Stage three and beyond.

[Hipmonlee]

The suspect test process is drawing to a close and this thread is really covering two issues.

The stage 3 testing process is beginning, but it isnt completely defined in terms of exactly how things will work out. So I am hoping we can work out something here.

Related to this, is what will happen in future when we want to test something. I think the suspect test process has strengths and weaknesses and I think now we have some experience to fall back on, I think hopefully we can plan for a solid system that can be applied even in future generations.

I think it is important to understand how we will be able to make decisions in the future when we decide how we will approach stage three voting. I mean, obviously we want stage three to be as final as possible, but how easily we will be able to revisit things I think should affect how comprehensive we should be in our testing before the end of this process.

So I am going to start from a philosophical level and then try and get more specific toward the end.

Ok, well the goal of our tiering is to create a ruleset that results in a balanced metagame. The assumption is that balance results in a good game, and that a good game is our goal.

This goal trumps other considerations such as what is wanted by individual players. The point here is we are essentially the caretakers of the rules of competitive pokemon, we need to protect the integrity of the game from the personal preferences of current players on behalf of all future players. If we fail to do this, then in the future Smogon will lose its market, either to another pokemon site, or to another game entirely. Seeking balance is the only way we know how to do this.

This is all well and good, but at some point we need to consider practicality. It is no good having a process that guarantees the best possible metagame, if that process breaks down every time we try to implement it. The example of the official servers failure to introduce Wobbuffet shows that philosophical purity is only worthwhile if people are able to accept it.

The point is we have to come up with a system that is philosophically pure, but still acceptable to the general population. I think the suspect test process has been mostly successful in this regard.

But, practicality goes beyond public acceptance. I think that, at the moment, the general consensus is that the suspect test process in its current form, while it is more or less working, has just been too hard. The impression I get is most people wouldnt want to go through this again. Perhaps part of that is the fact that this is our first time through, and that maybe the second time it would be easier.

So I am going to identify some areas where I think the suspect test is not adequate for the future, at least in its current form.

• Deciding suspects. We havent really got an effective system for this, we just kinda wing it.. People seem to be accepting it though, so this seems kinda minor.
  
• Voters. There havent been enough, at any point. It has been a big dissapointment, and I would say that of anything, this is the real problem with the suspect test process. But I also know how difficult it is to make the commitment to qualify. Several times I tried and gave up just because of the time needed.
  
• Timeframe. It has taken us a year. To be honest I think that a year is probably a reasonable timeframe, to take an untested metagame and create something with a sense of finality about it.
  
• Effort. I feel so relieved this process is finally finishing, and I havent even been heavily involved since the deoxys test. People like Aeolus, Jumpman and Doug have really given a fuckload to this, and I wouldnt like to be the one to ask them to go through it again.

There have been some other less tangible reasons for opposition to the process. Suggestions like the tests are artificial and are cultivating biases for various reasons. These are hard to prove but are surely worth noting. I'll leave it to others to suggest these in this thread.

The solution I have to issue 2 and 4 and the claims of artificiality is we need to test and have votes based on the standard ladder. It will result in more voters, and it asks less effort of them, as they presumeably already battle on the standard ladder. I originally was going to justify this by arguing that people are already accepting regular changes to the OU tier throughout this process, but then I realised that if we just continue with the process as it is, and then whenever the status quo is changed, we change the standard ladder and a month later test based on standard ladder results within that month. What we need to do is during this period calculate two seperate ratings for battlers on the OU ladder. The normal rating, which wont be affected at all. And a second "suspect rating" which would be reset to the default position at the beginning of each test. This way people wouldnt complain about losing their ratings, but people also wouldnt qualify without a lot of battling during the month.

Also for future generations I would recommend that we start with all suspects unbanned in OU. Then preferably test by subtraction rather than addition. IE test the stage 3 metagame first, and then test removing pokemon rather than adding them. This is just because the philosophy states that we ban only when we are certain. Obviously we couldnt do this pracitcally this time, but in future I think it would be a better idea. Mainly because this way we force changes we make into OU, like if we unban a pokemon on the OU ladder, and people just dont use it, we dont gain anything by testing. But if we unban something, then people are guaranteed to be testing what we want them to..

So when a new gen comes out, the process would be:
Theorymon OU and some suspects. This will probably be largely based on >600 base stats are probably uber.
Then allow all the suspects, and test the metagame for a month or so, to make a priority list of pokemon. This can be done by Badgeholders/PR members whatever. Ideally if a suspect is an obvious counter to another then that pokemon should be tested after the pokemon it counters. Hopefully you wont get circles lol.. Perhaps someone can come up with better details..
Then we have a suspect test for the first suspect for a month, if it is voted uber, we change the rules for OU.
We repeat this process.
And as we repeat this process, whenever that process results in a change to OU we have a months worth of keeping track of the scores of the OU ladder as though ratings had been reset at the start of the month. TRhen at the end of the month we have a vote whose result will be final.

I am unsure if a super majority is necessary or not, remember this pokemon would have been ou at first, so to be voted uber it needs to pass two votes, which will both probably will have reasonably different voters as the second vote can be expected to be significantly larger.

And now that the suspect process is more or less finished and we want to make changes, we just need a method of deciding when to suspect test something. Once we do that, we can have a month of suspect testing, then a month of OU ladder with a second suspect rating and a vote. Now that we have the groundwork in place I think the work needed should be seriously reduced, the hardest part will be deciding what to test. I dont have an elegant solution for this..

What I would also suggest is that if the stage three vote should overturn the original vote, then we can set up a suspect rating like I suggested on the on the OU ladder for a month (or two?) after that vote, and then have an additional vote that will be the final decision on the positioning of that suspect. Hopefully that vote will be large enough to be conclusive (worst case scenario it has won 2 out of 3 votes [a supermajority!])..

Have a nice day.

 

[Tangerine]

I agree with your approach, Hip. However, I disagree with the notion of deciding suspects and what not.

If we are to have suspects, I rather not have a formal testing scheme like we had before. Rather, I would prefer to have them cycle through the metagame as in "seasons", culminating in a Smogon Tournament or a Tour (Sound familiar?). I think this will keep the metagame interesting and changing.

I also don't think we should even bother voting. Having the staff decide is literally just as good, and some might argue, a better decision, with a lot less work. People can definitely have a voice - they can complain, start discussions, and the staff can read and consider them when they make their decisions. I don't see why we should involve the public again this time around.

Tiering should also be emphasized to be practical and not so idealistic like we have now to avoid nonsense. I disagree with anyone who mindlessly apply Sirlin to Pokemon thinking that fighting games will fall under the same format. We need to decide based on what is practical and what we know instead of appealing to impossible ideal and attempting to work with it.

