Why do people want to ban more and more things?

just posting to say that I could not agree any more with what reachzero just said.

clearly broken and annoying are also undefined in your argument :p
Click to expand...
what competitive game/community does define this sort of thing though? I mean I know nothing about Magic or whatever, but when it comes to, say, the fighting game community (and, you know, the pokemon community up until around early DP), bans happen "when it's obvious." and obviously those games aren't "all statistics" like pokemon is or whatever, which is why I would have loved to know what MtG actually does since it's more easily comparable. regardless
 
I'd be interested to see how many people would vote Scizor or Heatran Uber, if they were tested. I think we can say that they're clearly not Uber, but I'm sure some people would vote them so. I'd just be interested to see exactly what happened.
 
SHUCKLE MAN said:
I'd be interested to see how many people would vote Scizor or Heatran Uber, if they were tested. I think we can say that they're clearly not Uber, but I'm sure some people would vote them so. I'd just be interested to see exactly what happened.
Click to expand...

People will always make stupid votes. Frankly, I don't care to know how many people would vote them uber. People's response to something they have trouble handling is to call for it's ban. Doesn't mean it should be given credence, or even given attention.
 
blame game i have no idea what you are arguing. you say "who does that? maybe these guys do i don't know".

my answer is, with a function this complex, why not optimize infinitely?

"when it's obvious" means we have a clear idea in our heads of what is "banworthy" ie huge stats/movepool (ubers) that make too many things not viable, bringing no net strategy to the metagame (double team/scarf spore loom).


essentially it's always an attempt to optimize diversity, and we're trying to get better and better at it. skymin was what many consider a "hiccup" but we're learning from our "mistakes".

i see no reason why we should not look at bans as a contributor to diversity optimization. optimal diversity clearly has a lower bound (form which we allow UU and BL to emerge) and an upper bound (Uber). so why not hunt for the ideal "metagame spot" with actual raw numbers?

People will always make stupid votes. Frankly, I don't care to know how many people would vote them uber. People's response to something they have trouble handling is to call for it's ban. Doesn't mean it should be given credence, or even given attention.
Click to expand...
so you ignore people struggling to come up with good answer to an apparently difficult question (ex: Is skymin uber)
but you listen when it's obvious (ohko moves are bad)? that seems like laziness =\

the fact is, we are going to figure out skymin's tier for sure and we are going to test SR. we just need time.
 
I haven't been around Smogon long enough to know, but what did we do before this? Were bans just easy until DP or something? Did we use stats then to find out if something was overused?

The more I think about this, the more it seems there is no good solution. I was under the same impression as Tangerine as to why we have public, non-bold voting, but people obviously don't vote for that reason.

I've heard a lot about how "bold voting can't work for Smogon", but rather than run around with a billion other, all fatally flawed systems, why don't we "fix Stark" to work around bold voting? You know, teach people about tiers, philosophy, and all that jazz.

I still think "bold voting" is the most workable of the many flawed setups we have, though an "objective" definition would work best if we could, you know, find one.
 
maybe for educational purposes, as it requires a certain level of education to bold vote, but i don't see how it's "useful" to bold vote. in the context of a suspect it's just a guess at what factors of a pokemon are the most important on the metagame.

"I want to enjoy this game the most i can" seems like a shaky premise to start with but most competitive battlers "enjoy" the same thing (namely an optimal amount of viable strategies, not too many that you can't teambuild with 6 pokemon, and not too few that you get bored using the same pokemon over and over)
 
I'm still not quite clear on the subject of what exactly bold voting is.

however, some kind of filter is required to keep noobs out of voting.

Maybe a 100+ post count or something, or mod approval.
 
Quick question from me:

If everything is overcentralizing, then are they really overcentralizing?

I hope you understand what I'm saying.
 
I think the post count idea isn't bad, but they've probably already toyed with it. Maybe post count in Stark Mountain? (In addition to the other requirements)
Moderator approval might be more difficult, though... They'd have to know the voter-to-be.

Meh, objective would provide a happy medium for most people, I think, if it was all sorted out. Thanks, X-Act!

To Osthato: Everything isn't centralizing at one time. It's only considered overcentralizing if one thing or a very few things is/are being centralized around at any given point in time.

I'm guessing I messed some definitions up, so someone please correct me. Thank you.
 
Nonono. Everything is "centralizing". We just don't know what "too much" (over) is.
 
Okay, I think I see. Is it that centralizing is having the opponent need something to take care of the thing and overcentralizing is having the same to an "unreasonable extent," which is what would need to be defined? Is that right? If not, my ideas can serve as exactly what not to say, at least.
 
@ Chris is Me: I seriously doubt that we'd be able to find a working defintion for terms like "overcentralization", "optimal diversity", and "uber".

