Bo3 would make your reqs run take at least twice as long, potentially even up to 3 times.
There's no reason to make it Bo3, please explain yourself instead of just saying it "sucks ass"
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
-
Congrats to the winners of the 2023 Smog Awards!
Metagame SV OU Metagame Discussion v4 [Volcarona Banned]
- Thread starter TPP
- Start date
There's no reason to make it Bo3, please explain yourself instead of just saying it "sucks ass"
logistically this is a nightmare, but i'm having a hard time disagreeing because every reqs run i try to do either goes like 2-1 before getting fucked by rng or ends up getting over halfway and then dropping 2 or 3 games back to back because of some matchup-fishing bullshit. maybe i should just give up on being a serious player and go back to testing shitpost teams or hang out in the 1200s ruining other people's reqs runs because lord knows it's impossible to actually try and do serious shit in this goofy-ass meta. how did we get a mechanic that lets you switch your matchups mid-match and still every game in this godforgotten gen is decided at preview
Last edited:
god i hate zote
edit: I just realized I forgot to delete my earlier rage-message but it's funny so I'll keep it.
I can kinda relate; losing a suspect run just because you ran into a bad matchup is always frustrating, but this is entirely the wrong way to go about fixing that. I actually think a Bo3 would actually discourage skill, because up until a point (1500s+ I'd say) I think people are either gonna never bother to switch teams, leading to the suspecter, who has a significantly higher motive to actually win, being able to pick a better team for the matchup, giving them an easier time, thus lowering the skill ceiling. And it's not like it significantly reduces the luck aspect: the more skill-based Bo3 format is balanced out by the fact that each suspect run is at least 2x longer than normal, meaning more chances for RNG to fuck you up.
edit: I just realized I forgot to delete my earlier rage-message but it's funny so I'll keep it.
Last edited:
if i had the time and energy and knowledge i'd spin up some sort of algorithm that somehow factors the amount of rng that happened and how much it mattered into how much or little elo/gxe players gain or lose, so losing a match due to rng doesn't automatically end a reqs run. this would have the side effect of not being able to go anywhere on ladder with hax-fishing teams, as god intended. unfortunately, i barely even have the time or energy to even ladder myself, and i might be able to whip some sort of basic program up in like c# or something but i don't have the first idea of how to implement that on showdown
Last edited:
The only time I've ever gotten reqs was going 19-0 before losing. I don't have the skill to win 20 games in a row. If you're losing a match after winning only 2 games, there's clearly a huge skill issue present in your teambuilding and/or your plays. Try to improve rather than slamming the game for being decided at preview since that is not the actual reality.
Best of 3 is like, objectively better for competitive integrity. But ladder was never about competitive integrity. The people who are consistently getting reqs basically do not have to sweat it that hard, and that is why they will never accept this. It'd waste their time. And the people who don't go for reqs will never accept having to play Bo3 when they don't want to.
I sympathize with the want for Bo3 because it is great (still think it's great for tournaments), but it's just not what ladder is about.
I sympathize with the want for Bo3 because it is great (still think it's great for tournaments), but it's just not what ladder is about.
dropping an early game against an actual 1000s player is a huge skill issue, but it's also not what's actually going on here. the times i go 2-1 or some abysmal record like that are the times when i get matched up early against another suspect account belonging to someone who actually knows how to play the game at some level of competence, or some top player who's just starting a ladder challenge, or some other instance of an actually good player being that far down. by far the biggest source of my failed reqs runs is other people's reqs runs
I've noticed this as well. Under usual circumstances, if I have a half decent team, there's almost no shot I lose before I hit at least the 1400s unless I run into a good player on an alt. When a suspect drops, the odds of that happening increases dramatically. It also might be confirmation bias but I feel like there's at least a handful of people who sit on alts in the ~1300-1400s and try to snipe people going for suspect reqs.
That's just how it is though. I don't really see a viable solution to that problem that doesn't compromise the integrity of suspect tests, and I think the requirements should be as harsh as they are.
Reqs aren't harsh lol just load a different smorgasbord of ho's and tweak em a lil.
