Metagame SV OU Metagame Discussion v4 [Volcarona Banned]

quziel

I am the Scientist now
is a Site Content Manageris a Forum Moderatoris a Community Contributoris a Smogon Discord Contributoris a Top CAP Contributoris a Contributor to Smogonis a member of the Battle Simulator Staff
I mean I'd argue clicking CC vs a Gliscor when its not revealed what tera type it is (low ladder may be running steel, may be saving it for their Volc, etc), isn't necessarily playing it safe. Frankly committing to a sweep early is probably a bigger risk that you necessarily need to be taking when in the pits.
 

viivian

beep boop
is a Tiering Contributor
Actually just don't bother playing it safe. I've learned hard from experience that players below 1400 will consistently make objectively the wrong choice. (Who the actual fuck doesn't click tera water on their gliscor when rain is up and the SD valiant has liquidation revealed and dies to a single stray hit are you insane?)
and why wouldn't you click liquidation in that scenario? what option did you go for that wasn't liquidation and why? i feel like making the obvious play of clicking liquidation would've benefitted you far more than going for an unnecessary read
 

hidin

What a kind young man
is a Site Content Manageris a Forum Moderatoris a Community Contributoris a Smogon Discord Contributoris a Tiering Contributoris a Contributor to Smogon
Not trying to offend anyone but if you’re questioning the reqs for a tier because you think it’s unrealistic for whatever reason you probably shouldn’t be voting in the first place. There is no reason to change the system to make things easier; your knowledge and skill in a meta is supposed to be judged with the requisites so you can actually make a valid vote, and if you can’t do that I really just think it is time for you to either get better or accept the circumstances.
 
and why wouldn't you click liquidation in that scenario? what option did you go for that wasn't liquidation and why? i feel like making the obvious play of clicking liquidation would've benefitted you far more than going for an unnecessary read
because I think too much despite knowing that these people will consistently make the wrong move. also hello corsola cove
 
Gonna add my own experience laddering... which is that I suck. I can't get over a 5 win streak since I am a bad player, so I have started going around with memey teams to stop other people. Honestly, for laid back people trying to get some fun, (myself included before I joined this site), suspects are unplayable-your favourite tier becomes swarmed with people with the same initials and you repeatedly get smacked. I remember a year ago complaining to my friend that these must have been top players, considering how they kept predicting me. He laughed me off.

Would it be against the fundamentals of tiering? Or could we use a seperate ladder, seperate thing which would be completely seperate from the main tier? Maybe the hassle would be too much, and if it would be sorry.
(also can someone advise me on the minior RMT in my signature?)
 
I don't think suspect tests should be easy.

Now, making sure suspect test accounts can't step on each others' toes sounds just like a silly-easy way to make it so reqs are more consistent for those who can do it, reducing the unnecessary grind and, as ausma said earlier, the stamina factor.

Let's imagine ouogtimmypeak#250 gets unlucky and matches up against ouogbigmanpeak#2 and ouoglargecylinderpeak#100. Timmy's suspect won't be entirely ruined, but it will be much shakier from then on and it will indeed increase the number of games he'll have to play. If he's REALLY unlucky, he'll have to play way more games and possibly double the hours to get reqs. On the long run, this won't stop him since he is a good player, of course, but removing this part of the equation will reduce the mindless grind.

People mentioned that this will open a venue for players to target lower-level players during suspect weeks. To that I say: unfortunately, assholes creating alts by the dozen with the sole purpose of keeping pulverizing new players on the 1000-1300s range are a constant, so I don't think this will make much of a difference.

All in all it's not good to make suspects easy (well, the entire point of the suspects is to create an organic barrier to gather competent voters). It is, however, on the interest of everyone to make it consistent to reduce the stamina factor of the suspect barrier.

edit for clarity
 
Last edited:
Not trying to offend anyone but if you’re questioning the reqs for a tier because you think it’s unrealistic for whatever reason you probably shouldn’t be voting in the first place. There is no reason to change the system to make things easier; your knowledge and skill in a meta is supposed to be judged with the requisites so you can actually make a valid vote, and if you can’t do that I really just think it is time for you to either get better or accept the circumstances.
It's not that it should be made easier, it's that the current system heavily incentivizes spiking. The whole reason it's even a consideration at the moment is because a single person playing through, instead of just resetting their account, has attracted significant attention for doing it "the hard way".

