Metagame Views From The Council

[Mimikyu Stardust]

c) The mod we currently use for sleep is unfaithful to cartridge mechanics, which would *maybe* be okay in exceptional circumstances to keep the game playable (such as the various RBY desync patches on sim), but we do not necessarily feel that this warrants such a mod.

Gonna be honest here, a sleep ban to keep it consistent with the cartridge should've happened a long time ago. I do not feel that sleep is unbalanced at all just annoying, but i do feel dirty spamming spore with amoonguss to stall PP with a bulkier team, knowing that i can't replicate this in the real game and it would break the gentlemen rule.

 

[Heatranator]

"it seems like everyone is just talking about Darkrai, clearly Darkrai is the issue, it's all Darkrai, everyone's just saying Darkrai, the only example is Darkrai"

the person who introduced the topic, in the very first post about it:

Okay, but non of these besides valiant maybe is really broken. Do people genuinelly think amoonguss or brute bonnet is broken?

 

[ianlazerbeem]

Okay, but non of these besides valiant maybe is really broken. Do people genuinelly think amoonguss or brute bonnet is broken?

Broken or uncompetitve? Because a 100% chance to force the opponent to win 1-3 dice rolls in exchange for free turns seems a bit like the latter!

An alternative answer to this question is, would they be without the mod we have implemented?

 

[Heatranator]

Broken or uncompetitve? Because a 100% chance to force the opponent to win 1-3 dice rolls in exchange for free turns seems a bit like the latter!

Well, is paralysis banworthy? By your logic, you can get 25% of the time each turn to not even move. And the condition does not have a timer. That's technically uncompetitive. I know that is a low hanging fruit, but we got to think about it in what you are saying.

 

[ianlazerbeem]

Well, is paralysis banworthy? You can get 25% of the time each turn to not even move. And the condition does not have a timer. That's technically uncompetitive.

Only if you can answer, what is the difference between a 25% game winning outcome if it happens at a particular moment out of the users control and a near 100% game winning outcome at the user's control of timing. The fact is, vs HO teams Spore is a 100% accurate ohko move

I can even answer, the difference is 75%, and control of the user. Free turns sometimes happening vs choosing to get 1-3 free turns is incomparable

 

[TheTraininator]

Are we really talking about banning a healthy Pokémon to keep a pet mod enabled that makes an incredibly degenerate mechanic slightly less degenerate?

I’m sorry, I don’t support a Darkrai ban in the slightest. I’m not even convinced the Hypnosis set it has is broken; it’s just kind of uncompetitive for the same reasons Hypnosis is always an uncompetitive move. If we really want to preserve sleep so bad, we can update sleep clause to remove the mod elements of it; allow partial chance sleep moves & abilities to be able to roll sleep multiple times, (of which there are only Relic Song, Effect Spore, and Dire Claw, the former two of which aren’t problems and the latter is banned by default because Sneasler is), and just like… prevent players from clicking sleep moves multiple times at all instead of allowing them to proceed as if they somehow didn’t use a move that turn, or make clicking a non-partial chance sleep move into a forfeit if you’ve already slept something.

Or you could just make the whole thing easier and ban sleep. Or just do nothing because I’m not convinced there’s even a problem that wasn’t here every gen we’ve had sleep legal.

 

[Ehmcee]

Well, is paralysis banworthy? By your logic, you can get 25% of the time each turn to not even move. And the condition does not have a timer. That's technically uncompetitive. I know that is a low hanging fruit, but we got to think about it in what you are saying.

That's not a very effective comparison, Paralysis has heaps more counterplay in two different types being immune to it (with the exception of glare vs grounds),

Additionally, you have way more control compared to Sleep, Paralysis is only a 25% chance to not move, compared to 100% -> 66% -> 33% -> 0% depending on the amount of turns asleep, Sleep has an extremely higher chance of just incapacitating your mon.
You actively have to do nothing to get rid of it as well, you're full on forced to waste turns just to gamble on wether or not you will wake up.

