Next Gen? (suspects).

What is going to be done about the suspects from next gen? This system takes horrendously long and I'm pretty sure everyone will end up wanting to "test" stuff like Chomp and Latios and Manaphy and all those guys over.

It's really going to suck if Smogon has to keep doing it this way again for future games.
 
With HGSS still not out in the US, I do not see Gen. 5 coming out for at least a year or two. I don't think we really have to worry about it yet seeing as we haven't even finished figuring out Gen. 4 suspects. I also don't really see why (going by the current suspect tests) a Pokemon such as Garchomp or Latios or Skymin would ever be re-tested in OU, even in Gen. 5. When a new Pokemon or two comes to UU, all the other previous BL Pokemon are not unbanned. (Although it makes sense that they should)
 
No time like the present for figuring out a more optimised process, though. Or at least after this; hindsight should aid us a great deal.
 
With HGSS still not out in the US, I do not see Gen. 5 coming out for at least a year or two. I don't think we really have to worry about it yet seeing as we haven't even finished figuring out Gen. 4 suspects. I also don't really see why (going by the current suspect tests) a Pokemon such as Garchomp or Latios or Skymin would ever be re-tested in OU, even in Gen. 5. When a new Pokemon or two comes to UU, all the other previous BL Pokemon are not unbanned. (Although it makes sense that they should)
There could be countless reasons to re-test them. For instance, what if Ice Shard becomes the new tutor move that every Pokemon and their mother can learn? Also, if a few viable OU Pokemon who are faster than Chomp and can OHKO him become introduced, I see no reason to not test him as a suspect again. Anyways, it's never too soon to discuss this sort of thing, seeing as we don't know how long it could take to possibly improve on our current system. It would be far better to discuss it now, than it would be to wait until the next generation comes out, IMO.

That being said, I have already thought about ways to expedite our testing processes, but I really can't think of a way to do so, that doesn't involve short-cutting key steps. Though, I think the initial individual tests can possibly be thrown out, judging by how things turned out in this current test.
 

eric the espeon

maybe I just misunderstood
is a Forum Moderator Alumnusis a Researcher Alumnusis a Top CAP Contributor Alumnusis a Tiering Contributor Alumnusis a Top Contributor Alumnus
If we want to reach the goal of a metagame with as few bans as possible without any Ubers then the fastest and most efficient way to go about testing may be to start with only that bans which are almost unquestionable, start with all potential suspects not banned. This way all suspects are tested at the same time, and all together so that the effects that each of them has on the metagame and in turn each other can be observed, rather than in "isolation" (which is not necessarily entirely relevant if another suspect is unbanned). From this we may find that our theorymon about some suspects was wrong, and they are indeed perfectly fair in the new generation, and we may find that some suspects are immediately and totally broken, these would be banned once they show themselves to be clearly broken (probably with a voting system similar to the current one).

The disadvantage of this style of testing could be that players may end up playing in a metagame that has several "broken" Pokemon during the initial testing period. However, the time saved (testing 10 Pokemon at once is much faster than testing 10 one at a time, then going back and stage threeing them), and the fact that it automatically and fairly takes account of the metagame forces applied by suspects on each other (much like OUs Stage Three test) makes it an appealing option.

Saying something like "I also don't really see why (going by the current suspect tests) a Pokemon such as Garchomp or Latios or Skymin would ever be re-tested in OU, even in Gen. 5." is absolutely ridiculous. It's possible that the new generation will not change the metagame much, but if you take a look at the huge shifts that each other generation caused it seems almost inconceivable that Gen 5 will not render any and all testing we do in Gen 4 obsolete.
When a new Pokemon or two comes to UU, all the other previous BL Pokemon are not unbanned. (Although it makes sense that they should)
The difference is a matter of scale, yes if it was practical then ideally each time the rules changed everything would be fairly retested but adding a couple of Pokemon is will only ever cause a relatively minor metagame shift (unless they are so powerful to be broken, which removes them and the problem). It's a world away from adding ~100 Pokemon, a ton of moves and redoing a significant portion of the game mechanics.
 

Hipmonlee

Have a nice day
is a Community Contributoris a Senior Staff Member Alumnusis a Smogon Discord Contributor Alumnusis a Tiering Contributor Alumnusis a Top Contributor Alumnusis a Battle Simulator Moderator Alumnusis a Four-Time Past WCoP Champion
This really should be discussed in policy review. There's no reason this shouldnt be public.

Have a nice day.
 