 

[Hipmonlee]

How do you propose the staff will make these decisions?

The staff have definitely screwed up in the past. Justin made all sorts of questionable tiering decisions when he was in charge. And in DP none of us wanted Latias..

I really like the vote concept, it isn't flawless, but it is good enough. And I think it is more pallateable to the general population. None of the decisions so far have been heinously bad, which I dont think you could say about having admins just decide by themselves (though to be fair, admins have had more chances to screw things up).

They key is just to simplify the process, so it runs a lot smoother..

Have a nice day.

 

[Tangerine]

How do you propose the staff will make these decisions?

The staff have definitely screwed up in the past. Justin made all sorts of questionable tiering decisions when he was in charge. And in DP none of us wanted Latias..

I really like the vote concept, it isn't flawless, but it is good enough. And I think it is more pallateable to the general population. None of the decisions so far have been heinously bad, which I dont think you could say about having admins just decide by themselves (though to be fair, admins have had more chances to screw things up).

They key is just to simplify the process, so it runs a lot smoother..

Have a nice day.

I don't think I ever promoted staffs just making decisions "because they feel like it".

The point is - listen around. Read posts, listen to what people say, and then see if they have merit or not based on some theory, and act accordingly. Surely, we can get staff members to do that.

 

[Hipmonlee]

I dont think justin was doing things just cause he felt like it..

I mean, for instance if Obi was making all the tiering decisions, we would have a very different ruleset than if I was making them. From my perspective, it is much easier for me to accept a tier list created by a vote than one created by Obi.

It seems to me thatthe hardest part of this process was putting it into place, (as well as getting people to participate, which I think I have come up with a great solution to). I dont necessarily think a vote will even be more work than having people sort through pages and pages of arguments and then making a decision themselves..

Have a nice day.

 

[lilyhollow]

I would prefer to have them cycle through the metagame as in "seasons", culminating in a Smogon Tournament or a Tour (Sound familiar?). I think this will keep the metagame interesting and changing.

There are infinite ways to keep Pokemon as a whole interesting and changing; it's well beyond unnecessary to force the same of the Standard metagame. In fact, I find it somewhat irresponsible to treat it as something that requires our constant prodding to "keep being interesting," unless it is proven through nothing less than extended, top-level competitive play that that is the case-- and at that point, I'd strongly expect something "obviously broken" to arise that we could deem responsible for such an environment anyway. The fact is, there are a good number of Advance players who will tell you day in and day out which generation of play far and away puts the others to shame (guess); if your suggestion were ideal, they wouldn’t be talking about the gen that has been significantly changed by Nintendo/Smogon a total of maybe three times ever, four tops. That isn’t to say that I have something against the idea of “constant change” itself, but as the standard method of play? With gimmick tournaments, multiple tiers, doubles, and suggestions like yours being perfectly available as secondary options to pretty much anyone who just happens to feel like playing with them, it’s a serious waste that effectively prevents us from exploring aspects of this game that clearly mean something.

As for the banning process itself, I pretty much agree with everything Hip has said so far. I don't think we stand to gain much by ditching the voting system. I do still believe firmly in supermajorities though.

 

[Tangerine]

I dont think justin was doing things just cause he felt like it..

I mean, for instance if Obi was making all the tiering decisions, we would have a very different ruleset than if I was making them. From my perspective, it is much easier for me to accept a tier list created by a vote than one created by Obi.

It seems to me thatthe hardest part of this process was putting it into place, (as well as getting people to participate, which I think I have come up with a great solution to). I dont necessarily think a vote will even be more work than having people sort through pages and pages of arguments and then making a decision themselves..

Have a nice day.

I'm not promoting a dictatorship here. I'm saying the staff - see: multiple users - should decide without attempting to "educate" the public every damn time within some limited time frame. People don't have to sort through pages and pages of arguments - if they're a mod, they should be reading through arguments anyway. Voting is a lot of extra work unless we're literally letting people vote on preference and that's that. If that's the case, then sure, you can do voting. But if you're serious about leading the metagame to a given direction based on a theory, then 99% of the people will remain ignorant of this given theory and it's not going to work out very well. Smogon needs to formulate rigorous theories about metagames instead of using weasel words to get out of it. If you want a head start I recommend looking at the Suspect Test, it's built on some pretty interesting theory and can be measured.

If something is obviously broken, then we can ban it... Of course, not everything is Kyogre, it's closer to Deoxys E. It takes time for people to learn how to properly use it and it took ages for Deoxys E. Sure, there are flaws within the "seasons" idea in that it's unlikely that people will ever develop indepth intuition to Pokemon, but at this point, I wonder if that even matters since good players can do it anyway. This is also a reason why I think our current analysis is just a lot of extra work and why we should go with a general analysis format. I would love to hear about how you can keep the standard metagame "interesting" instead of letting it lead to some obscure equilibrium point.

I also think we should completely revamp the tiering system since it's flawed with the current philosophies at hand. The Tiering system will prove to be broken if we ever started with a "no bans" environment within months. The current restrictions within the Tiering system is way too strict and affects the metagame adversely and really it's sort of funny how people can't see this. This will prove to be obvious, again, if we ever go with a no ban approach. But even if we are going with Hip's approach, I propose that the Tiering system gets a heavy revamp.

but hey, i'm done making policy here, and at this point, also done making suggestions to this site so...

 

[Hipmonlee]

Ok, but the staff are never going to agree.. We could have a vote of staff members, but is that really going to be easier than having a vote of people who accomplish some battling criteria? I still believe that if we give good direction, then at least the majority of people will vote in good faith.

I think the key to keeping the standard metagame interesting is for it to be balanced. There are plenty of useable pokemon to be experimented with, the key is to have a ruleset that just gives them half a chance.

Have a nice day.

 

[Tangerine]

Ok, but the staff are never going to agree.. We could have a vote of staff members, but is that really going to be easier than having a vote of people who accomplish some battling criteria? I still believe that if we give good direction, then at least the majority of people will vote in good faith.

I think the key to keeping the standard metagame interesting is for it to be balanced. There are plenty of useable pokemon to be experimented with, the key is to have a ruleset that just gives them half a chance.

Have a nice day.

The staff never needs to completely agree... I don't see why that stops the staff from making decisions. Which is still, pardon me, a lot easier than going through with suspect tests again. I don't mind people voting in good faith, that's not the issue - the point is either way we'll get similar results and one is obviously a lot less effort.

Please define balanced. If they are to be balanced, how many Pokemons should make up this balance? Balance is another euphemism for "good" considering how poorly it's defined.

If there are untested Pokemon, then the metagame isn't balanced yet, obviously. Balance will occur only when every combination has been tested and there's an equilibrium of certain number of kinds of teams/Pokemon/however you want to quantify it. Even then, it probably needs a cutoff anyway. Balance is a poor goal until you can define it, and find variables that shows the definition.