If it's about "putting in the work", what makes resetting your account a dozen times after an early loss more impressive than just grinding your way up to an Elo target, rather than a GXE one?

Grinding is grinding, and if you give people incentive to not start over, you'd see a whole lot fewer "low-ladder" matchups between two suspect accounts.
 
This conversation is way off track at this point but let's do away with the "git gud" non-sequiturs; you're entirely missing the point. Nobody is suggesting that suspect tests should be easy or complaining that it's not fair if they lose to a 1300s player. The only slight issue that's been brought up is when two suspect players get matched very low in the ladder. It can be very demotivating and detrimental to a run to lose that early on, and if you match up against another suspect player, one of you is going to lose. You could be the two best players on the entire ladder, it doesn't matter. If you lose to an actual 1300s player, skill issue, get better, etc. If you lose to a suspect alt of a player who would usually be playing up near the top of the ladder, that's... less of a skill issue.

Regardless I really don't care enough about this issue to argue any more about it, but anyone saying something along the lines of "get better" or "suspects are supposed to be hard lol" is missing the point and shadowboxing an argument nobody is actually making.
 
Not trying to offend anyone but if you’re questioning the reqs for a tier because you think it’s unrealistic for whatever reason you probably shouldn’t be voting in the first place. There is no reason to change the system to make things easier; your knowledge and skill in a meta is supposed to be judged with the requisites so you can actually make a valid vote, and if you can’t do that I really just think it is time for you to either get better or accept the circumstances.
absolutely trying to offend everyone but this is a pompous, elitist position to take. none of the proposed changes would open the floodgates to uninformed idiots getting reqs, which is the clear subtext here. uninformed idiots don't even know a suspect is happening. there are plenty of people with enough knowledge and skill that their voices should matter but who don't have the time, stamina, or luck to get reqs under the current system. just say you're scared of me being able to vote, cowards
 
Question.
Can we have two different suspect things going on? Like, one democratic vote and one laddering challenge. We already have 'qualified' and 'unqualified' in surveys, can't we have the same?
 
Random idea and I haven't thought it through, but what if you do "first runs" on a new Alt to get to a lower GXE?

This then allows you to access a different ladder (and maybe related chat room) where you can do subsequent runs and then your voting right depends on the GXE in this second ladder?
 
its not unknown tech to try to get to be slower than other pivoters in general. here its used with chilly reception, but this was known w regular slowking and slowbro in ss. and of course it goes down further to 98 etc. @ some point it stops doing its job as a defensive piece and loses the point. the reason why you want to be slower is that you get to choose what to go into based on what your opponent goes into
Guy like me intentionally tried to get paralyzed with corviknight so i would get slower u-turns
 
With the power of throat spray plus roar, kyurem can do something funny stuff

Kyurem @ Throat Spray
Ability: Pressure
Tera Type: Dragon
EVs: 252 SpA / 252 Spe
Timid Nature
IVs: 0 Atk

-Roar
- Ice Beam
- Draco Meteor
- Earth Power

it WAS a shame kyurem couldn’t utilise set up on special sets, now it can get a free boost and forces a pokemon, which may not able to handle kyurem or a pokemon that devours it

use this as much as possible before the kyurem ban, maybe you can get reqs can save kyurem? Or make sure he gets arrested for his crimes

the valentines day themed gas pass for £13.99 , unlock the generation 4 uu ladder after reaching level 42, elo gained in the event will be lost next season

and for just £23.99, you can use magearna in ou again, maybe magnezone won’t be a lonely ru mon anymore
Not to mention leaks show they will be adding the option to kick ban shroom god from games and gain some free elo next season
On one of my kyurem gimmick sets, I used noble roar kyurem with throat spray. I guess great minds think alike.
 