We shouldn't rule out a massive reduction in the size of the Uber tier. With the likelyhood of more 'statistical Ubers' (Pokemon with 670 or higher BSTs and no hindering ability) appearing, it's within the realms of possibility that we could see a well balanced and diverse game with most of the current Uber population.

So I think the first step should be testing the metagame with absolutely no Pokemon bans whatsoever. If it looks like a small number of bans would create a balanced metagame, proceed by testing individual suspects.

If, on the other hand, it looks like the statistical Ubers will remain incapable of being present in a balance metagame, then I would suggest the following:

Anything new with a BST at or above 650, and no hindering ability, is default Uber.
Anything new with a BST below 600 is default OU.
Anything new with a BST between 600 and 650 inclusive goes into the second round of suspect testing. (The first round being the test of the open metagame).

I know that BST is a very crude measure of a Pokemon's strength. But it rings true with the present situation - everything with a very high BST except Regigigas and Slaking (hindered by their abilities) is Uber, while everything with a BST below 600 except Wobb and Wynaut is OU or lower; the Pokemon at 600 are pretty evenly split.

One advantage of this idea is that a 'conservative' OU metagame - one that is fairly well balanced and diverse, but lacks several Pokemon that rightfully belong in it - could hopefully be maintained while the suspect testing is going on. I think this is preferable to letting a bunch of new stuff with a good on-paper chance of being broken straight into the OU ladder.

Another matter: I really think our testing process needs to be more dynamic and responsive. At the moment it's very thorough, and consequently very slow. But when new games come out, the metagame will be in flux, with new sets for Pokemon new and old turning up a lot. A young metagame demands a suspect testing process capable of effectively responding to new developments.

One possibility for a dynamic process.
1 Suspects are chosen. Initially anything with a BST 600 to 650 is automatically a suspect.
2 Other suspects should be added by having a discussion to marshall ideas, followed by a formal debate and vote. This step should be ongoing.
3 All suspects are tested together for a period of time. Suspects previously OU are removed at the start of the test. Temporary absence of rightful OUs is IMHO less damaging that prolonged presence in OU of rightful Ubers.
4 All suspects are voted on. Those deemed Uber by a defined majority are made Uber and removed from the suspect pool.
5 Return to step 3, adding any new nominations passed by the ongoing step 2.
 
I agree with Eric's notion of introducing a big clusterfuck of a metagame of suspects, wait for the metagame to die down(after a year or so), and then actually start the elimination process. If anything becomes obviously broken(such as how Deoxys-s did after a certain time), then we eliminate it mid-way.
 
I highly support a system where everything begins unbanned and figure out what is uber from there, as I am of the opinion that the current uber list is unnecessarily big. The tests could be run in a similar manner to the current uu tests.
 

Tangerine

Where the Lights Are
is a Top Team Rater Alumnusis a Community Leader Alumnusis a Smogon Discord Contributor Alumnusis a Tiering Contributor Alumnusis a Top Contributor Alumnusis a Smogon Media Contributor Alumnus
I'm not sure why we're worrying about a system like there is a "correct" metagame we should aim for (ie "uber banlist is too big") and what justification we have of "aiming for the least bans possible". I'm also not sure why we're discussing this now when we really dont know what Nintendo will pull off in Gen 5.

Just take Nintendo's beginning list, adjust it with theory, and let it run a bit, see how it plays out, analyze the quirks of the new metagame, and adjust metagame to fit a specific metagame that highlights a quirk we want. "Least bans" is not something we should inherently aim for, but a "good metagame". If you're going to worry about Gen 5, then you should spend that energy thinking of what a good metagame is for gen 4 even before worrying about gen 5. Why worry about something later when most people have no fucking clue about the generation now? Sounds pretty empty and pretentious to me.

Got time to kill? Go figure out what makes a pokemon metagame good instead of worrying about future generations that we dont have any information on.
 
When I read this thread I was kind of thinking something along the lines of what Tang pointed out. (Of course he said it first and probably said it better but I felt like voicing my thoughts)

We already have enough problems with the current gen, and we should focus on that first. Especially since it's kind of a waste of time to try and go over multiple scenarios of "what if GameFreak adds [pokemon with so-and-so qualities]?". And then there's of course the problem that most of us (myself included because I have the critical thinking capacity of an 8-year old) don't really know what makes a pokemon metagame good [apparently].
 
what Tangerine said. Elevator Music emphasized the "we've got enough on our plate as it is" part so I just want to clarify that I'm more interested in the "it's valuable and interesting to talk about what makes a pokemon metagame 'good' and to incorporate that into the way we structure things in the future" part.
 
