting is probably something to consider; when people hear "...and then you get to vote to decide whether it stays in the game or not!" that's a pretty big incentive to speak your mind about something you don't like, which I think is important. Tournament play has not really been particularly fleshed out in our community either, and the possibility of, say, Swiss/Best-of-3 becoming standard, or "pay to play" becoming more popular, could make interesting differences in how we deal with policy changes. Certainly everything you've mentioned in this thread regarding our "flawed" initial tiering could make a world of difference as well... there are just a ton of factors to possibly consider because this community does work differently from other communities, but I think the actual differences between the games themselves are either irrelevant, or are already reflected in the community anyway, and that this isn't really a matter of "how do we handle this game from here on out" but "who are the people, and what are the aspects of this community that we need to pay closest attention to?"

    Maybe I've just been describing exactly what you mean by "developing a new philosophy," and in that case you should just take this part of my post as "I guess 'develop' was just too strong a word for me." Either way I'm probably getting into "someone make a new thread for this" territory so I'll stop for now.

    I'm just not really sure what to think right now. If 5th gen were announced tomorrow, I know that I'd be pushing for us to start off with no bans; it's both logically and practically the best solution when you consider that there are no other relevant communities (not even Nintendo) that we really need to conform to, nor are there any particularly valid complaints considering that "this isn't 4th gen, go back there and play with Jolteon if you want." Banning would be a relatively straightforward process of us getting acquainted with the "pure" metagame and then banning anything we deem too powerful. We wouldn't have to deal with "but wait let's not ban Garchomp, let's just unban Latias!" for one thing, and there wouldn't be any reason to hold on to the "true metagame" considering that there wouldn't be one. So I don't know about "redefining OU," because where I'm standing the "there's a stigma associated with OU" argument is pretty weak in the context of an entirely new game.

    That said, I disagree with Obi that "it's never too late to do things right." If we're talking about the actual, current situation in 4th gen pokemon, I think I pretty much agree with Tangerine entirely. We have to recognize that there is incentive to "move slowly," and we could even be justified in just halting the testing process altogether (not that I'd agree with that by any stretch of the imagination), because allowing the metagame to actually settle permanently definitely has plenty of advantages. On a practical level, a 4th gen reset would definitely be iffy, and it's only getting iffier as time goes on. If nothing else, it's definitely not productive to try and undermine the Suspect Test process entirely at this stage, considering that we're theoretically moving towards the same destination anyway.
  4. Aldaron

    Aldaron wu tang clan ain't nothin to fuck wit
    is a Tournament Directoris a Battle Server Administratoris a Smogon IRC SOPis a Tiering Contributoris a Contributor to Smogonis an Administrator
    OU and IRC Leader

    Joined:
    Aug 5, 2007
    Messages:
    4,494
    The biggest thing I want to point out is that:

    We HAVE unbanned everything (by everything I mean Pokemon, not clauses). Note, the Ubers ladder. What we did not do is intentionally proceed from there with our suspect process and rigorously ban / unban 1 Pokemon each time...supposedly.

    However, I argue that, in fact, we DID do this (kind of). See, stage 1's most important aspect is that for a potential move, we test in a suspectless metagame. Therefore, we (referring to the collective, not any individual or groups of individuals, but literally the collective policy makers at smogon) determined the suspects with nearly 2 years of competitive DP knowledge, and we then banned them all. This may not have been done intentionally, but the beauty of it all is that it was done regardless.

    The initial ubers list was actually a primitive form of a suspect list (note the comparison to current UU). We spent 2 years of DP competition trying to weed out a successful suspectless metagame. We accomplished this with the banning of Garchomp.

    We also playtested a "no bans" environment for many months with the unveiling of the Ubers ladder. This served to confirm our primitive list of suspects.

    We are now rigorously testing each individual Pokemon for banning or unbanning.

    So for those claiming we didn't start this off properly...I ask...what are you talking about? If you ignore the timeline and intentions...our actions do represent a perfectly implemented suspect process.

    1.) We spend a LONG time determining suspects. Eventually, we have our list of suspects.

    2.) We unban everything in the Ubers ladder. This confirms our suspects.

    3.) We proceed with our suspect project.

    In fact, the only thing to really complain / comment on is the actual suspect process, which is a different matter altogether.
  5. Jumpman16

    Jumpman16 np: Michael Jackson - "Mon in the Mirror" (DW mix)
    is a Smogon IRC SOPis a Site Staff Alumnusis a Team Rater Alumnusis a Battle Server Admin Alumnusis a Researcher Alumnusis a Tiering Contributor Alumnusis a Contributor Alumnusis an Administrator Alumnus

    Joined:
    Dec 19, 2004
    Messages:
    4,768
    I already addressed in the UU PR thread why the "unban everything" mentality shouldn't be given credit until UU settles down and we're sure that there are only six Suspects (I personally wouldn't bet on it).

    Even if it were easy to define a set cutoff point like banning 600+ BST pokemon we would still need "subjective" characteristics to make this cutoff point work. I'm sure you realize that this would ban Slaking and Regigigas yet allow Wobbuffet, and were just using it as an example, but it nonetheless underlines the fact that keen thought still needs to be put into whatever cutoff we initially arrive at. And if we're going to have to do this, why not just use our intuition when Gen 5 comes around and ban the obvious Ubers but retest stuff like Garchomp and Latias if most of us can agree that they are, according to how the standard metagame is at the time, Suspects. As for right now though I don't see a reason to stray from our current cutoff or try defining one outside of what math stuff X-act can cook up.

Users Viewing Thread (Users: 0, Guests: 0)