Works ever tim
Works ever tim
what about a simple tweak to the matchup algorithm: "if players have the same first 4 characters in their names, don’t match them up against each other on ladder." boom, now we can grind for reqs against normal players without having to step on each other's heads to reach them
the problem is that for at least a few days now you have a ladder where you can essentially opt in to not playing (usually) better players lmao
I would say this is a bit of a fundamental issue with using Elo and what have you for a game series like Pokemon.
I just really hate it when I play a game, get absurdly unlucky, lose despite making every right play, and also lose as much Elo as I would if I constantly make mistakes.
Imo randomness should alter the increase or decrease of Elo. I don't have an exact equation on how it would be done, but generally;
-If you freeze the opponent's Pokemon multiple times and you win, you should get less Elo/your opponent loses less Elo
-If you freeze the opponent's Pokemon multiple times and you lose, you should lose even more Elo/your opponent getting more Elo.
-The increase/decrease being determined by expected probability, in this case when freeze is used 16 times, it should be expected to have frozen Pokemon 1.6 times, or if it was only used once it'd be 0.1 times. You'd compare number of times you froze a Pokemon to number of times you expected to freeze the opponent.
I think something like that would ease the frustration of Pokemon so long as Smogon is cart accurate, and makes Elo more accurate. You don't get people getting insane Elo jumps after winning games they had no business winning, or lose massive amounts of Elo after losing through no fault of their own.
I just really hate it when I play a game, get absurdly unlucky, lose despite making every right play, and also lose as much Elo as I would if I constantly make mistakes.
Imo randomness should alter the increase or decrease of Elo. I don't have an exact equation on how it would be done, but generally;
-If you freeze the opponent's Pokemon multiple times and you win, you should get less Elo/your opponent loses less Elo
-If you freeze the opponent's Pokemon multiple times and you lose, you should lose even more Elo/your opponent getting more Elo.
-The increase/decrease being determined by expected probability, in this case when freeze is used 16 times, it should be expected to have frozen Pokemon 1.6 times, or if it was only used once it'd be 0.1 times. You'd compare number of times you froze a Pokemon to number of times you expected to freeze the opponent.
I think something like that would ease the frustration of Pokemon so long as Smogon is cart accurate, and makes Elo more accurate. You don't get people getting insane Elo jumps after winning games they had no business winning, or lose massive amounts of Elo after losing through no fault of their own.
I don't think suspect players should always be exempt from facing each other, though, only that it doesn't feel organic when it happens at a level way lower than either of you would usually be playing at. Maybe you could deprioritize matching players with the suspect characters against each other up until like, the 1400s, and then from there it's sink or swim, motherfucker.
true, but normally you never match up into vastly better players anyway because the matchup algorithm is designed to match you against someone around your own elo, explicitly to prevent the sort of massive skill-gap matchups that are prone to happen in suspect tests
fair enough, i think 1400 would be a good cutoff for that. by that point i've gotten enough wins that it no longer feels like bullshit to match up into another suspect account and lose. i just didn't want to make it too complicated because i felt like a simple solution would be more readily accepted. not that it matters, no one's ever gonna consider this as a serious proposal because my name is attached to it
trick room under misty terrain… #FraudDetected
Problem is that the prefixes in other tiers can be quite a bit more than 4 characters.
Yeah Natdex UU I’m looking right at you.
Speaking for myself and not the council here.
The current reqs system works well for what it is, funneling in consistent and competent players for voting reqs. It isn't an unreasonable system by any stretch of the word, especially with the new +/-0.2 GXE system that has been implemented last generation to reward a highly successful reqs session and give leverage to those who have a harder time, making them more accessible than ever. As Finch already said, this system is definitely solid and works fairly well, as shown by the incredibly high voterbase in recent months.