 

[Heatranator]

Only if you can answer, what is the difference between a 25% game winning outcome if it happens at a particular moment out of the users control and a near 100% game winning outcome at the user's control of timing. The fact is, vs HO teams Spore is a 100% accurate ohko move

Okay, that's against one team which is known to be ride or die. HO teams need to have ways to deal with spore ammonguss otherwise they haven't constructed a good team. If a darkrai/valiant misses a hypnosis against a HO team, they most likely die on that turn or the next turn. The thing is, hypnosis always has low accuracy, you are taking a risk in using it. If it does hit, yes, it does have high reward. However, if you miss, it has no reward and you get punished by the opposing player. Every time you use it, you have a chance to miss and get punished by your opponent. I know the paralysis arguement was dumb, I intentionally said it to show its dumbness, because uncompetitive is a weird topic. Paralysis may have more counterplay, but it also is way more spammable, meaning it can be put on multiple mons at the same time.

Do I think paralysis should be banned? No way in hell. But sleep has immense downsides if you miss, which has always held it back.

 

[ianlazerbeem]

Okay, that's against one team which is known to be ride or die. HO teams need to have ways to deal with spore ammonguss otherwise they haven't constructed a good team. If a darkrai/valiant misses a hypnosis against a HO team, they most likely die on that turn or the next turn. The thing is, hypnosis always has low accuracy, you are taking a risk in using it. If it does hit, yes, it does have high reward. However, if you miss, it has no reward and you get punished by the opposing player. Every time you use it, you have a chance to miss and get punished by your opponent. I know the paralysis arguement was dumb, I intentionally said it to show its dumbness, because uncompetitive is a weird topic.

Bro the HO teams are the ones using sleep bullshit to get the free turns. The very first post was about the disturbing trend of lead Amoonguss, and re hypnosis it's not "balanced because of the low accuracy", it is complete bullshit because of the low accuracy, you could say the very same thing about ohko moves. In fact:

If a [Ting-Lu] misses a [Fissure] against a HO team, they most likely die on that turn or the next turn. The thing is, [Fissure] always has low accuracy, you are taking a risk in using it. If it does hit, yes, it does have high reward. However, if you miss, it has no reward and you get punished by the opposing player. Every time you use it, you have a chance to miss and get punished by your opponent.

How different at the end of the day are these "sleep moves" from double team and ohko moves? Sleep moves functionally ARE ohko moves when used by hyper or bulky offense teams.

 

[Ehmcee]

Paralysis may have more counterplay, but it also is way more spammable, meaning it can be put on multiple mons at the same time.

Do you realise the problem with this argument, the only reason it's more spammable is because we artificially nerfed sleep to make it only usable on one pokemon. Sleep is infinitely more "spammable" but the only reason we can't do it is because we had Sleep Clause applied in the first place.

Uncompetitiveness isn't a black and white "does this require luck to happen", the degree at which it happens is important to consider, as well as how problematic the community tends to see it.

 

[nimzowitsch]

I'm late to the party but I agree with most of the points here. While I think there are bigger fish to fry right now, I would be completely happy to see Sleep-inducing moves be entirely banned. Firstly, it would cut the proverbial Gordian's knot about Sleep Clause being not cart accurate--if Sleep is completely banned then the whole debate about Sleep Clause becomes irrelevant. In my view Sleep is also an uncompetitive RNG clown show as Sleep duration varying from 1-3 turns can have a huge impact on the game, not to mention that most sleep moves (esp. the ones on the most problematic abusers such as Darkrai and Iron Valiant) are just straight up gambling that your move can land. In some cases, if you have a good MU and you hit the jackpot with Sleep (as in you don't get an early wake) you pretty much just win on the spot and your opponent can't do much about it. There's a huge difference between a Darkrai/IV hitting Hypnosis on a key check, setting up with the free turns and killing the check as it stays asleep, and Teraing through the team's other check to win, and a Darkrai/IV that misses Hypnosis or gets an early wake and just dies. The difference between the former and latter literally comes down to just luck. I get that there is counterplay to Sleep such as Sleep Talk and such but with the Tapus gone not every team can fit Sleep counterplay, and we shouldn't have to go out of our way to deal with this mechanic, especially not in a Tera metagame.