Indeed. I feel we of Smogon ought to work out WHAT we want for our metagames. A mission statement as it were. Something very general, that when we then consider more specific aspects of tiering policy, it all refers back to helping us achieve our mission statement.
 

Jumpman16

np: Michael Jackson - "Mon in the Mirror" (DW mix)
is a Site Content Manager Alumnusis a Top Team Rater Alumnusis a Battle Simulator Admin Alumnusis a Smogon Discord Contributor Alumnusis a Researcher Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis an Administrator Alumnus
It would be ideal to begin Gen 5 without any bans save for the ones on the titular cartridge legendaries. Anyone who wants to argue that keeping Mewtwo and Rayquaza banned is a bad idea can do so either now or then, but "good luck". One of the few precedents we've set that has made sense in hindsight is to ban the 680 BST legendaries. Again, and as I've suggested in a previous thread here, anyone who wants to argue to the contrary has virtually all the evidence in the world to make the case right now, today, that pokes like Mewtwo and Rayquaza were not indeed fairly tiered in Gen 4 from the beginning. I see no reason to entertain the idea of beginning Gen 5 with absolutely no pokémon bans until someone can make that argument.

Anyway, Stage 3 isn't quite done, but I am of the opinion that Manaphy's impending Stage 3-3 vote will have a ton of implication on how right everyone thinks they were about their predictions on our final tiers. Tangerine, for example, is convinced it will be voted uber by a supermajority. If this is true, then its failure to receive any kind of majority for uber in either Stage 3-1 or 3-2 would be rather "puzzling", to say the least. (It's 13-10 uber vote in Stage 1 isn't exactly a good sample size.) People have argued that it's because "everyone was focused on Garchomp" (81.5% of teams) and therefore Manaphy was largely ignored as an offensive threat itself. If this is true, then there is no logical reason to believe that future Suspects won't similarly be "ignored" in favor of the more popular Flavors of the Month. This is an problem with the "let's test everything at once" model, since it suggests that the more suspects tested at once, the more likely something will be overlooked since it was being "tested" in a less-true metagame.

And if Manaphy is ultimately voted OU, then one could wonder how wrong we were to assume it was so broken to ban it in the beginning of DP, or ban it after a very small amount of testing. Then you could argue that yes, of course we need to start Gen 5 with every non-titular legendary unbanned (as we will unless someone has a good reason), but this also underlines how crucial it is to NOT start Gen 5 with pokemon that a large consensus of us feel are uber, since it will likely hamper the potential of a Gen 5 Manaphy (or the Gen 4 one for that matter) until enough "Suspects" have been rightly banned.

All that said, I'm aware that this process is almost two years old now and has gone on for a very long time. Many of the processes will be streamlined by Generation 5, though, and I feel that the Suspect Test in Gen 5 will be able to begin with Latios, Garchomp, Skymin, Manaphy, and the ~1-2 legit Gen 5 Suspects we're bound to have. This is assuming Latias will no longer be a Suspect in Gen 5 (what a stretch), but take note that this also gives the same treatment to such almost-Suspects as Salamence and Scizor. I think a Gen 5 Suspect Test could be wrapped up in about six months once we agree on what pokes deserve to be considered Suspects.
 
Jump said:
such almost-Suspects as Salamence and Scizor
When was Scizor almost-suspect? =S (Edit: This was never intented to be a 'gotcha,' I know you wouldn't do it by mistake. I wanted to know where it came from. '=S' is a confusion face, though I'll admit its not appropriate)

Anyhow I'm confused about your stance Jump; you go from saying 'unban everything, good luck convincing autouber' to saying we need a list of suspects for various reasons. Were you just laying out pros and cons of each?

Jumpman16 said:
This is an problem with the "let's test everything at once" model, since it suggests that the more suspects tested at once, the more likely something will be overlooked since it was being "tested" in a less-true metagame.
Another argument against this (and most of my opinions regarding this approach) is summed up here. Reason I'm linking rather than explaining is because it's already laid out well, and the thread has plenty of reasoning to overlook the 'unban everything' approach. The loophole is that uber is not a genuine tier, so comparing how OU may turn out to a banlist metagame isn't really valid, and in fact "uber-lite" may (big may) be the elusive "true metagame."

Jumpman16 said:
I think a Gen 5 Suspect Test could be wrapped up in about six months once we agree on what pokes deserve to be considered Suspects.
I'm not sure how much a preemptive list will accomplish until we have validated mechanics on the next game. Subtle things can change a lot (more things may resist dragon or more moves of a certain type may appear), and our general idea of borderlined pokemon should suffice until we do have verification.