 

[Jumpman16]

For the record, it was never my initial intention for the standard ladder to change with every decision on a Suspect. I never wanted the standard ladder to change or have the potential of changing every month, as the latter is actually worse when you consider x Suspect voter possibly voting Suspects uber because the voter is enjoying his or her highest ladder ranking ever. This is a kind of bias that may not necessarily surface intentionally, since you can hardly blame anyone for wanting to fix something that, according to their own performance, ain't broke.

So in the future, I planned on making it more clear that decisions we make on Suspects should have as little impact on the Standard metagame (and, therefore, rankings) as possible. Whether we make this possible by having two different rankings on Standard or by keeping Suspects off of the Standard Ladder till they've been tested in Gen 5 is something we can decide on later and when (and how) we actually decide on Suspects. I think it's obvious that starting out with Suspects like Garchomp, Latias, Latios and even Manaphy in standard is something "we" should consider before just starting Standard play on Gen 5. Doug can tell us how practical or even possible it is to have two different rankings on the Standard Ladder.

It makes the most sense to not completely forget the Stage 3 results of any Suspect, and use this knowledge in deciding which, if any of them, need to be retested when Gen 5 starts. If, using what we learned about Deoxys-S, we do not see any new moves or pokemon to warrant labelling DXS a Suspect in Gen 5, we won't, until and only until it becomes incredibly obvious that DXS will play any different in Gen 5 than it did in Gen 4. The same goes for something like Latias that is very likely going to be voted OU by a wide margin in Stage 3 as it was in Stage 2. If Latias doesn't get moves Aura Sphere and Heat Wave in Gen 5, there is little reason to believe that the supermajority it will likely have as OU votes for Stage 3 aren't good enough for the start of Gen 5. In addition, we will have the benefit of analyzing any Suspects that are given Stage 3 OU tags in the actual Standard metagame, because they will, by definition, no longer be Suspects.

Something like Skymin that may get a vote of, say, 54% uber in Stage 3 is what we really care about. First, we have to decide whether this is enough to keep it out of Standard until Gen 5. This is exactly the reason why its own Suspect EXP is more important than any other Suspect's, and will thus count for more when Aeolus and I finalize the Stage 3 SEXP equation. We will have to have these numbers before deciding which of the six (five) Suspects deserve to be retested in Gen 5, though. And as far as the new Gen 5 pokemon are concerned, we're smart enough to ban the 680 BST Legendaries, and also smart enough to see the potential of something like a pure fighting Darkrai Gen 5 "movie Legendary". I also hope we're smart enough to NOT ban something like a Gen 5 Manaphy until we actually play with it, this was a mistake made by a group of people at whom pointing fingers does us no good now. But when we have our Stage 3 SEXP for the five Suspects, we'll have a great idea of which of them deserve to be retested.

Finally, I'm still not convinced we necessarily need more people voting, or that even if we do, we would need to do anything but adjust the SEXP thresholds Aeolus and I have decided are good enough for the current Suspect test process. People may think it's a pain to have to play it out, and would rather just vote, or have the voting done amongst badgeholders or whatever, but when you realize that there stand to be less Suspects in Gen 5 than there were this time, you realize that it may not be in our best interests to completely abandon actual experience with Suspects or would-be Suspects. And especially when you realize that "we" are not going to have a "Gen 5 Manaphy" banned from the start, or do stupid stuff like "ban Kureseria!" or act on "is Dosaidon uber???" before having any experience with such pokemon.

 

[cim]

I don't think we should be making any assumptions about Gen 5 based on the previous metagame. So far, each generation of Pokémon has been drastically different, so using Gen 4 as a starting point for Gen 5 will just raise the exact same issues that led us to begin the Suspect Test in the first place: premature bans on Pokémon we were never sure were broken in the new generation.

Then again, we fundamentally disagree here, as I don't think there's such a thing as an obvious uber and would support starting the Uber list over with a new generation, as the previous generations have been almost completely different games compared to the gens before them.

 

[Jumpman16]

We may have never been sure before this generation on the Suspects (kind of why they were branded Suspects by definition), but after the Suspect Test process is complete, we will be more sure about how the current Suspects will play in Gen 5 than we ever were about any Suspect besides Latias and her brief stint in Advance. The assumptions we have previously made about Suspects were never, ever made with any experience behind them, and that is an incredibly important distinction to make.

Again, only Latias is a pokemon that is considerably less uber than it was in the previous generation, and we have actually "suspected" that Latias is a Pokémon, move or clause that respectively may benefit competitive standard or uber battle if moved or implemented elsewhere for quite some time, and only got to it so "late" because that is what was dictated by our Order of Operations. We were able to deduce after just a few months of play in DP that Latias is a Suspect, and more importantly willing to brand it as such. If Latias' movepool doesn't change or if there's no crazy competitive change like an attack/special attack split, Choice Scarf or Stealth Rock, it probably shoudn't be a Suspect in Gen 5. There may be such competitive changes, but nothing that should indicate she should be a Suspect again right off the bat.

We will still have to responsibly arrive at our list of Suspects no matter what we do. I don't think you're going to get many people on board with not assuming anything is obviously an uber. To say that is to imply that we should test pokemon like Rayquaza and Kyogre and, yes, Arceus and Wobbuffet. You're going to have to clarify such a blanket statement with regard to uber assumptions, because some of the assumptions we have made have saved more time than we've used up with this inaugural Suspect Test process (and had to because it was inaugural, I'll stress again).

 

[Lemmiwinks MkII]

I think what Chris is trying to say is that a new generation potentially brings such drastic changes to the metagame that we should not allow previous generation status to be a definite indication of a Suspect's initial status in the new generation, but rather decide things on a case-by-case basis.

For example there may well be the introduction of a levitating, special defensive Fire / Electric type with Fire priority that completely walls and counters Skymin like no other Pokemon could in Gen 4. Maybe there will be such a thing as an Encore immunity or more accessible trap immunity (through ability or otherwise), or even a possible change to the type chart that makes Dragon no longer such a broken type. Please don't bash me on the possible ridiculousness of my examples, they are just to demonstrate the point that there will probably be a number of factors to consider based on how the game has changed between generations when it comes to deciding our initial Uber list in Gen 5. To make concrete assumptions based purely on Gen 4 experience would be a big mistake IMO.

 

[Jumpman16]

And what I'm saying is that, since there was no Ground/Grass pokemon with Recover and 130 Base Defense released in Gen 4, there was no reason to believe that we should test Groudon. We're not going to base anything purely on Gen 4 experience—we're going to use our intuition to make "99.9% correct assumptions" like the 680 BST legendaries should not be included in standard play. Not every assumption we've made about the tiering of competitive pokemon so far has been incorrect.