If we are talking about pure RNG, I think a good starting ground for accounting for it is deviation from expected probability. I made a post before about it before, and to summerize it’s about checking how often you Freeze/Para/Flinch/whatever to the expect number of times you’re supposed to be haxed.
If you use Ice Beam a total of 16 times in a match, you should be expected to freeze 1.6 times, so 1-2 times. If you use Scald 16 times, you should expect 4.8 Burns or 4-5 burns (assuming each time the target was eligable to be frozen or burned). If you use Scald 16 times and 6 of those times the target couldn’t be burned, it’s be 3 expected burns.
Something like ((constant)*(actual hax #))/(expected hax #) being a multiplier for GXE/Elo would totally work.
In 2009, there was a huge policy debate about changing the outcome of matches based on "more skillful play" to eliminate losses based on severe luck. It turned out to be a giant April Fool's prank. I recommend reading through the script; it's pretty fascinating, and hilarious to see the "roles" that both contrived and genuine forum posts tend to gravitate towards.

In your case, while the statistics themselves are objective, interpretation of them is not. If I Mist Ball exactly once and get the drop, is that hax? Or would it be hax to not get the drop? If I use Iron head twice and it flinches once, is that hax? That's a 50% flinch rate, but it's as close as possible to its innate 30% chance, so should it be considered the expected outcome or 20 percent higher (or 50/30 = ~67% higher?). Moreover, not all instances of hax are equally impactful, or can even be subjectively impactful—a crit on a move that already KO'd wouldn't matter at all, and a crit dealing 90 instead of 60 might not matter since it was a 2HKO anyway, but it could have impacted the decision not to switch due to being in Stealth Rock KO range etc. Trying to create a formula that accurately measures luck and applies an outcome to your ladder ranking is nightmarishly complex and subject to endless argument.
 

Srn

Water (Spirytus - 96%)
is an official Team Rateris a Forum Moderatoris a Community Contributoris a Top Tiering Contributor
Moderator
Now that we're talking about reqs, I'd like to offer some tips on getting them!
To be realistic here for a second, if you're not comfortably and consistently at least 1800s on ladder, I think it's going to be very difficult to reliably get reqs. A lot of luck or a fire team can get you there once, but there's no replacement for skill.
These tips aren't gonna improve your skill level and giftwrap you reqs, it's just gonna make things less of a grind and help you be more efficient.

1) Early losses kill your gxe
Gxe is the chance you'll beat a random player, so when you win or lose vs somebody with lower or higher gxe, there's a few ways this can play out:
a) You win against somebody with a higher gxe than you --> your gxe improves a lot
b) You win against somebody with a lower gxe than you --> your gxe improves a little
c) You lose against somebody with a higher gxe than you --> your gxe drops a little
d) You lose against somebody with a lower gxe than you --> your gxe drops a lot

This is why some people can qualify with 29-1 and others with 30-0, or some with 34-3 and others with 32-3. The same W-L record does not gaurantee the same results, it depends on what kind of players you fought.

Ideally we avoid d) as much as possible, because we want the highest gxe we can get, so I always recommend going at least 10-0 to start. Ideally you can manage something like 16-0 or better, but 10-0 is the bare minimum imo. From there, I think 22-2 into 35-5 are good benchmarks to try and hit. Once you're at 35-5, you're almost done, so just keep going without letting your gxe drop below 78%. After 40 or so games, you're gxe isn't gonna drastically change, so if you're stuck at 77.8% or something (and we generously assume that every win improves your gxe by 0.2%) then you would need 11 straight wins to make it back to 80%. This is unlikely if you ended up below 78% in the first place, so try not to get screwed here. These are just general benchmarks I use, not some hard and fast rule. Don't be afraid to abandon your current alt and grab another if you fail to reach these benchmarks by a longshot. I went 2-1, 3-1, and 3-1 for kyurem reqs before going 34-3. Random shit gets you sometimes, don't let it tilt you.

In summary: Go 10-0 --> 22-2 --> 35-5 --> Finish without <78% gxe.

2) Any teamstyle can work, at every stage of the ladder. Use what you know best.