 

[nimzowitsch]

My take is either ban sleep or ban nothing. Darkrai is annoying but not that bad and I'd rather keep it than letting it go, especially over hypnosis of all things + the fact it wouldn't change the annoyance of val hypnosis cheese

Yeah I agree with you completely here. Darkrai would be a completely fine addition to the metagame if Sleep moves were banned.

 

[MeepBard]

seeing a lot of sentiments in this thread that the choices are either to uphold a dated, non-cart accurate clause or to ban sleep entirely and i wanna point that sleep clause can just be reworked. as previously brought up in the numerous PR threads about sleep clause and even in this thread:

I would much prefer keeping sleep clause but coding it in a way that is cart accurate. Like a basic "you can't click a sleep move unless it's literally your only option" like you just got encored and it's your last Pokemon.

the only possible downside to doing this is the removal of a strategy in which you click a sleep move anticipating a wake up, but that's something that's only possible on sim anyways and if you did that in a cart tournament with a sleep clause rule you'd risk straight up getting DQed should your opponent switch. i think this is a near perfect solution that would appease both camps.

but as for whether or not i think sleep is broken in SV OU: it's clear to me that darkrai is what even put a sleep ban on the table, so i would say to ban darkrai. if sleep is still deemed uncompetitive, then we can reassess a sleep ban. i'm not convinced sleep is the common denominator when it really seems to just be darkrai that's an unreasonably strong abuser of it

 

[Ibaragi]

Fully support strengthening the limitations over sleep. Not sure on whether to allow exceptions for Yawn, or Dire Claw and moves like that, but some action should definitely be taken.

This conversation has been brought up by Darkrai, but as the post that opened the debate explained, this is just the last straw. Sleep has been contentious and a target of discussion for a very long time that has simply been ignored by past councils and even this council in the past due to the many complications that this topic brings up (policy reviews, touching on the already complicated topic of Sleep Clause itself, the massive level of support that would be required to pass such a great change...).
Darkrai is not really doing anything ban-worthy on its own; it's doing the same thing Iron Valiant was doing before, just with the difference that Iron Valiant has a wide enough movepool to justify not running Hypnosis at all often enough and most people don't want to play the luck game of Hypnosis anyway. But these issues can and will be repeated by any top speed-tier mon with access to Hypnosis, or any sleeping move, and will bring the problem where they're "too inconsistent to be banned, but too impactful to be ignored".

I'd also like to point out that Sleep is also stronger now than it has ever been; the faster the game becomes, the more power crept our attackers are, the closer Sleep just becomes effectively the same as OHKO moves. Moves that were banned not because they were too good, but because of how non-competitive the impact they gave on a random chance was. I think Sleep Moves now are just as toxic as OHKO moves when you factor in the accuracy difference and should not exist at all.
But I have particular hatred towards Hypnosis and similar "unstable" sleep moves. That move is very wide-spread, no constraint has ever been taken on the speed of its users and has a fully random, but high enough chance to mess up any fight, turning it into a coin toss. Compare that to 4 Fully Evolved mons getting access to Spore, with the two new ones being A) Toedscruel: locked to using it at -1 priority and B) Brute Bonnet: low base speed, stats fine-tuned to NEVER get a +1 Speed from Booster Energy, can only boost its speed with Trailblaze while sporting one of the worst defensive types in existence. Yawn is also much more acceptable by giving no direct advantage and giving the receiver a chance to react, a decision to make.
I feel Sleep Clause is enough to balance moves that are delayed like Yawn or that are very limited in distribution like Spore, but moves like Hypnosis are "balanced" solely around them being subject to randomness, which should be unacceptable from a competitive perspective given their impact and are, again, akin only to OHKO moves.