If you or anyone else would like to argue that we should start Gen 5 with Mewtwo, Rayquaza, and Kyogre unbanned and available on the Standard Ladder that's fine (good luck), but if that happens, you can sure as hell forget any "badgeholder vote" scenario where we'd be using theorymon to decide how much more or less Mewtwo is in Gen 5. Again, you might have an argument about pokemon like Manaphy and Latios regardless of what's decided on them after Stage 3. But you both seem to be making either a bold statement in saying we shouldn't at the least keep the title legendaries out of Gen 5 Standard play, or an incomplete one that doesn't also start to lay the framework of how and where we should go about testing 493 + "100" pokemon at once in Gen 5 Standard play.

 

[cim]

If you or anyone else would like to argue that we should start Gen 5 with Mewtwo, Rayquaza, and Kyogre unbanned and available on the Standard Ladder that's fine (good luck), but if that happens, you can sure as hell forget any "badgeholder vote" scenario where we'd be using theorymon to decide how much more or less Mewtwo is in Gen 5.

I don't really see that as a problem. If something is as obviously broken as Gen 4 Mewtwo is, it would be banned in like 2 weeks from the Standard metagame, with the benefit of knowing that with very little doubt it was a broken Pokémon. If it takes longer than that, it wasn't obviously broken and deserved to be a Suspect anyway. Sure, it's a lot more work, but I'm sure people are willing to put in the effort.

Rather than go with the preconcieved notion that there is something inherently broken about (most) 680 base stat Pokémon, no matter what other Pokémon are around it, this solution ensures there is no "legend / previous uber bias" on a Pokémon and makes sure that it's clearly out of line.

Theorymon is, in my opinion, unreliable enough that one can't use it to make lasting tiering decisions at any level, in any generation. Theorymon predicted a Skymin-dominated Standard metagame in Platinum, and that wasn't the case a month after introduction. Theorymon predicted a BL Heatran and a Rhyperior dominated OU at the beginning of DP (as well as an Uber Latias, before we revisited that). None of us are so smart as to be able to predict and calculate exactly how a Pokémon will impact a metagame without even playing it, and the inconsistent accuracy of theorymon has proven this time and time again.

Again, you might have an argument about pokemon like Manaphy and Latios regardless of what's decided on them after Stage 3. But you both seem to be making either a bold statement in saying we shouldn't at the least keep the title legendaries out of Gen 5 Standard play, or an incomplete one that doesn't also start to lay the framework of how and where we should go about testing 493 + "100" pokemon at once in Gen 5 Standard play.

We don't even know if there will still be the same type chart, movepools, PP, base power of attacks, and game mechanics in Gen 5. They could change Paralysis to affect Ground types, make the Type Resist berries only cut down damage 25%, introduce a third type of damage outside of physical and special, have a party of 9 Pokémon...

I'm not opposed to keeping the title legendaries out of Standard, I would just like a few weeks to confirm, without theorymon and with little doubt, that the lowest-ban balanced metagame that we call Standard is truly upset by these Pokémon.

 

[Tangerine]

I don't really see that as a problem. If something is as obviously broken as Gen 4 Mewtwo is, it would be banned in like 2 weeks from the Standard metagame, with the benefit of knowing that with very little doubt it was a broken Pokémon. If it takes longer than that, it wasn't obviously broken and deserved to be a Suspect anyway. Sure, it's a lot more work, but I'm sure people are willing to put in the effort.

Remember how I told you like 50000 times that what is broken is relative to the metagame? Gen 4 Mewtwo wont be broken in a metagame filled with... other Gen 4 Mewtwos.

Secondly, "Wobbuffet is obviously broken, yet it took months before people utilized it properly". What makes you think the same won't happen here, considering there will be same biases?

Rather than go with the preconcieved notion that there is something inherently broken about (most) 680 base stat Pokémon, no matter what other Pokémon are around it, this solution ensures there is no "legend / previous uber bias" on a Pokémon and makes sure that it's clearly out of line.

And you bias the metagame to be centered around these ubers.

Theorymon is, in my opinion, unreliable enough that one can't use it to make lasting tiering decisions at any level, in any generation.

If you really don't believe we can't theorymon Kyogre being Broken in OU then you have some serious issues.

Theorymon predicted a Skymin-dominated Standard metagame in Platinum, and that wasn't the case a month after introduction.

I'm sort of amused that you guys seriously believe a month is a long time, especially in a game like Pokemon where you don't need to be heavily competitive to win.

Theorymon predicted a BL Heatran and a Rhyperior dominated OU at the beginning of DP (as well as an Uber Latias, before we revisited that).

Theorymon also predicted Garchomp would dominate the metagame along with Gyarados and Salamence, and most of the standard sets. Rhyperior dominating OU was based on wrong information on Solid Rock. Latias is Uber is based on Soul Dew, there was literally no talk of Dewless Latias until the Eon Tournament.

None of us are so smart as to be able to predict and calculate exactly how a Pokémon will impact a metagame without even playing it, and the inconsistent accuracy of theorymon has proven this time and time again.

Inconsistent accuracy? Really? I'm pretty sure theorymon is more correct than wrong! Secondly, obviously theorymon isn't going to be able to predict the changes that happen after the initial environment. Maybe you're just terrible at predicting. And if you really aren't smart enough to predict obvious ubers then you really shouldn't be posting here

I'm not opposed to keeping the title legendaries out of Standard, I would just like a few weeks to confirm, without theorymon and with little doubt, that the lowest-ban balanced metagame that we call Standard is truly upset by these Pokémon.

Standard isn't the lowest-ban balanced metagame, I'm not too sure why you keep calling it that. Standard is the most popular metagame people will agree to play in. Please stop applying random theories about what is "competitive" - competitive doesn't mean simply lowest bans, that's a ridiculously shallow way of thinking of what is competitive. I'm surprised you can keep talking and continue to attempt to make those statements when you've been shown to be wrong... over... and over.... and over... and over... again. Your entire argument comes from this simpleton idea of competitiveness and that's laughable at best, but then again, when haven't you made an argument that could actually be taken seriously? Maybe it's good that everyone knows better and ignores you

I guarantee that if Smogon ever follows the our current strict restrictions on usages combined with the theory of the lowest-ban balanced metagame as standard, Smogon will no longer be the dominant authority of Pokemon within two months.

