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.
I never felt it was overcentralizing from a teambuilding perspective or unfun to play against. On the builder side I've never slapped venu or victrebeel on teams carelessly, sure that I would get some value due to their ability to sleep. In fact, I find victreebel quite weak in the current meta. Haunter falls in the same category, and outside of the lead position I get most of its value from walling dnite and persian. Hypno is of course stellar, but I think we would be using Hypno even without sleep. I agree that hypnosis adds variance to the Hypno mirror, but I don't think we can get away from it. Mirrors are flippy by definition and even without sleep games where one Hypno gets full parad 3 times in a row will feel bad to play. It's true that if both Hypnos go for sleep there isn't a lot of skill involved if any, but I'm not sure that going for flips is the optimal play. Some teams do not carry sleep Hypno because they'd rather have a more consistent sponge and want to spread paralysis to win via wrap later in the game. Moreover, allowing Hypno to carry hypnosis allows more options in the builder. Choose hypnosis and seismic toss and you'll better off in the mirror, choose rest and s toss and you will be a better sponge and improve the match-up vs kadabra, choose hypnosis and rest you match up vs kadabra gets worse but you improve vs tentacruel.
Speaking from the sensations of playing the game rather than building teams, sleep feels to me the status that give me the most counterplay. I can choose which mon is slept, allowing me to sacrifice the one I feel is least important to my strategy, deters my opponent from clicking wrap too eagerly, for fear of giving me free turns to wake up and gives me slight odds to flip a bad endgame if I manged to paralize all my opponent mons. Compared to para+wrap, where I'm just waiting for a miss, or freeze, where I might lose my defensive cornerstone for all the game sleep feels to me like a strong status to suffer but not unbearable.
hope to hear more thoughts from others but ig ill say a few things now, in addition to agreeing with everything emma and amaranth said.
~ first from my perspective sleep is modded, the default based on tiering policy would be to ban sleep unless it adds something that is ‘worth’ (by whatever standard) modding the game for.
~ i dont understand what sleep adds to or benefits the tier in any way. i agree with some of stunner’s comments such as that victreebel is bad, and u dont add even haunter just for sleep esp since haunter isnt at all reliable at getting sleep.
i dont rly understand why u ‘cant get away from the variance in the mirror,’ outspeeding twaving and getting 3 fps is obviously far far less than the chance to outspeed and land hypnosis. 1/2 * 1/64 is 1/128, that is 255-miss realm probability and nowhere near the 30% of getting outsped and hypnosised.
i dont understand why being able to choose between moves for hypno in the builder benefits the metagame, u dont know what opp hypno has until it clicks moves and u have to decide how to respond to it before that so playing against hypno is almost a bit random. u can make guesses based on current meta trends about the correct play vs opp hypno but thats kind of all u can do. or, u just assume it has hypnosis and attack it even if that means persian gets twaved or whatever. on paper i could see the argument that the options are interesting in the builder (idk if thats what stunner meant), but in practice, ig since the metagame is so offensive, u dont rly have the control over gameplay to implement that. eg u can run a moveset on hypno thats designed for kadabra matchup, but even assuming opp brings kad, they might j keep kad in back til endgame. u cannot rly realistically choose hypno’s matchups like that, what makes hypno so good is its ability to matchup vs whatever it is needed against in a specific game state.
~ there was also a suggestion on the discord that sleep provides structure to uu games, i didnt really understand it but stoss + twave hypno dittos provided sufficient ‘structure’ during period(s) when hypnosis hypno was rare,
~ i think it is worth noting that some ppl want sleep banned for different reasons. for myself, and i think for emma and amaranth as well, the problem is the severe variance it brings, particularly in hypno mirrors but not only there, that takes ‘too much’ control out of the players hands. (& for me its also a cart accuracy thing.) however, my understanding is there are other rby uu players who support sleep ban but have different reasoning for it.
~ hipmonlee’s math about rest-less hypno beating hypnosis-less hypno in a 1v1 about 70% is interesting, this is obviously quite a bit above 50%, but a 30% chance to instead lose a matchup that u were supposed to win is hardly predictable. now ofc this is 1v1 in a vacuum while in practice u can eg go kang after landing sleep so its prob a bit more than 70% favored to hypnosis. still, we’ve all seen the variance including games where hypnosis hypno j loses to the opp hypno.
~ im not sure if i understood everything but the second half of stunner’s post, as well as other convos eg on the discord, seems to be ‘there is other stuff that is bad too, why focus on sleep.’ i think bringing up other potential things to look at (whether mons or moves or w e) is valid but dont understand why it is an argument to “preserve” [a mod to cart mechanics]. removing anything like blizzard bslam wrap or twave would obv have drastic metagame impacts and there is a lot to consider with any such change. banning sleep is very narrow, all it does is knock a few low ranked mons out of the tier + remove a mechanic that to me at best contributes nothing and at worst takes a third or so of games out of the players hands. (and yes there are also games that get taken out of players hands in other ways, but its not like preserving sleep is somehow reducing those other forms of variance, its just added on top of them).