I honestly don't care between maintaining part of Sleep in moves that aren't random (Spore, Yawn), but absolutely want moves like Hypnosis to be gone entirely. I prefer to be put to sleep 100% of the time by an Amoongus or a Breloom specifically than to have to fear the possibility of an Iron Valiant, Darkrai, Alolatales or any random 100+ Speed mon to win a better-than-a-coin-flip chance to get free turns and probably win the whole match out of it.
If we're using this opportunity to get rid of the Sleep Clause too, though, then yes, Spore also needs to go, of course. Still conflicted on Yawn, but would take either result.

 

[Heatranator]

Bro the HO teams are the ones using sleep bullshit to get the free turns. The very first post was about the disturbing trend of lead Amoonguss, and re hypnosis it's not "balanced because of the low accuracy", it is complete bullshit because of the low accuracy, you could say the very same thing about ohko moves. In fact:



How different at the end of the day are these "sleep moves" from double team and ohko moves? Sleep moves functionally ARE ohko moves when used by hyper or bulky offense teams.

Lead ammonguss is an interesting trend for sure, but if a HO team has ammonguss, you know what it is going to do, meaning you can hit it extremely hard with your wallbreaker, which will most likely destroy ammonguss on its own. The difference between sleep moves and double team/ohko moves is that it can be procced multiple times, something that is not possible with sleep moves right now. That may be because of sleep clause, but the simple fact is right now, you cannot spam sleep reliably. Spore users are slow, meaning the opponent has a turn to wake up and hit you. Hypnosis has 60% accuracy, meaning they have to get lucky. You cannot spam it, that's the difference. There is no stacking it, you can't click it and make multiple mons fall asleep.

Do you realise the problem with this argument, the only reason it's more spammable is because we artificially nerfed sleep to make it only usable on one pokemon. Sleep is infinitely more "spammable" but the only reason we can't do it is because we had Sleep Clause applied in the first place.

Uncompetitiveness isn't a black and white "does this require luck to happen", the degree at which it happens is important to consider, as well as how problematic the community tends to see it.

This is a better arguement, they say that sleep is not broken because of the mechanic right now in place. Which means that at this moment it is not banworthy. I agree that uncompetitiveness is not black or white, but sleep feels like in the middle of the scale. Yes, it can be destructive, but the main abusers have downsides that keep it in check, ammonguss is slow and has to take a hit first, while hypnosis users are frail. If sleep gets banned, I won't cry over spilt milk, but it feels like to me we are either cherrypicking and not banning darkrai which is the main problem to me, or we are trying to get rid of something that doesn't really need changing because it's worked well so far.

 

[658Greninja]

https://replay.pokemonshowdown.com/gen9ou-2031210136-txw9iyy6mvk7mgs483y90pox77liox4pw

Whether or not this is the doing of solely Darkrai, we could all agree that there is something up with teams just getting 6-0d on turn 1 due to sleep procs. Take this game for example.

Darkrai gets sent out first and lands a Hypnosis on the Clefable. Then it proceeds to get off 2 Nasty Plots because Clefable won’t get out of bed and Darkrai one-shots with Ice Beam boosted by Terastilization. Dozo comes in and Darkrai lands another Hypnosis. After the Dpulse, Dozo wakes up prematurely.

Wanna know what happens next?

Darkrai lands another fucking Hypnosis!

Afterwards, Darkrai wins the speed tie with the opposing Darkrai and at that point, it was joever.

You could argue sleep is the entire problem, but here is the thing. Darkrai can get up multiple sleep attempts at once thanks to Focus Sash and its speed, while it gets insane rewards for landing Hypnosis.

Valiant is tied to Booster and it loses the 1v1 against Ghold, the main sleep absorber.

Venu and H-Lilligant are only threatening under sun.

Amoonguss is the best out of them, but it is also much slower and passive.

Plus Darkrai was the one that sparked this debate. If we are to address the issue of sleep, we should suspect Darkrai.