 

[lilyhollow]

Agreeing with Tangerine. I can't imagine a balanced "Ubers Lite" metagame being difficult to construct with another Darkrai, Manaphy, Arceus, and Garchomp added into the mix "at least," especially when I could possibly see a 4th gen Ubers tier with say, Kyogre, Groudon, and Arceus banned becoming acceptably balanced; I also think that such a tier of play would be desirable, not only for "philosophical purity" but because a "balanced Ubers tier" would arguably become the second or third most popular mode of play pretty easily. With that said, I find myself agreeing more and more with Tangerine's frequent complaints regarding our tiering system and specifically his "Unban Everything" Mentality thread, which partially illustrates the issue we're faced with when considering the "untapped potential" of the "playable Ubers" due to the nature of our tiering system, which makes all balanced metagames dependent on usage in the "first" balanced metagame. In other words, under the current system, if we want a balanced tier that allows Pokemon like Palkia, Lugia, Darkrai and Mewtwo (which almost certainly exists), we also have to accept that the tier with Starmie, Snorlax, Jolteon and Salamence is essentially playing second fiddle-- we're "officially" pointing to what everyone views as an "Ubers-lite" tier and calling it "Standard." That doesn't make any sense, but neither does abandoning two dozen balanced, popular and iconic Pokemon and letting them waste away in a faux tier... so despite being ridiculously skeptical initially, I'm going to agree 100% that Smogon's tier system should be completely revamped for 5th gen.

I guarantee that if Smogon ever follows the our current strict restrictions on usages combined with the theory of the lowest-ban balanced metagame as standard, Smogon will no longer be the dominant authority of Pokemon within two months.

I'll agree to the extent that having an "Ubers Lite" metagame as Standard would not exactly do wonders for us "politically" or whatever, but if you think that this would actually be enough to "kill" Smogon, I honestly don't follow. The way I see it, if we really forced people to deal with something that unpopular, they'd simply begin to start voting to ban things; sounds like an argument for "uselessness" to me, one that I don't disagree with, but we'll "no longer be the dominant authority of Pokemon"? Come on lol.

That said, I definitely think that people will see "Jolteon, Starmie, Snorlax" as the baseline of Standard almost no matter what we do, and that that "taints" any vote that potentially has an effect on the perceived Standard metagame, whether we start with or without the information gained from 4th gen in mind. I think that should be really important to anyone who supports scrapping the current Suspect Test process for 5th gen and starting with zero assumptions, and serves as a pretty good reason for them to revamp the tier system to make each mode of play (mostly) independent of the usage of the other ones. In particular, I would support having one tier which uses our 4th gen knowledge/intuition as a "baseline" as Hip/Jump have described, and then another completely independent tier that serves as a "balanced Ubers." Honestly, the only thing stopping our current Ubers tier from being "Philosophically Pure Pokemon" is the fact that it's by definition imbalanced-- I don't think it would take very much to make an "Ubers Lite" tier that is both balanced, and popular enough to justify its existence.

But whatever, I'd love to see you elaborate more on what you'd like to see the tier system to look like come Gen 5, this is just my vague interpretation but I think it has plenty of philosophical and, more importantly, practical merit. I also seem to remember Colin having some "naming issues" with the tiers of some kind, I kind of wish you guys would speak more about these things in PR because you seem to have your ideas pretty fleshed-out outside the forums but whatever.

 

[Jumpman16]

I don't really see that as a problem. If something is as obviously broken as Gen 4 Mewtwo is, it would be banned in like 2 weeks from the Standard metagame, with the benefit of knowing that with very little doubt it was a broken Pokémon. If it takes longer than that, it wasn't obviously broken and deserved to be a Suspect anyway. Sure, it's a lot more work, but I'm sure people are willing to put in the effort.

Rather than go with the preconcieved notion that there is something inherently broken about (most) 680 base stat Pokémon, no matter what other Pokémon are around it, this solution ensures there is no "legend / previous uber bias" on a Pokémon and makes sure that it's clearly out of line.

First, I am definitely not sure that people are willing to put in the effort you're talking about. What makes you so sure they are? I'm basing my skepticism on literally over a year of watching the community's lackadaisical approach towards the things they actually have to power to fix in competitive pokemon, from the refusal to use Wobbuffet February-May 2008, to Deoxys-S's very shallow voting pool in October, to people constantly complaining about "getting ToF'ed" when they though they were good enough to make requirements in the last 2-3 days of a 30-day test. What evidence do you have to support your optimism?

Second, Tangerine already addressed the larger issue with starting out the metagame with pokemon like Kyogre and Darkrai and Rayquaza and Groudon. Do you really think it's going to take "two weeks" to determine that they all really are uber? What would you be confirming, and how? Stage 3 is going to be at least seven weeks, and this is with five pokemon that are actually Suspects. How can you even begin to think that when you retest these Suspects, and add Mew, Mewtwo, Ho-oh, Lugia, Groudon, Kyogre, Deoxys-D, Deoxys-A, Deoxys (spare me a "no one will use normal Deoxys argument" since this will kind of be a problem if every other "uber" is banned but then normal Deoxys isn't actually banned because of, you know, experience), Rayquaza, Palkia, Dialga, Darkrai, Giratina, Arceus and the 3-5 Gen 5 title legendaries, that this will take 14 days? That's less than one day a pokemon, are you serious?

Theorymon is, in my opinion, unreliable enough that one can't use it to make lasting tiering decisions at any level, in any generation. Theorymon predicted a Skymin-dominated Standard metagame in Platinum, and that wasn't the case a month after introduction. Theorymon predicted a BL Heatran and a Rhyperior dominated OU at the beginning of DP (as well as an Uber Latias, before we revisited that). None of us are so smart as to be able to predict and calculate exactly how a Pokémon will impact a metagame without even playing it, and the inconsistent accuracy of theorymon has proven this time and time again.

As already pointed out, a lot of this was silly theorymon that nobody actually acted on because it wasn't based on any experience (I pointed out the "Dosaidon???" thing already). Further, none of these pokemon are title legendaries. But most importantly, how would 14 days of playing with over two dozen title legendaries and Gen 4 Suspects be even remotely better than using theorymon to save us the straight up year that this would take to sort out?

We don't even know if there will still be the same type chart, movepools, PP, base power of attacks, and game mechanics in Gen 5. They could change Paralysis to affect Ground types, make the Type Resist berries only cut down damage 25%, introduce a third type of damage outside of physical and special, have a party of 9 Pokémon...

I already acknowledged this. Don't you think it will be a hell of a lot more difficult to figure out Gen 5 with "over two dozen new variables" when you add the possibilities of the above? Isn't this exactly why we aren't testing Evasion and OHKOs and Species Clause simultaneously with pokemon Suspects?

I'm not opposed to keeping the title legendaries out of Standard, I would just like a few weeks to confirm, without theorymon and with little doubt, that the lowest-ban balanced metagame that we call Standard is truly upset by these Pokémon.