That being said, I do believe the current system by itself, while good, could be improved, or at least supplemented in some way. At the end of the day, it's true that your experience and skill in the metagame is tested, but I do subscribe to the idea that its just as much of (if not more of) a stamina test that some people are better suited to approach than others. This admittedly has made it a problem for me in the past prior to me joining the council, and while part of it definitely was a skill issue at the time, I (and I know a good deal of other players who are experienced and I think would definitely deserve the right to vote but struggle with stamina) can have a frustrating time when you factor in fatique and tilt that have the potential to punish you with a greatly damaged (and sometimes even ruined) run off of one or two mistakes and/or the luck of the draw. When the primary initiative behind the current system is to gauge your experience, that can be extremely upsetting, for sure. I think for that reason it's absolutely worthwhile to have a dialogue about; I was having a chat with viivian about it the other day, and I was waffling a bit on opening a policy review thread. Though I have ultimately decided against it since, like I said, the current system is definitely not broken and works quite well, and allocating resources toward addressing the system when there are other priorities is a bit rough to justify imo. Unless there is a massive wave of support I probably won't be doing so.
___
However, while I'm here, I did have two ideas on how to, specifically, supplement the reqs process. I don't believe removing a functional system is a particularly great idea regardless of its flaws, as there isn't anything fundamental that makes it broken or overtly and arbitrarily skewed toward one group of players.
1: My first suggestion is to implement suspect tours in some way. This is a pretty common practice in other/unofficial metagames and RoA suspects mostly because they don't have active ladders, but to me it's an excellent voting alternative that focuses exclusively on metagame and skill knowledge. The primary caveat of this is that the scale of OU suspects comparatively can complicate things for the OU room staff who already have a ton on their plates, and voting reqs cutoffs are things that have to be sorted out too. It's not a perfect alternative given the circumstances behind OU but I do think they excel when utilized properly.
2: My second suggestion is one that's a bit rough around the edges, but specifically is a bit inspired by the crazy run of LusterSN who had literally >100 games before qualifying and hit 1900 before reaching 80% GXE. Another potential alternative could be having an ELO cutoff alongside a GXE cutoff. This one is one that would need a lot of sorting out regarding what thresholds are truly "correct" so to speak, but there is potential in it. I think a lot of people often dismiss ELO as a way to evaluate someone's experience in the tier. There comes a point where you get past the garbage and truly have to contend with competent players and the metagame at its fullest force. Even if your GXE isn't particularly the best, being at high ELO means you are still effectively winning enough games to even be there in the first place, which requires competency and some degree of experience to achieve and maintain. I personally don't like this one as much as the suspect tour option because of the fact that it's far more fickle to establish boundaries for and could lead to ladder saturation, but whether these are dealbreakers or even problems at all I think is up for debate.
___
My own stance is that having voting alternatives alongside the current GXE-based system could be an actively positive thing to encompass a greater range of skilled/experienced players who have a harder time with the stamina-oriented demands of the current system. These are things I personally would support implementing if we have the resources/time, but it is not something that I believe is mandatory or at least not worth prioritizing compared to actually working on the tier. Regardless, I hope this can provide some food for thought.
The current reqs system works well for what it is, funneling in consistent and competent players for voting reqs. It isn't an unreasonable system by any stretch of the word, especially with the new +/-0.2 GXE system that has been implemented last generation to reward a highly successful reqs session and give leverage to those who have a harder time, making them more accessible than ever. As Finch already said, this system is definitely solid and works fairly well, as shown by the incredibly high voterbase in recent months.
That being said, I do believe the current system by itself, while good, could be improved, or at least supplemented in some way. At the end of the day, it's true that your experience and skill in the metagame is tested, but I do subscribe to the idea that its just as much of (if not more of) a stamina test that some people are better suited to approach than others. This admittedly has made it a problem for me in the past prior to me joining the council, and while part of it definitely was a skill issue at the time, I (and I know a good deal of other players who are experienced and I think would definitely deserve the right to vote but struggle with stamina) can have a frustrating time when you factor in fatique and tilt that have the potential to punish you with a greatly damaged (and sometimes even ruined) run off of one or two mistakes and/or the luck of the draw. When the primary initiative behind the current system is to gauge your experience, that can be extremely upsetting, for sure. I think for that reason it's absolutely worthwhile to have a dialogue about; I was having a chat with viivian about it the other day, and I was waffling a bit on opening a policy review thread. Though I have ultimately decided against it since, like I said, the current system is definitely not broken and works quite well, and allocating resources toward addressing the system when there are other priorities is a bit rough to justify imo. Unless there is a massive wave of support I probably won't be doing so.