 

[Joseph Gomes]

Okay, that's against one team which is known to be ride or die. HO teams need to have ways to deal with spore ammonguss otherwise they haven't constructed a good team. If a darkrai/valiant misses a hypnosis against a HO team, they most likely die on that turn or the next turn. The thing is, hypnosis always has low accuracy, you are taking a risk in using it. If it does hit, yes, it does have high reward. However, if you miss, it has no reward and you get punished by the opposing player. Every time you use it, you have a chance to miss and get punished by your opponent. I know the paralysis arguement was dumb, I intentionally said it to show its dumbness, because uncompetitive is a weird topic. Paralysis may have more counterplay, but it also is way more spammable, meaning it can be put on multiple mons at the same time.

Do I think paralysis should be banned? No way in hell. But sleep has immense downsides if you miss, which has always held it back.

Sleep is already nerfed and it's still more broken than paralysis. You can put a same mon to sleep as many times as you want. During that period, the opponent mon completely incapable of doing anything. You can't paralyze the same mon twice unless it's cured. No mon also is immune to sleep condition by default unless they have an ability to block status move(s). Electric types are completely immune to paralysis (even glare), and ground types are immune to thunder wave.

 

[TheTraininator]

MeepBard I agree with the rest of your post wholeheartedly but

but as for whether or not i think sleep is broken in SV OU: it's clear to me that darkrai is what even put a sleep ban on the table, so i would say to ban darkrai.

There’s a sentiment in this thread that Darkrai is the main problematic sleeper and I just… don’t agree with this when Valiant exists? I think Darkrai’s introduction into the tier made people curious to try sleep more extensively and then they rediscovered “oh wow, this is an extremely degenerate, luck-based, high-variance, unfun mechanic”. But it’s always been this, both before and after Darkrai has been in the tier. I’ve seen plenty of people complaining and warning others of Hypnosis Valiant, but it was just kind of never used consistently enough and largely dismissed as a meme. The only real difference between the two, other than Bad Dreams which is just kind of a bonus, is that Gholdengo switches better into Valiant than into Darkrai. Better is the operative key word here because Swords Dance -> Knock Off is always a thing that you’re risking.

TL;DR: If you think Hypnosis Darkrai is banworthy but Hypnosis Valiant isn’t worth talking about, I think you’re splitting hairs.

Valiant is tied to Booster and it loses the 1v1 against Ghold, the main sleep absorber.

Don’t switch in on a Swords Dance instead of a Hypnosis then.

+2 252 Atk Tera Dark Iron Valiant Knock Off (97.5 BP) vs. 252 HP / 196+ Def Gholdengo: 494-584 (130.6 - 154.4%) -- guaranteed OHKO

 

[Divine Retribution]

Why are we even discussing Paralysis? These two mechanics are so far apart in scope it's not even funny, and even outright banning Sleep moves is controversial. Can we keep this discussion remotely realistic, please?

 

[Azery]

Opinion of someone who hasn't delved into DLC2 yet but I think this is kind of pointless (?). If Hypnosis is in the conversation because Darkrai got unbanned and is something that actually has the space to run the move, then handle Darkrai. I don't much consider it broken on any of its other users; As said above, Valiant has better options + loses to Ghold, Ninetales is niche, Venu and Lilligant are sun-exclusive and there are immunities. Only Amoonguss has reliable sleep among them and even then, there are immunities + its slow and passive.

Editing a gentleman's agreement that has historically been considered fine and understandable -- if slightly controversial for its scope-- outside of Gen 5 where Sleep was a literal KO button simply due to a Pokemon who has historically been banned for abusing fast, reliable sleep to sweep, because it... starts abusing fast, somewhat less reliable sleep to sweep is perhaps more an indicator that the issue at hand isn't Sleep but Darkrai itself.

 

[Joseph Gomes]

Opinion of someone who hasn't delved into DLC2 yet but I think this is kind of pointless (?). If Hypnosis is in the conversation because Darkrai got unbanned and is something that actually has the space to run the move, then handle Darkrai. I don't much consider it broken on any of its other users; As said above, Valiant has better options + loses to Ghold, Ninetales is niche, Venu and Lilligant are sun-exclusive and there are immunities. Only Amoonguss has reliable sleep among them and even then, there are immunities + its slow and passive.