To underline Tangerine's point: yes, I seriously think that anyone who either cannot or is not willing to recognize that Kyogre is uber should honestly reconsider posting here. You should be both willing and able to tell us all why you think Kyogre and Rayquaza, right now, in Gen 4, aren't uber, because we will be using the same "theorymon" to keep them out of standard Gen 5 play and these pokemon have never seen a lick of standard play in any generation.

 

[cim]

First, I am definitely not sure that people are willing to put in the effort you're talking about. What makes you so sure they are? I'm basing my skepticism on literally over a year of watching the community's lackadaisical approach towards the things they actually have to power to fix in competitive pokemon, from the refusal to use Wobbuffet

What evidence do you have to support your optimism?

Mostly UU. With a metagame with 1/10th of the players, there is enthusiasm to test and vote on things with just one ladder; as everyone has to play, it becomes even more obvious that THIS WILL AFFECT ME RIGHT NOW and thus gets more people involved.

to people constantly complaining about "getting ToF'ed" when they though they were good enough to make requirements in the last 2-3 days of a 30-day test.

I don't mean to sidetrack, but I think this comment's stepping on a few toes. I've heard several testers, testing the entire month, talk about getting ToFed, and it's an insult to those people to pass off the remark as whiny battlers who play for a short period of time. Sure, some did, but I've heard people like reachzero (one of the most active suspect testers) use it, and yeah.

Second, Tangerine already addressed the larger issue with starting out the metagame with pokemon like Kyogre and Darkrai and Rayquaza and Groudon. Do you really think it's going to take "two weeks" to determine that they all really are uber?

If it doesn't take two weeks for one Pokémon to dominate a metagame, they're not obviously Uber then. Staraptor in UU, a Pokémon less drastically broken than any Uber, took a good 5-7 days for everyone to figure out was too much.

What would you be confirming, and how?

Well, I'm not saying we're specifically looking for pokemon x to do this, then it's Uber. We're not actually "confirming" anything, we're banning things when they start breaking the metagame.

Stage 3 is going to be at least seven weeks, and this is with five pokemon that are actually Suspects. How can you even begin to think that when you retest these Suspects, and add Mew, Mewtwo, Ho-oh, Lugia, Groudon, Kyogre, Deoxys-D, Deoxys-A, Deoxys (spare me a "no one will use normal Deoxys argument" since this will kind of be a problem if every other "uber" is banned but then normal Deoxys isn't actually banned because of, you know, experience), Rayquaza, Palkia, Dialga, Darkrai, Giratina, Arceus and the 3-5 Gen 5 title legendaries, that this will take 14 days? That's less than one day a pokemon, are you serious?

You're misunderstanding me. I didn't really make my point clear enough, but I never intended just one round of voting or whatever. It would take 14 days to knock off the most obvious of the obvious Suspects, because we could go "Kyogre is tearing shit up, as well as Groudon" and get them knocked off. Then we ban them, rinse and repeat until we stop banning things.

So, no, it would be longer than 14 days. I'm envisioning a 2 week "wave" or two, then a transition into three month cycles that continues endlessly with each new list, or at least until it's apparent nothing is broken. This is open to discussion with this kind of system, though.

As already pointed out, a lot of this was silly theorymon that nobody actually acted on because it wasn't based on any experience (I pointed out the "Dosaidon???" thing already).

We made Skymin a Suspect almost purely on theorymon, when I personally think the hype and quick ideas that it was unstoppable should at least settle a little. If we waited even 2 weeks to make that declaration, I bet at least a few people would have been opposed to even making it a Suspect.

Further, none of these pokemon are title legendaries. But most importantly, how would 14 days of playing with over two dozen title legendaries and Gen 4 Suspects be even remotely better than using theorymon to save us the straight up year that this would take to sort out?

See above, the test doesn't end in 14 days.

I already acknowledged this. Don't you think it will be a hell of a lot more difficult to figure out Gen 5 with "over two dozen new variables" when you add the possibilities of the above? Isn't this exactly why we aren't testing Evasion and OHKOs and Species Clause simultaneously with pokemon Suspects?

I'm saying we should prepare for changes so drastic that every single Pokémon becomes a new variable.

To underline Tangerine's point: yes, I seriously think that anyone who either cannot or is not willing to recognize that Kyogre is uber should honestly reconsider posting here. You should be both willing and able to tell us all why you think Kyogre and Rayquaza, right now, in Gen 4, aren't uber, because we will be using the same "theorymon" to keep them out of standard Gen 5 play and these pokemon have never seen a lick of standard play in any generation.

I do not think that Kyogre and Rayquaza tests would end in a conclusion for anything but Uber, based on experience in the "no bans" metagame that is Ubers... meaning if we started the tier list with no bans, we would start by banning what currently unbalances Ubers (those two!).

If I had literally zero experience with them, I would not say.

 

[Tangerine]

You missed the point, lol.

If you think UU testing is the same as starting with the Uber tier you are terribly, terribly mistaken. Theorymon's project uber is perfect evidence that Uber tier doesn't work as you think it would. Then again you'll probably ignore me too :) Skymin would have been a suspect whether you would have liked it or not. Few people would have opposed in two weeks? Give me a break. Skymin wasn't that underwhelming.

Chris, i'll put it nicely. Your idea is stupid. We're not going to make your metagame standard. Dodging around the objections aren't going to work. You're missing why we're objecting to your tiering idea.

Of course, if we revamp tiers anyway we can always make a "balanced Ubers" tier if we felt like it, and you're free to call it standard. No one else will. Just you. So please, drop this nonsense. I'm not going to trust opinions coming from someone who have demonstrated over and over in Stark Mountain that they have little understanding of the competitive game, or even how people operate, considering you have this tendency to miss points and fail to understand people many times and that has been your main problem within this site. Without this understanding, you're never going to have a serious impact on this site.

 

[Jumpman16]

Mostly UU. With a metagame with 1/10th of the players, there is enthusiasm to test and vote on things with just one ladder; as everyone has to play, it becomes even more obvious that THIS WILL AFFECT ME RIGHT NOW and thus gets more people involved.