Some may disagree with me on this one, and I used to spam HO until I hit 1500s or so before switching to something I'm more comfortable with. That can work too, but I don't think you need to pigeonhole yourself into anything specific. HO, BO, balance, and even stall can be used from 1000 to 1700, for the whole run. What matters the most is how well you know the ins and outs of your team, and if you were able to reach and stay 1800s with it, you should be good. Gen9 in particular really demands you learn how to use your not only your 6 mons but also your 6 teras, and this takes practice, so use what you know. I've gotten reqs for every suspect this gen by just loading 1 single team the whole time that I was familiar with.

3) Be familiar with samples, RMTs, TOTW, and other public teams. And don't use them.

More than any other generation, I feel that teams get MUCH worse the second that they go public. Tera types and other surprise techs provide immense value in this generation where every turn matters, and the gap in information makes a huge difference. Knowing which tera types to expect, what moves your opponents have, and what items to play around makes your gameplan much stronger. Don't give these advantages to your opponents. You should absolutely know mimikyu stardust's and pinkacross's HO RMTs inside out by now, and definitely just snag the TOTWs and successful RMTs you see. Put them in your teambuilder, not because you're gonna use them, but so that you have the advantage when you run into them.

This does kinda require you be proficient at teambuilding yourself, or just have good friends who are, but that's how it goes. You can also steal teams from tournament replays (especially SPL), but keep in mind these are not necessarily built to be consistent on ladder, and you will have to fill in some blanks yourself (especially tera types).

4) After 30 games, keep close track of where you are.

Type /rfaq reqs in the OU room or refer to every suspect test OP for the chart that tells you the gxe requirement that corresponds to total games played. Do not screw yourself by achieving reqs, loading another battle, and then losing. It's a really frustrating mistake to make, so just keep that chart open and make sure you do not miss the moment you get reqs.

The current system of suspect tests and getting voting requirements is not perfect, but I do think it's very accessible and fair, while being just difficult enough to serve its purpose. We can discuss other ways of getting them, like attaining specific elo benchmarks with a fresh alt or winning special suspect tours, but ultimately there should not be any "easy" way to get them.
 
Last edited:
Honestly, the majority of people that complain about certain mons can’t qualify for reqs and that’s a good thing. I don’t want to see unqualified players have their say and the game gets to the point where it’s unplayable. Keep your opinions here and let the qualified vote what’s best for the game. :)
apologies for the harshness, but there's no way to say this gently: i don't think you're in any position to be talking down on "unqualified players" when you've personally been on the losing side of the palafin, annihilape, espathra, chien-pao, volcarona, gliscor, ursaluna-bloodmoon, and most recently sleep debates
The current system of suspect tests and getting voting requirements is not perfect, but I do think it's very accessible and fair, while being just difficult enough to serve its purpose. We can discuss other ways of getting them, like attaining specific elo benchmarks with a fresh alt or winning special suspect tours, but ultimately there should not be any "easy" way to get them.
i'm not looking for an easy way to get them, i'm just looking for a way that isn't such a slog and doesn't depend on getting lucky enough to not match up into better players' suspect alts for 30+ games. give people an elo goal so they don't have to spin up a new alt every time they lose 3 games. give the good tour players suspect tours so they don't have to slog their way through ladder and other players don't have to be sniped by them, it's a win-win. both of these are good options and should be implemented
 
Last edited:

Ehmcee

A Spoopy Ghost
is a Tiering Contributor
If this request isn't fulfilled, it's totally fine, but thought to ask:

Would the required GXE percentage please be put in this thread title? So "SV OU Metagame Discussion v4 [Kyurem Suspect GXE%##]".
It'd just be a nice quick reference.
The required GXE percentage is fluid and depends on your amount of games played, /rfaq reqs command exist on Showdown, while the table listing all the possible GXE and Game amounts are displayed in both the suspect test discussion and ID thread.
 
Splatoon community went back to forums lately after Twitter drama. A friend talked to me and was like, "This means there'll be way more thoughtful posts!" And they seemed so dismayed when I thought of this thread, and thought... No. It's 90% Twitter, 10% good posts.

Hell, my post here is bad. This post sucks. It's basically a tweet.
 

Users Who Are Viewing This Thread (Users: 12, Guests: 35)

Top