~ the best argument made so far to me is that the uu metagame is in a period of rapid development and maybe we have insufficient information to evaluate this rn. altho we have to test lapras and learn how to adapt to that anyway. ig my main thing here is 1 even if adaptation were to prove able to handle sleep (esp hypnosis hypno) decently consistently, sleep still doesnt add anything beneficial and def to mod the game for 2 no amount of metagame development can address the fact that u dont know hypnos set when it first comes out, and even if u did know the moveset its still a 50/50 with a outcome that massively influences the outcome, even more so than the 50/50s vs potential apt dnite, as was brought up during that suspect by players who saw/see hypnosis hypno as 100x more problematic than agiliwrap.
i wouldnt say im 100% voting ban but i havent heard any reasoning as to what sleep adds to the rby uu metagame (in any way).
some maybe notable replays:
https://replay.pokemonshowdown.com/smogtours-gen1uu-636614 unown gets an ‘ideal matchup’ vs my clef lead, maybe predicted me to lead clef idk, but instead hypnosis misses and hypno gets crit paraslammed into hbeam range, lol. hypno is unowns only sleeper and there isnt anything wanting to bring into the clefable hbeam either. also imo hypnosis gives clefable extra reason to t1 bslam rather than twave, as a 74% parad hypno would be arguably more annoying for my hypnosis hypno than 74% unparad, while the chance to paraslam and get a second hit is worth a lot more; tho maybe u click bslam in clef vs hyp regardless im not sure.
https://replay.pokemonshowdown.com/gen1uu-1606893745 phoopes plays very well, also lands hypnosis on my hypno partly due to me being risky and not assuming it was hypnosis (since the games i had to go off of hypnosis hypno wasnt at all universal from phoopes; but ofc during rapid metagame development going off of past games isnt rly reliable), im way behind til turn 26 when i go to slept hypno on golem’s sub (seemed to be my only win path), i insta wake and hit hypnosis flipping a pretty completely lost game to a win.
also in our game two, i go hypno on predicted dugtrio sub to sleep it, i think it was overall the best play available but obv high risk as if hypnosis doesnt hit before hyp dies then its gg. another example of how sleep encourages high variance plays. https://replay.pokemonshowdown.com/smogtours-gen1uu-635442
some clarification on the paralised hypno comment. Juoean is absolutely correct in saying that getting parad multiple times in a row as opposed to loosing the speed tie and getting slept. The point is that if we ban sleep Hypnos can run only one set (twave, rest, psychic and stoss) and the matchup is a literal flip. Both players know the optimal line and the one getting more unlucky with paras loses. With sleep allowed Hypnos have to drop either rest or stoss, making them weaker in either the matchup vs tentacruel/general sponge if they drop rest, or vs psychics by dropping stoss, or drop hypnosis and be worse in the hypno ditto.
Regarding the "just because other broken thing exists we should not give sleep a pass" argument, I sit on the opposite side of the fence. Sleep does not reduce the occurence/likelyhood you will be haxxed, but gives a way for you to counter it. You lost the persian ditto, but you can retaliate with sleep on the following turn, your tenta got frozen, but at least you manged to sleep theirs. Sometimes it happens that all the variance goes against you (multiple paras in crucial turns, you lose the sleep ditto and get frozen), but I am fine with calling that a cursed game and move on.
If we ban sleep we reduce one instance of uncompentitive variance, but we increase the impact of the others.
some clarification on the paralised hypno comment. Juoean is absolutely correct in saying that getting parad multiple times in a row as opposed to loosing the speed tie and getting slept. The point is that if we ban sleep Hypnos can run only one set (twave, rest, psychic and stoss) and the matchup is a literal flip. Both players know the optimal line and the one getting more unlucky with paras loses. With sleep allowed Hypnos have to drop either rest or stoss, making them weaker in either the matchup vs tentacruel/general sponge if they drop rest, or vs psychics by dropping stoss, or drop hypnosis and be worse in the hypno ditto.
Regarding the "just because other broken thing exists we should not give sleep a pass" argument, I sit on the opposite side of the fence. Sleep does not reduce the occurence/likelyhood you will be haxxed, but gives a way for you to counter it. You lost the persian ditto, but you can retaliate with sleep on the following turn, your tenta got frozen, but at least you manged to sleep theirs. Sometimes it happens that all the variance goes against you (multiple paras in crucial turns, you lose the sleep ditto and get frozen), but I am fine with calling that a cursed game and move on.
If we ban sleep we reduce one instance of uncompentitive variance, but we increase the impact of the others.
j want to respond to the first thing, hypno would not have only one moveset bc counter exists and currently counter is legal based on desync mod. would be a somewhat similar situation, u have to decide what move to drop for counter and some similar aspects in terms of matchups.
ofc i also did/do not support legalizing counter via desync mod, but j responding to this for accuracy.
~ and the ‘both players know the optimal line’ thing, not certain how to interpret what u meant here but obviously there are the switching to normal type on the predicted rest when hypno is at 47% and paralyzed, kang ofc also takes stoss well on switchin, all of the lines of play that were used when hypnosis hypno was rare. both players j sitting hoping for better rng wasnt what happened bc u gain an advantage if u can come in on predicted rest (and dont lose much switching kang into stoss either, can even do this at 74% if u want to.)
it definitely is not a accurate description of the metagame during those periods of time to say that hypno vs hypno was just rng coinflips bc the matchup wasnt in a vacuum even if psy twave stoss rest was the by far most used moveset