If we decided to start with no bans on the Standard ladder, your enthusiasm argument goes out the window because there's by your assessment at least 10 times the number of players that will "have to" play the "Standard" ladder which would really actually be ubers. And as for the "this will affect me right now" part:

http://www.smogon.com/forums/showpost.php?p=1740685&postcount=93

"Say I play well enough on the suggested "Standard Ladder + Suspect" to make the upper bound. I use said Suspect on my team and owe a lot of my success to it. But what if I only made the upper bound because the majority of the people I faced did not also use the Suspect? Ok, sure, a good team created by a user who is cognizant of the presence of the Suspect does not necessarily have to contain said Suspect, but...is that why I won? Did I beat "Suspect-less but Suspect-cognizant teams" or did I beat "Suspect-less and Suspect-oblivious teams"? I would argue that if I most played and won against the latter, I had an undeniable, clear advantage, and that those battles should in no way reflect my experience with the Suspect, because, to the Suspect-oblivious player, the metagame very literally did not contain the Suspect. When you couple that with my argument that apprising virtually everyone of important information about the server is impossible, you start to see why I don't think putting the Suspect on the Standard Ladder is a good idea. If a player can owe even one victory to the fact that his or her opponent "didn't know ubers were allowed" (and that statement is a lot less ridiculous than you may be inclined to think), his or her rating will be tainted."

Please address this (and read the whole post if you need context). When you are talking about starting an entirely new generation on the Official Server for online competitive pokemon, you are talking about hundreds and hundreds of players new to our community who will be joining our server looking to play online. This number is probably going to be in the thousands when you're talking about the people who will find and join our Official Server over the first few months of Gen 5.

I don't mean to sidetrack, but I think this comment's stepping on a few toes. I've heard several testers, testing the entire month, talk about getting ToFed, and it's an insult to those people to pass off the remark as whiny battlers who play for a short period of time. Sure, some did, but I've heard people like reachzero (one of the most active suspect testers) use it, and yeah.

I'm not talking about those people though? I'm talking about the people who either admitted to just playing the first and/or last few days of the test, or exaggerated their efforts in my np: threads evidently without the knowledge that I have access to the entire log of Suspect Battles to crosscheck what people said in my threads or in PM of their Suspect activity.

If it doesn't take two weeks for one Pokémon to dominate a metagame, they're not obviously Uber then. Staraptor in UU, a Pokémon less drastically broken than any Uber, took a good 5-7 days for everyone to figure out was too much.

Your brevity is a little off-putting because you are either not thinking through to the next step (able) or you are deliberately not posting what would likely happen next (willing). So I will just ask you this: how uber do you think Mew is going to be in a "no banned pokemon" Gen 5? What about Deoxys-S? And Latios? What about Dialga? Again, there are going to be over 24 "Suspects" if we start gen 5 with no bans, and only the most obvious of these would actually break the "metagame" in under 2 weeks.

Well, I'm not saying we're specifically looking for pokemon x to do this, then it's Uber. We're not actually "confirming" anything, we're banning things when they start breaking the metagame.

There is no "metagame" when you start with no bans. This is why our current uber "metagame" is unbalanced (and will be even more ridiculous in a few months when Arceus is released).

You're misunderstanding me. I didn't really make my point clear enough, but I never intended just one round of voting or whatever. It would take 14 days to knock off the most obvious of the obvious Suspects, because we could go "Kyogre is tearing shit up, as well as Groudon" and get them knocked off. Then we ban them, rinse and repeat until we stop banning things.

So, no, it would be longer than 14 days. I'm envisioning a 2 week "wave" or two, then a transition into three month cycles that continues endlessly with each new list, or at least until it's apparent nothing is broken. This is open to discussion with this kind of system, though.

And would take every bit as long as starting with the 680 BST legendaries banned, I'd argue. Which is practical because as we learn how the actual generation changes have altered how competitive pokemon is played, we will be much more able to determine what new Gen 5 factors would make Latios OU or Skymin uber or whatever.

We made Skymin a Suspect almost purely on theorymon, when I personally think the hype and quick ideas that it was unstoppable should at least settle a little. If we waited even 2 weeks to make that declaration, I bet at least a few people would have been opposed to even making it a Suspect.

I think you're forgetting that I purposely had us test DXS before testing Skymin so we would have a month of actual experience with it in Standard before testing it. And since it got its fair shake after some experience, its very close vote is more valid than one that would have taken place if we'd acted on theorymon instantly (which we've never done in the Suspect Test Process).

I'm saying we should prepare for changes so drastic that every single Pokémon becomes a new variable.

I do not think that Kyogre and Rayquaza tests would end in a conclusion for anything but Uber, based on experience in the "no bans" metagame that is Ubers... meaning if we started the tier list with no bans, we would start by banning what currently unbalances Ubers (those two!).

So why would we waste our time with them in a no bans metagame if they're 100% uber?

If I had literally zero experience with them, I would not say.

Okay, so you are indeed able to base Gen 5 tiering on Gen 4 experience? Kind of like I have been proposing this entire thread? Do you see now why I asked you this question?

 

[obi]

I have to question why you're calling certain theoretical gen-5 Pokemon "uber" already. No Pokemon is inherently uber. You seem to think that, if Gamefreak introduces another 20 or so Pokemon that can compete with the current ubers, you can make a "balanced ubers" tier and have fun with that, but they're still uber because of some magical, intangible status. That is not how the tiers work. Ubers is defined as the set of Pokemon incapable of being in a balanced tier. In other words, if you can have a balanced uber tier, then it's not ubers...

If you think this isn't the case, you should update the tiers article on the site to include non-competitive terminology.

 

[lilyhollow]

In other words, if you can have a balanced uber tier, then it's not ubers...

Which is why it seems dangerous to start testing Standard without any bans. As wonderful as it is on paper, the likely results are either practical inconvenience or, yes, philosophical impurity.

Tangerine has talked about this for ages and I feel like I want a neon sign or something because you're still talking about the same shit that has pretty much been firmly established for months now. He has a gigantic thread about it, go read it or something. Or don't, but please recognize that there is a stigma associated with "Standard" that has nothing to do with Dialga, Mewtwo and Giratina. Someone needs to address this; I'm sure even Tangerine would be happy if someone stood up and said something relevant instead of mindlessly spouting philosophical nonsense that we are all perfectly aware of (I personally would probably gravitate towards an "Ubers Lite" tier, and have argued in support of one, even as Smogon's "Standard" metagame, since before DP's release), but considering that his thread has received a whopping four replies in three months, I'm not exactly surprised at his frustrations anymore. You are literally ignoring the point; whether that's because of insufficient attempts to communicate it (I kind of doubt it) or because mechanical testing procedure debate is simply more interesting somehow, I'm definitely getting tired of the fact that this isn't getting proper recognition. I'm not referring to chris is me or obi or anyone in particular when I say this; just anyone who can contribute to this discussion without repeating things with which we are all familiar enough to preach ourselves, I'd love to see you speak up so we can stop running around in circles for once.