___
However, while I'm here, I did have two ideas on how to, specifically, supplement the reqs process. I don't believe removing a functional system is a particularly great idea regardless of its flaws, as there isn't anything fundamental that makes it broken or overtly and arbitrarily skewed toward one group of players.
1: My first suggestion is to implement suspect tours in some way. This is a pretty common practice in other/unofficial metagames and RoA suspects mostly because they don't have active ladders, but to me it's an excellent voting alternative that focuses exclusively on metagame and skill knowledge. The primary caveat of this is that the scale of OU suspects comparatively can complicate things for the OU room staff who already have a ton on their plates, and voting reqs cutoffs are things that have to be sorted out too. It's not a perfect alternative given the circumstances behind OU but I do think they excel when utilized properly.
2: My second suggestion is one that's a bit rough around the edges, but specifically is a bit inspired by the crazy run of LusterSN who had literally >100 games before qualifying and hit 1900 before reaching 80% GXE. Another potential alternative could be having an ELO cutoff alongside a GXE cutoff. This one is one that would need a lot of sorting out regarding what thresholds are truly "correct" so to speak, but there is potential in it. I think a lot of people often dismiss ELO as a way to evaluate someone's experience in the tier. There comes a point where you get past the garbage and truly have to contend with competent players and the metagame at its fullest force. Even if your GXE isn't particularly the best, being at high ELO means you are still effectively winning enough games to even be there in the first place, which requires competency and some degree of experience to achieve and maintain. I personally don't like this one as much as the suspect tour option because of the fact that it's far more fickle to establish boundaries for and could lead to ladder saturation, but whether these are dealbreakers or even problems at all I think is up for debate.
___
My own stance is that having voting alternatives alongside the current GXE-based system could be an actively positive thing to encompass a greater range of skilled/experienced players who have a harder time with the stamina-oriented demands of the current system. These are things I personally would support implementing if we have the resources/time, but it is not something that I believe is mandatory or at least not worth prioritizing compared to actually working on the tier. Regardless, I hope this can provide some food for thought.
If Suspect tests become BO3 before Team Tours (the actual thing that would be better in that format) become BO3, I will be mad. Suspects are meant to be done in 1 or 2 days if you have the right team, BO3 would make it a long nightmare. Pretty sure that if Kyurem Suspect was BO3, only the biggest Kyurem haters or coolest Kyurem defenders (like me) would bother to even try that.
Okay, so after testing out a galarian articuno team, I believe it is so close to be great. It can very easily destroy stall teams (unless they run a tera psychic clodsire, which yes, I did run into and it stonewalled most of my team), while also being great against balance teams.
The partners it is best with are primarina and ceruledge (sidenote, bulk up ceruledge is quite great, though it does get hardwalled by tusk and gliscor, though if you can get rid of those it can run wild). Primarina can check the dark and ghost types that threaten garticuno, while ceruledge can counter the most common steel types that threaten it, except for gambit.
It only needs a bit more speed to be great, because it sometimes feels like it misses out on crucial benchmarks, for example it speed ties gliscor, which can get off a toxic and ruin a sweep. It also really needs psyshock to not get walled by clodsire or blissey, but wants the healing denial of psychic noise. I would say run psyshock because it is more valuable. I've been running a tera fire blast set, with calm mind, and it can be great, but does require support.
Articuno-Galar @ Heavy-Duty Boots
Ability: Competitive
Tera Type: Fire
EVs: 252 SpA / 4 SpD / 252 Spe
Timid Nature
IVs: 0 Atk
- Psyshock
- Hurricane
- Calm Mind
- Tera Blast
(Also, air balloon gambit with tera ghost can be great against tusk, just to make tusk stabs not work, which can stonewall any bulk up rapid spin sets, lmao)
The partners it is best with are primarina and ceruledge (sidenote, bulk up ceruledge is quite great, though it does get hardwalled by tusk and gliscor, though if you can get rid of those it can run wild). Primarina can check the dark and ghost types that threaten garticuno, while ceruledge can counter the most common steel types that threaten it, except for gambit.
