Using simple majorities for tiering votes

[Cathy]

Banning a pokemon from a tier is always a serious decision. It is not something any of us does casually as it removes an aspect of the game. Even if that pokemon is broken, it deserves to be seriously considered before it is banned.

Traditionally, we have used a system where a simple majority is required to reach a conclusion on an issue. However, there is nothing significant about a simple majority when it comes to banning a pokemon -- potentially half the sample disagreeing a pokemon should be banned is a big chunk of the relevant people, and not enough to justify banning it from the tier it is presently in.

In order to recognise the seriousness of moving a pokemon into a higher tier, I recommend that we switch to a 2/3 majority system. Specifically, if exactly 2/3 of the voters, or fewer, vote to move the pokemon into a higher tier, it should stay in the lower tier; only if more than 2/3 of the votes are to move the pokemon into the higher tier should it happen.

Requiring a supermajority for serious decisions is a long-standing tradition of many decision-making bodies, and I recommend that it be adopted here, in order to reduce "splits" in the community over what should be banned, and to recognise the seriousness of banning a pokemon.

So far among the previous suspect votes, there have been a few pokemon where the votes have been very close to 50-50. These controversies could all be avoided with a higher bar to ban a pokemon, while simultaneously making the choice to ban a pokemon more legitimate.

 

[Tangerine]

In order to recognise the seriousness of moving a pokemon into a higher tier, I recommend that we switch to a 2/3 majority system. Specifically, if exactly 2/3 of the voters, or fewer, vote to move the pokemon into a higher tier, it should stay in the lower tier; only if more than 2/3 of the votes are to move the pokemon into the higher tier should it happen.

I agree. While our current system doesn't necessarily move them to a higher tier since they will be retested in the later stages - I think the bar should be much higher than a simple 50/50. When no clear majority can decide in one way, then it's best to assume "not broken" and let people tinker with the "suspect" more.

 

[RBG]

While you mention moving a pokemon to a higher tier, you forgot to include that we are currently testing pokemon that were originally in the upper tier for OU, does this mean that it needs 2/3 of a vote to move it down a tier?

 

[Cathy]

What's important is making it harder to move pokemon to a higher tier. In a debate about whether to move a pokemon down, keeping it in the higher tier would still require the bigger majority; if only 1/3 of the voters wanted to move it down, it should be moved down.

 

[imperfectluck]

Requiring supermajorities would turn voting into something akin to the United States Congress, in which it would become incredibly hard to ever change anything. Sticking with simple majority for making changes is what I prefer. Besides, why should it apply only for "moving pokemon to higher tiers?"

 

[Veedrock]

Shouldn't it be the other way around? In decision-making bodies, if 2/3 don't approve the measure, it doesn't pass. Since the measure in the suspect process is technically "do we unban it?", shouldn't it require 2/3 majority to be unbanned? Your proposal seems backwards to me at the moment. (Perhaps the "measure" can be debated either way, that's just how I see it).

Basically what DP-C sees. I disagree with Tangerine that if it's not supermajority it's "not broken;" it's moreso broken than not, and the votes reflect this. And isn't Stage 3 where the extra tinkering comes in?

EDIT: I haven't decided either way, but I don't understand why the unban requires the minority. It reduces error in one regard, but increases it in another.

 

[Cathy]

Banning something is a serious decision--that's why. In competitive gaming, we are careful to avoid banning something unless it should really be banned.

If more than 1/3 is required to move a pokemon down, what you are saying is that there is a really good reason for it to be banned in the first place--but if there were such a really good reason, then 2/3 of the voters would agree it should be banned, so it would not be possible to collect 1/3 votes to move the pokemon down.

This is about making it harder to move pokemon up in the tiers, not about making it harder to change the status quo.

 

[reachzero]

I strongly agree that unless there is clear community consensus that a Pokemon is broken after testing, it is unwise to ban it. If only 51% of the community feels that it is broken that is not consensus, only an indication that community opinion is inconclusive. 2/3 sounds very fair, although I think as high as 75% would not be unreasonable. If a Pokemon is actually broken, it should be clear enough that more than half of the competitive community can see it.

 

[Syberia]

It has always been my opinion that the majority of people who play the game should make the decisions concerning the game, without anything more than slight (and on objective grounds) interference. It's better to have 49% of people disappointed with an outcome than 66%.

 

[SoT]

I kind of agree and disagree here. 2/3 would reduce the amount of controversy in some aspects, however as IPL said not much at all will really be accomplished if we raise the bar. And if we do go with this "super majority" we need a set 2/3 for both, or 1/3 for both. It seems silly having 2/3 for moving up, and 1/3 for moving down.

Edit: After reading Obi's the supermajority makes more sense for being banned. I still say a 1/3 for being moved down is a bit odd, if we do a 2/3 moving up, 50/50 for moving down, that seems like it'd be a better idea to me. It would provide a more "accurate" decision in my opinion.

 

[JabbaTheGriffin]

What Syberia said. How can we leave something in the game that 64% of players think is broken. We're out to please as many people as we can, not to piss off just about everyone.