To spell out exactly what I think needs to be addressed, not that it hasn't been done before: "a Standard metagame centered around a number of 680+ BST Legendaries would be significantly less popular than one centered around the Pokemon that we currently deem OU." I haven't seen any objections to this, nor can I think of any. Why is nobody talking about this, which, if true, indicates a direct conflict between the well-being of our community and the "standard competitive gaming philosophy"? Assuming that the balanced metagame which bans the least amount of Pokemon is essentially an "Ubers Lite," which is absurdly possible so long as Gamefreak doesn't outright reset the series, there are two possible results:

1) we're driving players away from our community, which I find much easier to call "fundamentally anti-competitive" than a metagame that starts from a different baseline, so long as we aren't also gaining players in the process, which is probably doubtful.

2) players simply begin voting with the same "erroneous" assumptions as to what the Standard metagame is supposed to look like. Would this result in a metagame identical to one that started with our 4th gen knowledge as the "baseline"? Probably not, but surely the philosophical purity of a metagame is placed in jeopardy when the voters find an entirely different-looking Standard more desirable in the first place.

Again, I'm not going to go so far as to say that this would be some sort of community suicide (though if #1 takes place at all that's certainly an issue to consider regardless). But for "purists," it seems to me like simply ditching our banlist isn't anywhere near sufficient in making Pokemon an "ideal" competitive game. You would have to take our democratic process into question, or establish a firm definition of "balance," or something to ensure that biases don't worm their way through and taint the construction of the "first balanced metagame." Either that or you're just saying "whatever, good enough" which is pretty anticlimactic considering that it's almost guaranteed to be a poor "politics" decision even in the long run, and a longer, more bothersome process.

I believe that an Ubers Lite metagame would have a lot of merit. It's the one metagame that would truly reflect standard competitive gaming philosophy, which is definitely nice. It would also be very popular relative to our other "secondary" tiers of play. But these facts are respectively false and irrelevant in the face of our current usage-based method of determining tier banlists, while very few people have actually called that method into question. This makes absolutely no sense to anyone with the actual well-being of the community in mind; if anything, those who desire a philosophically pure metagame should be jumping all over Tangerine's suggestion that usage-defined tiers are flawed and should be reworked by Gen 5. This allows you to achieve everything that haphazardly reseting the banlist would, better, and without the risk of shitting all over the rest of the community. Why aren't you talking about it?

 

[Caelum]

This post will probably be confusing since I'll pick up on bits and pieces as I go. Probably won't make any sense but whatever.

Deciding suspects. We havent really got an effective system for this, we just kinda wing it.. People seem to be accepting it though, so this seems kinda minor

I'm personally a fan of the way RB and myself have managed it in UU. We allowed players to make an argument in a thread based on the characteristics and eliminated the poor arguments. My only concern with doing that for OU would be the (likely) number of suspect nominations we would receive and it could be a big mess. That said, I think a similar version for people with Policy Review access would suffice to nominated suspects and could probably produce a better result than just petitioning the general public. I guess, how democratic do we want this to be?

Voters. There havent been enough, at any point. It has been a big dissapointment, and I would say that of anything, this is the real problem with the suspect test process. But I also know how difficult it is to make the commitment to qualify. Several times I tried and gave up just because of the time needed.

I agree that more voters would be nice; but we don't want to sacrifice quality or quantity. The requirements could probably be relaxed a tad, but I don't think we should drop them significantly in attempts to create a greater voter pool. I'd rather have 20-30 qualified voters than 20-30 qualified voters plus 40-50 unqualified voters.

Timeframe. It has taken us a year. To be honest I think that a year is probably a reasonable timeframe, to take an untested metagame and create something with a sense of finality about it.

The streamlining of the process has significantly improved since the beginning, as expected, so next generation I would expect the process to take a bit less time assuming a similar number of suspects (assuming we replicate the same method which it appears we will).

I have to question why you're calling certain theoretical gen-5 Pokemon "uber" already. No Pokemon is inherently uber. You seem to think that, if Gamefreak introduces another 20 or so Pokemon that can compete with the current ubers, you can make a "balanced ubers" tier and have fun with that, but they're still uber because of some magical, intangible status. That is not how the tiers work. Ubers is defined as the set of Pokemon incapable of being in a balanced tier. In other words, if you can have a balanced uber tier, then it's not ubers...

If you think this isn't the case, you should update the tiers article on the site to include non-competitive terminology.

I think you are misreading Jumpman, Obi. I think he agrees with you for the most part. The disagreement arises from that Jump thinks you can reasonable deduce which Pokemon would be uber without a formal test of every single one.

We'll use what knowledge we can from Generation 4 to talk about Generation 5 Pokemon. Obviously, if the mechanics entirely changes and our experiences in Generation 4 become worthless in our deduction of Gen 5 suspects, we won't use our gen 4 experiences. It's as simple as that really.


I also want to bring up an issue of userbase for why we would need, at a minimum, at least a prototype uber tier in gen 5. Are we really going to start with everything unbanned and dick around while we force new players to start playing with broken things like Kyogre. I think that's unfair to them. Possibly a suspect ladder with everything unbanned would be alright and then transition the results of that to the standard ladder over time, but we'd have to start with some base uber list.

The philosophical purity of "test everything" is nice, but it's not entirely necessary for the success of any test. As I explained above, a prototype list is probably necessary and practical. My policy has always been if I could see a slight possibility of it being usable in standard play, even if I believe it's uber, it should be tested. Could we test something like Mewtwo? Sure, but I can't say that it would accomplish much that I couldn't figure out before hand.

The "uber-lite" point or something. I'm not even sure what that means. If you mean other ubers checking other ubers I don't buy that. Kyogre can be dealt with by Soul Dew Latias or Special Defensive Dialga; however, I think it's clear to most players that Kyogre is still obviously broken. I don't think we could ever create a balanced metagame out of most of our ubers by introducing other ubers; at least at this time.

 

[cim]

I also want to bring up an issue of userbase for why we would need, at a minimum, at least a prototype uber tier in gen 5. Are we really going to start with everything unbanned and dick around while we force new players to start playing with broken things like Kyogre. I think that's unfair to them. Possibly a suspect ladder with everything unbanned would be alright and then transition the results of that to the standard ladder over time, but we'd have to start with some base uber list.

This is something I'm not sure on (and by "not sure" I really mean "not sure", not "I disagree entirely"). The 4th gen user base started out tiny and grew as time went on. It's still growing right now. I heard that something similar happened with the third generation and NetBattle activity. I don't think we'll start out with a huge swarm of 5th generation players who then get scared off, I think it will be a gradual transition, giving us a neat window to test things.

The "uber-lite" point or something. I'm not even sure what that means.

The idea is that if Nintendo introduces 20 or so BST 680 Pokémon, that would essentially be enough to create a balanced metagame based on Ubers. If that happens, most of the above posters still want the traditional OU bar to be "OU", rather than the first balanced metagame.

Jump, I'll respond to your post later, I'm quite busy at the moment.