The issue is, before we create these well-reasoned goals, our definitions must be solid and as least flawed as possible. The Smogon philosophy and the way in which members of the community, and by their influence, the community as a whole, has conducted itself through bold votes for suspects are not working off the exact same motive like they should be. This is what we should call "bias". Voter bias is a problem when prevalent because it misrepresents the communities goals and weakens the authority of Smogon (due to supposed hypocrisy).

The term metagame implies a certain degree of centralization, where there is an accepted, standard, strategy that people can use. The majority of players then adapt to this strategy finding a way to beat it or using it themselves. As centralization decreases, the possible viable choices to win in a metagame increase (which is most people's idea of a "most fun" metagame). It is a viable point to discuss if we are playing in a metagame since centralization will always be there. Bans *can* be justified by this decrease in centralization. We need mathematics and statistics to determine this decrease because unlike the popular vote, mathematics is always consistent in the way that it applies to various instances. X-Act's way of measuring centralization is by far the best way thusfar suggested to go about this.

Eliminating user bias is impossible, but it can be decreased to a point where it is insignificant. Again, such an insignificance would have to be dictated by a statistical calculation. One possible way to go about this is to have a questionaire about comprehension of Smogon's philosophy PM'ed to the voters who have identified themselves. These PM's would be monitored by admins if necessary (optimally, there would be a third party, but that becomes more complex). If they get x% or higher and demonstrate knowledge of the current metagame and suspect, then their vote will count. Unfortunately, this would hurt our voters that don't speak English as a first language. However, this is a forum that caters to mostly people who speak English well, so this would be an inherent hardship that cannot be worked around.

Infinite optimization is, in theory, the best way to approach this, but we don't have a large enough number of Pokemon in OU to viably do this. If we *HYPOTHETICALLY* removed Heatran from the metagame and replaced it with Manaphy, it may not change the centralization significantly. If it doesn't (a mathematical approach towards this is needed), then according to our philosophy, a ban isn't justified.
 
Okay, I think I see. Is it that centralizing is having the opponent need something to take care of the thing and overcentralizing is having the same to an "unreasonable extent," which is what would need to be defined? Is that right? If not, my ideas can serve as exactly what not to say, at least.
Click to expand...
that is essentially it mathematician.


@kingdrom i don't understand your argument. you admit that the level of diversity is bounded by the number of strategies people consider acceptable but then you say we'll never find a formula for "acceptable"?

you can only check so many threats with one team. your team needs to be able to check threats in one way or another. "team size" is probably a pretty good indicator of "how diverse we want our metagame" as it has a discrete upper bound.

i'd like to add the parameter:
centralization of a strategy=f(#viable checks to a strategy)
centralization=f(n_vc)

here as n_vc gets large, centralization decreases since the metagame doesn't have to "pool it's resources" to stop a threat.


in general, threats with more viable checks are less potent. the fewer the viable checks, the larger the threat.

garchomp's set is not "uncheckable" as long as you run a mono ice shard toxic spikes team you're fine right? but it just tstupid the stuff we had to do to stop it so we banned it. wobbafett is not uncheckable just run taunt pursuit spiritomb, sableeye and baton passers.

now all we need to do is approximate #viable checks≤amount of possible checks on a team≤24

I'm setting 24 as moves*pokemon since this is "theoretically" the "maximum number of strategies you can check" but there's often some redundancy. the idea would be to calculate an "average damage a team does to the metagame" (complicated but totally doable, xact has done it for individual pokemon!)

something like that.


overcentralization, what is healthiest for the metagame, etc are all subjective
Click to expand...
i have to disagree =0we just dont have explicit funcitons for them yet, unlike the things you mentioned ("goodness/battling knowledge"
=f(deviation, rating) )
 
luxormaniac said:
I'm still not quite clear on the subject of what exactly bold voting is.

however, some kind of filter is required to keep noobs out of voting.

Maybe a 100+ post count or something, or mod approval.
Click to expand...

Bold voting was a system where people posted "X Pokemon is Uber/not Uber" (in bold) along with an explanation. In order for your vote to count, your explanation had to be judged as reasonable and intelligent. The biggest problem with it was how hard it was to make judging truly objective - some people would judge an argument as good while others wouldn't, and the fact that not everyone is as good at putting their arguments into words as others hurts too. As far as the potential alternate account problem goes, your account had to be registered beforehand for a certain time period in order for you to vote.