 

[obi]

I agree with having more than a 50% requirement to ban things for one simple reason:

If a Pokemon is banned, it is no longer being tested in the relevant tier. In other words, banning ends the discussion.

When a Pokemon is not banned, it continues to be tested in that tier, and can be brought up for a vote later with everyone more informed on the subject. If a Pokemon does prove to be broken, it can be banned at a later date. Votes to not ban are not permanent because the Pokemon's presence remains.

In other words, I support only requiring the supermajority to ban a Pokemon.

 

[lilyhollow]

I agree with this idea, but I'd just like to throw it out there that purely preserving the status quo has its practical merits too (so I can see RB's viewpoint that a supermajority should be required for any change in tiers whatsoever). I still agree with Obi and Colin that a supermajority should be required to make/keep anything Uber, but if 4th gen Pokemon were destined to die in twelve months or something, I'd argue that the healthiest decision would be to pretty much just stop changing the game period.

 

[QibingZero]

I'd first like to say that I wholeheartedly disagree with the idea that any bias should be placed on moving pokemon up or down in tiering. Pokemon is wildly different from most competitive games, and mostly so in the fact that it is actually very uncompetitive at base level. It's certainly reasonable to assume that a game which is less competitive from the start would require more rule-based changes to gameplay, and thus making bans more difficult to obtain is often counterproductive to the goal of creating a more competitive game. I believe we should be open to changes which further that goal, regardless of whether they end up being bans or unbans.


Noting Obi's argument, I don't believe that a banning should even end the discussion absolutely (though it would likely make it more difficult). Technically, we originally banned Latias - albeit in a different fashion - and yet we now consider it to be fine in OU. There is nothing to say that a pokemon might not change tiers after events like Platinum come about, either. Or that after the first round of a suspect process is completed, it would be more reasonable to consider other bans or unbans. I'm sure that if Skymin ends up staying Uber after stage 3, plenty of people will want to discuss it again.

This also has huge implications for UU, which will likely see at least one new pokemon from OU every 3 months in addition to whatever happens during it's own suspect processes. It's not hard to imagine a scenario in which one of the pokemon banned to BL could be reasonable for return to UU when top-tier threats are added every time a pokemon drops from OU.


That aside, requiring a 2/3 majority overall is a reasonable line of discussion. I'm just not sure I agree that it's needed. There are troublesome scenarios for both, and it's enough that I would say there's no reason to change what we have going. Maybe I just disagree with the premise though. The only problems I have with the current system are not at all touched by changing the way the voting is counted.

 

[haunter]

I agree with Syberia, demanding a majority of 2\3 to move up a pokemon could end up with the paralysis of the testing process.

About banning pokemons, instead, I beleive Obi is right (though for a different reason).
Since banning a pokemon should be an extrema ratio according to Smogon philosophy, a larger consensus would justify more the possible ban toward the community.

 

[X-Act]

What about unbanning a Pokemon? Would it require the same criteria?

 

[matty]

Requiring supermajorities would turn voting into something akin to the United States Congress, in which it would become incredibly hard to ever change anything. Sticking with simple majority for making changes is what I prefer. Besides, why should it apply only for "moving pokemon to higher tiers?"

I agree that 2/3 of the votes needed is the best course of action (e.g. the latest Froslass vote which was such a narrow margin). IPL makes an excellent point however in that it would make the likely hood of things being banned increasingly hard.
I also don't agree with this policy directly related to just moving pokemon to higher tiers; it should be equal moving them up and down.

 

[Jumpman16]

I strongly agree that unless there is clear community consensus that a Pokemon is broken after testing, it is unwise to ban it. If only 51% of the community feels that it is broken that is not consensus, only an indication that community opinion is inconclusive. 2/3 sounds very fair, although I think as high as 75% would not be unreasonable. If a Pokemon is actually broken, it should be clear enough that more than half of the competitive community can see it.

tell that to latios, which would be in ou right now if this had been in place

When a Pokemon is not banned, it continues to be tested in that tier, and can be brought up for a vote later with everyone more informed on the subject. If a Pokemon does prove to be broken, it can be banned at a later date. Votes to not ban are not permanent because the Pokemon's presence remains.

this is still a direct application of what i mean when i say "this is why we have stage three" even if the reasoning is more subtle. if latios had been voted ou by failure to reach a supermajority for banishment, then what? if it does "prove to be broken", at what point do we ban it again? and would this not require the same supermajority? and does this not ignore the entire purpose of stage three, to see whether latios or any suspect is correctly tiered in the "true" metagame, where we play everything out with suspects that may belong in ou like garchomp and skymin? instead of pretending that we can attribute any lasting meaning to any kind of "consensus" before stage three plays out, and pretending that latios breaking a metagame with just it and latias as suspects means anything?

now, i have planned for almost a year now to either recognize or utilize majority (the former referring to why i want everyone who's given rights to vote). i haven't posted about it yet because it doesn't matter and wont for a while yet

 

[X-Act]

tell that to latios, which would be in ou right now if this had been in place

I think what Colin is saying is that a Pokemon goes to OU if more than 2/3 vote for it to be OU, which isn't the case for Latios.