It only needs a bit more speed to be great, because it sometimes feels like it misses out on crucial benchmarks, for example it speed ties gliscor, which can get off a toxic and ruin a sweep. It also really needs psyshock to not get walled by clodsire or blissey, but wants the healing denial of psychic noise. I would say run psyshock because it is more valuable. I've been running a tera fire blast set, with calm mind, and it can be great, but does require support.
Articuno-Galar @ Heavy-Duty Boots
Ability: Competitive
Tera Type: Fire
EVs: 252 SpA / 4 SpD / 252 Spe
Timid Nature
IVs: 0 Atk
- Psyshock
- Hurricane
- Calm Mind
- Tera Blast
(Also, air balloon gambit with tera ghost can be great against tusk, just to make tusk stabs not work, which can stonewall any bulk up rapid spin sets, lmao)
BO3 ladder makes sense for smth like a fighting game where there is a guarantee that the set will be over within 15 minutes, but it would be dogshit for this because this is Pokémon and as such you could be randomly matched up vs a stall team and end up being stuck switching around unproductively for 30+ minutes per game. Getting locked into a BO3 for smth like that would be torturous and would completely kill any motivation to ladder for anyone who has literally anything else competing for their time.
That said, having more ways to qualify is always better bc ladder grind is really not that appealing regardless if you aren’t a specific sort of player. It does have the positive side effect in that it forces badgewhores who only play the tier for the suspect test/TC badge to at least have a bare minimum number of games played before voting, but realistically, as long as ladder is used (which should not be changed), it will always be extremely exploitable and only really serve as a basic competency check.
Whether it being a mere competency test is good enough is up for debate, but short of completing gutting the process, making it harder to run, and consequently having it yield a smaller, worse voter pool, there isn’t really a good argument that it makes sense to make it stricter. I think that any argument that it shouldn’t just be a basic competency test is bad regardless though, as making it so relatively straight-forward to qualify for reqs leads to a decentralised voter pool and the test is, as such, much less vulnerable to manipulation and a bit further removed from just reflecting the opinions of a biased and interested in-crowd than it would be otherwise. We’ve had cliquey tiering practice/policy before (sometimes by design, sometimes inadvertent) and it has, without exception, been very bad and led to a lot of community discontent.
That said, having more ways to qualify is always better bc ladder grind is really not that appealing regardless if you aren’t a specific sort of player. It does have the positive side effect in that it forces badgewhores who only play the tier for the suspect test/TC badge to at least have a bare minimum number of games played before voting, but realistically, as long as ladder is used (which should not be changed), it will always be extremely exploitable and only really serve as a basic competency check.
Whether it being a mere competency test is good enough is up for debate, but short of completing gutting the process, making it harder to run, and consequently having it yield a smaller, worse voter pool, there isn’t really a good argument that it makes sense to make it stricter. I think that any argument that it shouldn’t just be a basic competency test is bad regardless though, as making it so relatively straight-forward to qualify for reqs leads to a decentralised voter pool and the test is, as such, much less vulnerable to manipulation and a bit further removed from just reflecting the opinions of a biased and interested in-crowd than it would be otherwise. We’ve had cliquey tiering practice/policy before (sometimes by design, sometimes inadvertent) and it has, without exception, been very bad and led to a lot of community discontent.
ok so apparently god hated this idea because i was locked out of work on my first day and showdown and discord are both down right now. i should have just not gotten out of bed
the people getting reqs *are* the median players. if they're consistently spoiling your climb then i have news abt your position on the bell curve
i'm capable of getting to the 1700s, i feel like i deserve a say
I have seen this team structure like 50 fucking times, low ladder is hell.
fr it's either this or rain
the worst part about it is that you can't even really tell from team preview what they have
is it LO or screens Deoxys
its split evenly between physical superpower, scarf, and tera blast stellar enam
gouging fire runs burning bulwark ONLY when I click first impression
I swear people didn't change their teams this much before whats even going on