Anyways, we're trying to create the most competitive metagame possible, and to that end, we've created a system that lets the players decide. The definition of Uber is debated a lot, but even when something definite is decided on, the conditions are not something truly mathmatical - overcentralization, what is healthiest for the metagame, etc are all subjective. Letting people who are experienced (deviation requirement) and fairly skilled (rating requirement) at the metagame seems like the best way to do it at this point - it's much better than the other way we've tried, and nothing else that's as good has been suggested. So it may seem like we're banning everything but the kitchen sink, but we're simply aiming for our overall goal, and doing whatever we need to accomplish that.

I feel the need to point out that the bans on Garchomp, Deoxys-S, and Shaymin-S are NOT permanent - they will be retested during stage 3 of the testing process (or in 1 month in Shaymin's case), so no moves regarding a specific Pokemon or attack have been 100% decided yet. So anything believed to be uber because "it's new so people are afraid of it" will have another chance once the hype has worn off.

Also, the "Shaymin-S Uber voters are crybabies who got butthurt because they were haxed by it" is stupid, largely untrue (reasoning in the voting thread doesn't have to be concrete, so it wouldn't matter if someone posted "Shaymin-S is uber because it looks gay", their vote would still count, and people who did seem to be voting due to one case of hax had better reasoning), and borderline trolling. Don't do it.
 
gorm said:
that is essentially it mathematician.


@kingdrom i don't understand your argument. you admit that the level of diversity is bounded by the number of strategies people consider acceptable but then you say we'll never find a formula for "acceptable"?
...

now all we need to do is approximate #viable checks≤amount of possible checks on a team≤24

I'm setting 24 as moves*pokemon since this is "theoretically" the "maximum number of strategies you can check" but there's often some redundancy. the idea would be to calculate an "average damage a team does to the metagame" (complicated but totally doable, xact has done it for individual pokemon!)

something like that.
i have to disagree =0
Click to expand...

I didn't focus on diversity as much as decrease in centralization, as when I made that post, I had not seen X-Act's diversity thread and was unaware that there was a determined way to compute diversity of a metagame. As you've implied before, centralization and diversity are somewhat inversely related (up until a point, a point you call optimal diversity).

Decrease in centralization would mean that the strategies needed to check against would become less prioritized, opening up diversity in team building. People want this diversity, and this similar motive that is still removed from Smogon's philosophy was the bias of voters that my original post was about.

What is the boundary to a strategy? Things like Garchomp and possibly Kingdra could be considered his own strategy without extra support. Stuff like Crocune + Toxic Spikes and Celetran + Gyarados could be considered a strategy, but when is the line drawn between things like Tyranitar and SD Luke with SD/CC/Extremespeed/Bullet Punch and SD Luke with SD/CC/Ice Punch/SE and Stealth Rocks ? These strategies are aesthetically similar but are different in execution and checks (one strategy gets destroyed by Gyarados and Salamence but the other cannot deal with Cresselia or Gengar well, etc.) It isn't the best example but it demonstrates my idea.

Lemme get this straight. Once the number of possible checks are determined, the usage of a threat would have to influence how, well, threatening it is. The number of checks possible on a team would then be a calculated value without the usage of the threat taken, and the viable checks would be the highest value with the usages taken into effect?
 
ok, let me try

a pokemon is uber if it forces the player to use checks and counters, but despite numerous checks and counters, it's usage continues to rise to an unacceptable level.

(all we have to do is determine what an unacceptable level is, I'm guessing on 60% of teams)

a pokemon is uber if it has no counters or checks, and has a visibly adverse effect on the metagame. not necessarily based on usage. (ie, deoxys-e and wobbuffet)
 
I always thought that a pokemon was Uber if it had checks and balances that failed some (if not most) of the time. That being said it also must be able to set up its team or its self with little or no cost to its fellow team members. Also if people start running a different set of a certain pokemon to counter the other regular set (Haban Berry Chomp).
 
skiddle said:
ok, let me try

a pokemon is uber if it forces the player to use checks and counters, but despite numerous checks and counters, it's usage continues to rise to an unacceptable level.

(all we have to do is determine what an unacceptable level is, I'm guessing on 60% of teams)

a pokemon is uber if it has no counters or checks, and has a visibly adverse effect on the metagame. not necessarily based on usage. (ie, deoxys-e and wobbuffet)
Click to expand...

What about say Heatran? Heatran was the most popular Pokemon in the game for a while because he countered so much, but he is also fairly easily countered himself. He even counters himself fairly nicely. If he hit 60% of usage on all teams though I'd argue that removing him would be worse for the metagame at large, not better, because he controlled so many viable threats.
 
Heatran wasn't #1 long enough to meet Skiddle's def. It has to continue to rise despite countertrends.

Serene Grace actually proposed that, skiddle.
 