Editing a gentleman's agreement that has historically been considered fine and understandable -- if slightly controversial for its scope-- outside of Gen 5 where Sleep was a literal KO button simply due to a Pokemon who has historically been banned for abusing fast, reliable sleep to sweep, because it... starts abusing fast, somewhat less reliable sleep to sweep is perhaps more an indicator that the issue at hand isn't Sleep but Darkrai itself.

We are not discussing whether Darkrai is broken. We are discussing whether sleep itself is inherently uncompetitive.

 

[Hydrametr0nice]

Ok, let's look at numbers!

Turn	Sleep % of failing to act	Accumulated nº of moves hit	Paralysis % of failing to act	Accumulated nº of moves hit
1	100,00%	0,00	25,00%	0,75
2	66,66%	0,33	25,00%	1,50
3	33,33%	1,00	25,00%	2,25
4	0,00%	2,00	25,00%	3,00

So, by turn 4 (when you are guaranteed to wake up), you are only expected to have missed one additional move compared to being paralyzed.
Let's just remember that you can only put 1 mon to sleep at a time and, If you were paralyzed, you remain paralyzed forever and have your speed reduced by 50%.

January stats:

Pokémon	Usage %	Move	Chance of having the move	Frequency
Iron Valiant	12.314%	Hypnosis (60%)	1.496%	1 in 543 battles
Darkrai	11.537%	Hypnosis (60%)	30.526%	1 in 28 battles
Ninetales-Alola	4.422%	Hypnosis (60%)	22.137%	1 in 102 battles
Amoonguss	3.850%	Spore (100%)	96.040%	1 in 27 battles
Venusaur	2.011%	Sleep Powder (75%)	2.308%	1 in 2174 battles
Lilligant-Hisui	0.661%	Sleep Powder (75%)	23.415%	1 in 645 battles
Brute Bonnet	0.094%	Spore (100%)	93.348%	1 in 1136 battles

Here, I only put OU (because I'm that nice to you, paralysis lovers) and removed Darkrai, Dragonite, Iron Valiant, Kingambit and Raging Bolt because their Thunder Wave values were too low (not lower than the multiplied % of Venusaur and BrUtE bOnNeT). And let's not forget to compare the moves' accuracies!

Pokémon	Usage %	Move	Chance of having the move	Frequency
Gholdengo	12.314%	Thunder Wave (90%)	8.745%	1 in 63 battles
Hatterene	15.788%	Nuzzle (100%)	43.721%	1 in 14 battles
Slowking-Galar	12.601%	Thunder Wave (90%)	30.166%	1 in 26 battles
Deoxys-Speed	12.246%	Thunder Wave (90%)	2.401%	1 in 340 battles
Archaludon	10.002%	Thunder Wave (90%)	3.299%	1 in 303 battles
Dragapult	9.797%	Thunder Wave (90%)	19.962%	1 in 51 battles
Clefable	8.558%	Thunder Wave (90%)	18.146%	1 in 64 battles
Serperior	8.451%	Glare (100%)	81.566%	1 in 15 battles

And did I mention that Zapdos has Static?
So, aside from the cartridge arguments (which are quite valid), let's just accept that paralysis and other mechanics are just as uncompetitive as sleep!

 

[Cuddly]

Opinion of someone who hasn't delved into DLC2 yet but I think this is kind of pointless (?). If Hypnosis is in the conversation because Darkrai got unbanned and is something that actually has the space to run the move, then handle Darkrai. I don't much consider it broken on any of its other users; As said above, Valiant has better options + loses to Ghold, Ninetales is niche, Venu and Lilligant are sun-exclusive and there are immunities. Only Amoonguss has reliable sleep among them and even then, there are immunities + its slow and passive.