EDIT: And it seems I read your post completely backwards. Whoops.
 

Jumpman16

np: Michael Jackson - "Mon in the Mirror" (DW mix)
is a Site Content Manager Alumnusis a Top Team Rater Alumnusis a Battle Simulator Admin Alumnusis a Smogon Discord Contributor Alumnusis a Researcher Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis an Administrator Alumnus
When was Scizor almost-suspect? =S
When it was mentioned in Maniac's latest thread before this one and in Stark posts reference by Blame Game, even though your would-be "gotcha!" is way besides the point.

Anyhow I'm confused about your stance Jump; you go from saying 'unban everything, good luck convincing autouber' to saying we need a list of suspects for various reasons. Were you just laying out pros and cons of each?
No. And I've never said "unban everything" or anything even close to that. All I've advocated is tests of the various clauses and pokemon that would otherwise not be considered. The reason anyone could say "good luck convincing autouber" with regard to pokemon like Rayquaza is because we have been using them for over five years and can very accurately predict how well they will fare in Gen 5 Standard. The reason the same argument can NOT be made for clauses is because we have no competitive experience without them, regardless of the fact that what I want is to test them, not do away with them.

For the record, if Species Clause ends up proving to have been a good idea, the same logic applies. We would be going into Gen 5 knowing that the only time we played competitive pokemon without it, it broke competitive pokemon, therefore the onus would be on the "lets test species clause again" people if they exist in Gen 5.

Another argument against this (and most of my opinions regarding this approach) is summed up here. Reason I'm linking rather than explaining is because it's already laid out well, and the thread has plenty of reasoning to overlook the 'unban everything' approach. The loophole is that uber is not a genuine tier, so comparing how OU may turn out to a banlist metagame isn't really valid, and in fact "uber-lite" may (big may) be the elusive "true metagame."
The thread you linked to should remind you of my stance on how I feel we should approach Gen 5. As I said: "Even if it were easy to define a set cutoff point like banning 600+ BST pokemon we would still need "subjective" characteristics to make this cutoff point work. I'm sure you realize that this would ban Slaking and Regigigas yet allow Wobbuffet, and were just using it as an example, but it nonetheless underlines the fact that keen thought still needs to be put into whatever cutoff we initially arrive at. And if we're going to have to do this, why not just use our intuition when Gen 5 comes around". I'll address your "may be" below.

I'm not sure how much a preemptive list will accomplish until we have validated mechanics on the next game. Subtle things can change a lot (more things may resist dragon or more moves of a certain type may appear), and our general idea of borderlined pokemon should suffice until we do have verification.
Again, if someone can make the case right now, today, that pokes like Mewtwo and Rayquaza were not indeed fairly tiered in Gen 4 from the beginning, then I can consider not starting Gen 5 with the Titular Legendaries banned. Otherwise it would be a big waste of time to "start with everything unbanned", since through experience we have every reason to believe that said Titular Legendaries are going to be too strong for Standard play. Adding a type is an incredibly drastic change that would give obvious merit to a "lets unban everything" mentality.
 
The thread you linked to should remind you of my stance on how I feel we should approach Gen 5. As I said: "Even if it were easy to define a set cutoff point like banning 600+ BST pokemon we would still need "subjective" characteristics to make this cutoff point work.
Which is why I advocated the use of BST merely as a starting point, an aid to choosing the initial suspects.

Again, if someone can make the case right now, today, that pokes like Mewtwo and Rayquaza were not indeed fairly tiered in Gen 4 from the beginning, then I can consider not starting Gen 5 with the Titular Legendaries banned.
This is not relevant, because
Otherwise it would be a big waste of time to "start with everything unbanned", since through experience we have every reason to believe that said Titular Legendaries are going to be too strong for Standard play.
the reason the titular legendaries are not the standard game is because there are very few of them. For the generations thus far, we've had 1, 3, 6, and 10. Keep adding Pokemon with strength on a par to these legendaries, and eventually there'll be enough that becomes a well-balanced and diverse metagame. Now where that point comes I don't think we can easily say.
 