Blame Game said:
just posting to say that I could not agree any more with what reachzero just said.

what competitive game/community does define this sort of thing though? I mean I know nothing about Magic or whatever, but when it comes to, say, the fighting game community (and, you know, the pokemon community up until around early DP), bans happen "when it's obvious." and obviously those games aren't "all statistics" like pokemon is or whatever, which is why I would have loved to know what MtG actually does since it's more easily comparable. regardless
Click to expand...

There are two instances in recent history that there was a Standard Ban. The most recent was the Affinity Deck. It was hands-down the best deck in the metagame. Even more than that, it composed at least 60% of the metagame for most of its existence, and the majority of the remaining metagame was "Anti-Affinity" decks that frequently packed upwards of 16 maindeck hate cards (when the number of non-land cards in a deck is usually 36-40) and still rarely managed a greater than 50% matchup. So the ban-hammer came down, removing eight key pieces of the deck (they really wanted it dead). This is a good example of something like Garchomp: it is so overcentralizing to the metagame that it shapes the entirety of every team (or in this case deck) around this threat.
The other instance was Skullclamp, which was a bit before the Affinity ban. Skullclamp was a card that could be used in every single deck. It allowed you to attach it to any creature and it would give them +1/-1, and when they died, you drew two cards. What this meant is that any 1-toughness creature could be cheaply traded for massive card advantage (card advantage is a primary factor in winning games). The two most powerful decks in the format (Goblins and Affinity, before the rotation that made Affinity the only viable deck) could use it very well, so they became most of the format. Wizards realized there was a problem when Tooth and Nail, a slower, control-based deck (so one without 1-toughness creatures to exploit Skullclamp) had a resurgence by switching its basis to one with 16 one-toughness creatures and Skullclamp. This would be more comparable to something like Stealth Rock, if Stealth Rock dealt twice as much damage. Every team could use it, and the teams that did not use it would be obsolete.
 
Chris is me said:
Heatran wasn't #1 long enough to meet Skiddle's def. It has to continue to rise despite countertrends.

Serene Grace actually proposed that, skiddle.
Click to expand...

cool beans, I guess that wasn't such a bad definition after all.

Anyways, the heatran thing doesn't have an adverse effect on the metagame, heatran counters a lot of stuff, but doesn't do much harm to the average team.

I need to upgrade those rules a bit. The second rule I don't like that much, it should correspond with the first more.
 
60 % of all teams is extremely high for the usage of one Pokemon, skiddle. Not even Garchomp reached that high. X-Act had published statistics up back in August or something that showed the individual usage of Pokemon in OU. Of all the Pokemon, Garchomp had about a 5% chance of being in one of the slots on a team. Because of species clause, Garchomp must have been on less than 30 % of teams in OU.

Would it really matter if a Pokemon was not #1 anymore but continued to rise in usage? Lets say that somehow, both Scizor and Salamence are on 60% of teams (again, unrealistic but whatever). Despite Scizor having many checks, he is a useful Pokemon for lategame sweeping, and his valuable resists (dragon, dark, and ghost in particular),and a decent offensive movepool to work with mean that he doesn't have trouble finding a place on a team. Scizor's popularity in a way is correlated to Salamence popularity because Salamence gets a relatively easy switch-in on CB Scizor and can deal with Bullet Punch Scizor reasonably well. They continue to rise in usage, with Scizor and Salamence being able to deal with different sets of each-other. We would know Scizor to be extremely useful but easily countered (like Heatran), but Salamence's power is unknown. Does the extra Scizor usage mean that Salamence isn't worthy of uber just because it doesn't have the #1 usage position?

In case this wasn't clear. I do NOT think that Salamence is Uber in any way (yeah, it's a bad example :P).
 
I thought there was a 60% chance of him being on a team near the end, I'm sorry, too high of %. Had I known it was lower I would have set the bar lower :P
 
I dunno, I was tired when I made the first rebuttle and anger kind of took over me. Saying sorry isn't much Tangerine but yeah, my arguments, after looking at them carefully weren't very good.

But on a more serious note, it's rather... unclear, how to handle it. I was fully aware why Garchomp had to be banned in Pre-Platinum. It was overcentralizing, something I won't argue against. And Deoxys-S, I wouldn't argue that it wasn't a bad vote because most of us were aware of its capabilities and such, and it overcentralized yadayada. It's just the Shaymin-S voting that kind of drew a question mark on how it was overcentralizing. I'm not feeling into arguing it down even further so I'll leave it to others.
 
skiddle said:
I thought there was a 60% chance of him being on a team near the end, I'm sorry, too high of %. Had I known it was lower I would have set the bar lower :P
Click to expand...

I think the statistic you are thinking of is from the top X (50 maybe?) teams on the Smogon ladder. Something like 95% of them ran Garchomp, IIRC.
 
You must log in or register to reply here.
Back
Top