Editing a gentleman's agreement that has historically been considered fine and understandable -- if slightly controversial for its scope-- outside of Gen 5 where Sleep was a literal KO button simply due to a Pokemon who has historically been banned for abusing fast, reliable sleep to sweep, because it... starts abusing fast, somewhat less reliable sleep to sweep is perhaps more an indicator that the issue at hand isn't Sleep but Darkrai itself.

It matters for multiple reasons, big 2 being
A. current sleep clause is not how cartridge gameplay works as sleep moves don't fail when used on a second Pokemon.
B. Sleep as a whole is uncompetitive and just encourages degenerate gameplay
If Sleep Clause didn't exist before, everyone would universally agree banning sleep moves would be the right call as some complex gameplay changing clause would be dumb

 

[Heatranator]

Sleep is already nerfed and it's still more broken than paralysis. You can put a same mon to sleep as many times as you want. During that period, the opponent mon completely incapable of doing anything. You can't paralyze the same mon twice unless it's cured. No mon also is immune to sleep condition by default unless they have an ability to block status move(s). Electric types are completely immune to paralysis (even glare), and ground types are immune to thunder wave.

Technically, grass types are immune to powder moves which includes spore and sleep powder. The other thing is that the only reliable sleep move is spore, which is restricted to slow mons and thus, they have to take a hit. Every other sleep abuser has to use hypnosis, which is inaccurate and while it can steal games, relying on a move with low accuracy isn't good.

Also, I want to ask, what is exactly cart accuracy, because I see people throwing it out there and I think it doesn't really work because cart accuracy by my definition would be not a good metagame as evasion clause technically is against cart accuracy.

 

[Wraks]

There are two issues here: 1) whether sleep clause is itself outdated/incompatible with modern tiering principles (a claim which may affect sleeps present in future gens); and 2) whether sleep is broken in current Gen IX OU, even with current sleep clause, or if it might just be darkrai that is the problem.

1) Should sleep clause be removed entirely on principle, including for future iterations of OU?

If you want the most brutal argument against sleep, just picture if Sleep moves were introduced in Gen IX and we did not have close to two decades of Sleep clause. How different would the discourse be on sleep?

^This train of thought is the nail in the coffin for sleep from a self-consistency standpoint, in my opinion. If sleep were a new status, there is absolutely no way we would go out of our way to form a complex ban to keep it around. Sleep clause's blatant contradiction with modern OU tiering philosophy has bugged me for a long time. The fact that it's current implementation requires deviation from cart mechanics is even worse. If it were up to me, I would just get rid of (non self-inflicted) sleep for all current gen OUs going forward.

That said, as much as tiering is based on clear principles and the logic that flows from them, it must be acknowledged to some extent that tradition/nostalgia plays a big part in deciding what things are allowed. To that end, while I do not like sleep clause at all, I do have sympathy for people who would like to keep it around because the availability of sleep moves in some form has historically been a part of competitive singles. If a large part of the Smogon community as a whole thinks that sleep clause is worth keeping solely out of tradition, I don't think it would be wise to remove it for all time, even if I dislike it. For the same reason, I want to be clear that should this line of thinking become policy, I think both I and most others would NOT want this to be a retroactive change to prior generations (this is obviously just an Gen IX thread but when making broad changes like this often eyes soon turn towards making old gens consistent with modern ones). There is no need to go overturning other settled generations to fit with our modern conceptions of what tiering should look like.

To touch on some sleep related odds and ends:
Dire claw can go with sleep, no need to make an exception (I feel likewise with relic song for consistency tbh but no one cares about meloetta so whatever). I do acknowledge that yawn has a more competitive flair than the rest of sleep moves, so I understanding wanting to separate it from the immediate sleep moves (you could imagine a ban on moves that cause instantaneous sleep), but at the same time I also think that the preference for less complex clauses in battle is a good one. So I can go either way here. Now what to do with effect spore causing sleep is a whole other issue; thankfully, right now there is no pokemon which has effect spore as its only ability, so I think it is fine to ban the whole thing, but one could imagine a future where an otherwise balanced pokemon with only effect spore getting booted from OU simply due to its ability (getting frosslass vibes here); I will leave that to future players to deal with this should the problem arise.