Jumpman16

np: Michael Jackson - "Mon in the Mirror" (DW mix)
is a Site Content Manager Alumnusis a Top Team Rater Alumnusis a Battle Simulator Admin Alumnusis a Smogon Discord Contributor Alumnusis a Researcher Alumnusis a Top Tiering Contributor Alumnusis a Top Contributor Alumnusis an Administrator Alumnus
What I can easily say is that we will reach the point where we will need more tiers before we have a standard metagame where there are enough titular legendaries that standard becomes a well-balanced and diverse metagame. If the game follows the trend for 3-5 such legendaries per generation, and we want to keep OU at 45-50 pokemon, and we assume that, say, 35 titular legendaries will eventually make a well-balanced and diverse metagame, we wouldn't reach this point until the 11th or 12th Generation is upon us.

By then, we would have over 1,400 pokemon. Do you think we would still only have five competitive tiers at this point? Or that, if not, we would let UU or NU swell to 600, 700, 800 pokemon?
 
Which is why I advocated the use of BST merely as a starting point, an aid to choosing the initial suspects.
BST should not be a starting point at all as to choosing initial suspects for the most obvious reasons. If anything we could choose to go by X-Act's BSR if it's entire purpose is just picking out suspects as it would be more accurate.

However choosing suspects was the easiest part and is something that shouldn't take anymore than a 1-3 months at most. To be honest I don't even see why we would need to use either BST or BSR to pick out a suspect anyway.

The duration of the tests as well as the sheer amount of tests smogon did on each suspect was what I was mostly picking on. Also with Gen 5 basically being announced now we should aiming to either fix or revamp the system of determining what will inevitably end up being OU, Uber, BL or whatever sooner rather than later.
 
Also with Gen 5 basically being announced now we should aiming to either fix or revamp the system of determining what will inevitably end up being OU, Uber, BL or whatever sooner rather than later.
Just posting to say that I wholeheartedly agree with this particular point. Doing so will help to streamline the Suspect process in Generation 5 if we can hopefully establish something by that time, which will allow more time for a stable metagame to develop.
 
So...is there any general consensus as to what we might be doing to the system? Throwing everything together at the beginning of the metagame is all fine and dandy but how would we determine suspects? How long would we let the metagame sit before we decide to take action? What method are we going to use to determine whether a Pokemon is actually Uber or not?
 
So...is there any general consensus as to what we might be doing to the system? Throwing everything together at the beginning of the metagame is all fine and dandy but how would we determine suspects? How long would we let the metagame sit before we decide to take action? What method are we going to use to determine whether a Pokemon is actually Uber or not?
I always thought that aside from banning name legendaries, we would just let things run their course unless we saw things becoming obviously overpowered, like we did for Deoxys-S during 3-1.

By then, we would have over 1,400 pokemon. Do you think we would still only have five competitive tiers at this point? Or that, if not, we would let UU or NU swell to 600, 700, 800 pokemon?
Well, by then I would be in my 40s, assuming 3 years per generation, and our community would either have gradually died off or ballooned to a massive size, in which case a splintering to further expedite testing processes might be in order.
 
What I can easily say is that we will reach the point where we will need more tiers before we have a standard metagame where there are enough titular legendaries that standard becomes a well-balanced and diverse metagame. If the game follows the trend for 3-5 such legendaries per generation, and we want to keep OU at 45-50 pokemon, and we assume that, say, 35 titular legendaries will eventually make a well-balanced and diverse metagame, we wouldn't reach this point until the 11th or 12th Generation is upon us.

By then, we would have over 1,400 pokemon. Do you think we would still only have five competitive tiers at this point? Or that, if not, we would let UU or NU swell to 600, 700, 800 pokemon?
I can't tell if you are exaggerating to make a point or not, but this is definitely a bit severe. The current OU of ubers contains about 30 pokemon, only ten of which are 680 BST legends. It is not unfeasibly to think that if gamefreak continues to increase the number of 680 BST legends per generation (RBY had one, GSC two, RSE three, and DPPt has five if you count the giratina forms, who play very differently, separately and arceus) that we could have a 45-50 pokemon tier that includes 680 BST legends, within the next generation, or at least the next two.

In light of this and my experience with the very fun tier of DPPt ubers I still think that the best course would be to at least try the next gen meta-game with all pokemon unbanned and if it becomes clear that it is too centralized or unstable go from there.

Also, auto-banning 680 BST legends and nothing else has a very noncompetitive feel to it. I can say with good confidence, both from my experience in ubers and through theorymon, that darkrai would be much harder to handle than some of the titular legendaries, like lugia.
 
I see nothing wrong with an autobanning of >680 legends, so long as we are open to the possibility of banning other things, such as Darkrai. I mean, such Pokemon (>680) would certainly be broken in OU anyways, so sending them away is no loss.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top