The way I understood it, what Colin is suggesting is that you move down a Pokemon from Uber to OU if more than 2/3 feel it should be OU, and move a Pokemon from OU to Uber if more than 2/3 feel it should be Uber. The same holds for UU and BL.

 

[Lemmiwinks MkII]

I agree that 2/3 of the votes needed is the best course of action (e.g. the latest Froslass vote which was such a narrow margin)

It wouldn't just be Froslass, but also Raikou and Gallade. This decision will have a massive effect on the current UU should we go ahead with this. That said, I'm in favor of such a proposal as I've always believed that a Pokemon should not be banned unless it is absolutely necessary, and a required supermajority among qualified voters would help reduce the uncertainty somewhat. What we're not striving to do is this:

We're out to please as many people as we can, not to piss off just about everyone.

We're out to do neither. I don't recall anything in Smogon's Philosophy that talks about making people happy with respect to the metagame.

I think what Colin is saying is that a Pokemon goes to OU if more than 2/3 vote for it to be OU, which isn't the case for Latios.

The way I understood it, what Colin is suggesting is that you move down a Pokemon from Uber to OU if more than 2/3 feel it should be OU, and move a Pokemon from OU to Uber if more than 2/3 feel it should be Uber. The same holds for UU and BL.

Not the way I read it:

In order to recognise the seriousness of moving a pokemon into a higher tier, I recommend that we switch to a 2/3 majority system. Specifically, if exactly 2/3 of the voters, or fewer, vote to move the pokemon into a higher tier, it should stay in the lower tier; only if more than 2/3 of the votes are to move the pokemon into the higher tier should it happen.

Doesn't say anything about the reverse process.

 

[Veedrock]

I think what Colin is saying is that a Pokemon goes to OU if more than 2/3 vote for it to be OU, which isn't the case for Latios.

The way I understood it, what Colin is suggesting is that you move down a Pokemon from Uber to OU if more than 2/3 feel it should be OU, and move a Pokemon from OU to Uber if more than 2/3 feel it should be Uber. The same holds for UU and BL.

What's important is making it harder to move pokemon to a higher tier. In a debate about whether to move a pokemon down, keeping it in the higher tier would still require the bigger majority; if only 1/3 of the voters wanted to move it down, it should be moved down.

Pretty clear that it doesn't work in reverse.

With the "2/3 majority for either decision," what happens when neither is reached? That's why that 2-way system wouldn't work, because if the process fails, the failsafe (no tier change or the opposite) is the result of a bias that the 2/3's system creates regardless. So either way, if this is decided to take effect, the bias would have to be chosen for 2/3 (unless there's some alternative to neither side reaching majority I'm missing).

EDIT: Alright, I misunderstood. Figured you meant on every case it'd need it either way, not a bias towards it's current tier on a case by case basis.

 

[X-Act]

My point is the following:

Say we're testing Mewtwo as a suspect for OU. Where was Mewtwo before the test? Uber. Hence you'd need to get 2/3 vote for "Mewtwo OU" in order for it to go to OU. Otherwise it stays Uber.

Say now we're testing Salamence as a suspect for OU. Where was Salamence before the test? OU. Hence you'd need to get 2/3 vote for "Salamence Uber" in order for it to go to Uber. Otherwise it stays OU.

Either way, it's pretty clear what happens to the Pokemon if the 2/3 vote isn't satisfied.

 

[RBG]

actually, I was talking to colin last night and (correct me if I'm wrong) he said, the default should be unbanning, and if more than 1/3 thought it should be in the lower tier, than it should be tested there until you can reach the 2/3 thinking it is uber

 

[X-Act]

actually, I was talking to colin last night and (correct me if I'm wrong) he said, the default should be unbanning, and if more than 1/3 thought it should be in the lower tier, than it should be tested there until you can reach the 2/3 thinking it is uber

This doesn't make sense. If "banning a pokemon from a tier is always a serious decision" (which I agree with), then unbanning a Pokemon from a tier is an equally serious decision. Which is why I would support either leaving everything as it is, or do what I'm suggesting two posts above this one.

 

[Cathy]

The idea is that in competitive gaming, we always err on the side of having a pokemon in the lower tier. If you don't agree with that then obviously you aren't going to agree with this proposal. I already explained why only 1/3 should be needed to move a pokemon down:

If more than 1/3 is required to move a pokemon down, what you are saying is that there is a really good reason for it to be banned in the first place--but if there were such a really good reason, then 2/3 of the voters would agree it should be banned, so it would not be possible to collect 1/3 votes to move the pokemon down.

The whole point of this measure is to make the status quo not matter, and to make it harder to have pokemon in higher tiers--the point is not to enforce the status quo.

With the "2/3 majority for either decision," what happens when neither is reached?

Actually in my proposal in the first post I said it would be in the lower tier or moved to the lower tier with exactly 2/3 votes.

tell that to latios, which would be in ou right now if this had been in place

Why is this a problem? If there isn't widespread agreement to ban it, my premise is that we err on the side of leaving it in the lower tier for futher analysis. The "harm" to leaving a controversial pokemon in the game is less than the harm of pre-emptively banning something before there is a large consensus (i.e. a supermajority).