2) Setting that aside, is sleep as it currently exists broken in Gen IX OU? Or is it Darkrai?

This is a tough call. I will start by saying that in general, I do NOT think that sleep cause in its current implementation is inherently uncompetitive, in general (although how it applies to hypnosis in this generation in particular will be discussed later) There is obviously a large element of luck in sleep, but even so several previous metagames have managed to remain skill based and competitive, even when sleep is a big part of the ecosystem.

I think generally sleep's power is greater in more offensive metagames, where it is harder to recover from the instant loss of a pokemon due to sleep, and the momentum gained by putting a pokemon to sleep can be immediately capitalized upon. I think the thing that sets Darkrai (as well as to a lesser extent Iron Valiant and A-Ninetales) apart is that these pokemon get you so much immediate advantage from a 60% roll that it gets to a point where the outcome of one or two attacks can be disporportionally game warping, leading them to be uncompetitive. As a result, I do think it makes sense to do something about this mechanic.

On the flipside though, in high power environments, the downside of just wasting a turn due to the low accuracy of most widely distributed sleep moves similarly becomes harder to justify. To that end, I am not sure if hypnosis darkrai (if it is allowed to exist as it does now) will ever take off consistently in a tournament environment; even when the roll is in your favor, missing a hypnosis can be a big risk when you have a lot riding on the line. This is not to say that I don't think that these sets will be used at all in tourneys (there are already some successful tourney results from hypnosis darkrai), but I suspect that should they be allowed to persist, in the long run these sets would fade out of popularity due to inconsistency (in contrast, in a ladder environment, where high variance is acceptable if the expected value is large enough, I think darkrai would remain a scourge). As a result, in the tournament metagame, I think ultimately it is spore amoonguss that will prove the most annoying. However, in the current moment, I doubt amoonguss alone is what is making people call for action on sleep. Iron Valiant and A-Ninetales are similar; I do think they have problematic elements with regards to sleep, but without darkrai I am not sure this issue would have generated as much attention.

On that note, I do think that in this particular instance, it is very difficult to separate darkrai from a total sleep ban. Is sleep clause really just largely uncompetitive in this gen, or is darkrai the one who is really pushing it over the edge? If I am trying to be dispassionate, I would maybe lean a little towards banning darkrai first before going all out on sleep (provided there is not a fundamental alteration towards the tiering policy on sleep, which I would prefer). I do think there is a good chance that sleep may remain uncompetitive even so, but there is no harm in revisiting this down the line and possibly freeing darkrai in exchange later. It just feels unfair to hit all sleep across the board when in this moment it is very unclear where the real source of the problem is. On the other hand though, I just don't like gimmicky sleep stuff lol (in the context of this gen), and if I were redesigning the tier to my own wishes I probably wouldn't include it, not to mention the fact that banning darkrai may just be delaying the inevitable. So ultimately I'm ok with either path.

TLDR
1) Sleep clause as it exists should be abolished in my view; it is bad enough that it conflicts with modern tiering philosophy, it is even worse that it breaks cartridge mechanics. However, I understand the community at large may feel some traditional attachment to sleep as it has been implemented historically, and I think it is ok to violate consistency for the sake of maintaining the status quo, provided that is what the majority feels (once again, if it were up to me, I would kick sleep out altogether

2) If we are not going to fundamentally rework sleep clause, and are just concerning ourselves with whether sleep clause is broken in Gen IX, I do think there is an issue with sleep that darkrai has brought to the forefront, but it is very difficult for me to say definitively whether it is sleep broadly or darkrai specifically that is the problem. I think the most "objective" option is to ban darkrai first, and then revisit this later if sleep remains an issue, but my personal feeling is that I do not enjoy sleep broadly right now and would not mind seeing it go entirely for this generation, so long as most others agree.