Rolls seem problematic in this game

[Conda]

To @ Prawn
I'm going to make the points clearer - not that I agree or support them fully and think they are true, but because it is the correct hypothetical view being presented in this thread.

The point being made is that, if you cannot use a kill move to punish a roll, then theoretically you can roll to avoid having a kill move landed on you. If this was a stamina/hp game, then you could whittle down someone's HP and win by using jabs or throws to punish rolls. But in Smash 4, those do not lead to kill moves on many characters. Thus, top level players who are using a character that cannot punish rolls via a kill move will be unable to land kill moves against skilled top-level players who know how to roll to limit their opponent's options.

This will likely be a large factor in tiers a year or two from now - if you can roll dodge intelligently to literally and mathematically prevent a character from being able to land a kill move, then that will make those characters less competitive as they can be countered by a strategy.

It's the same as how characters who can be chaingrabbed uniquely - like Fox and Ness in Brawl - end up being low in their tiers because of their mathematical shortcomings. This will be the same case, if things shape up the way we are hypothesizing.

Thus, only once the meta develops will we see if every character can, in fact, mathematically punish anyone's roll with a kill move. If not, then it will matter for that characters' matchups. The only way this will not be the case is if every character can punish rolls with a kill move (which is likely, but also not confirmed yet).

 

[Prawn]

Yknow I thought I might've been being a **** but I just read a lot of this again and I still think most of this thread is silly.

Carry on with the discussion, I'm gonna bow out gracefully. But any new players heed my words. Rolling, even if your characters roll is OP, is a dangerous game. And the more you get away with it the harder the habit will be to break when you need to

That being said plz roll on me all day

Bye

 

[Conda]

Yknow I thought I might've been being a **** but I just read a lot of this again and I still think most of this thread is silly.

Carry on with the discussion, I'm gonna bow out gracefully. But any new players heed my words. Rolling, even if your characters roll is OP, is a dangerous game. And the more you get away with it the harder the habit will be to break when you need to

That being said plz roll on me all day

Bye

Nobody in the competitive scene should be confidently thinking anything is factually true about rolling's effect on high-level play until the meta develops, we do maths, and we see how things turn out. This thread is about rolls 'seeming' problematic - as if they may or may not have an effect on the meta down the road. The OP is not arguing that rolls are 'DEFINITELY' problematic. It simply seems as if they may be more powerful to the point of making them much safer to use than before, and thus more integrated into the meta.

We have to be agnostic and humble about it until we have hard facts. You can have a hypothesis, but there's no validity in confidence and assuredness at this early period in Smash 4's metagame.

 

[PizzaWenisaur]

I don't understand threads like this. We have quite a few that talk about various topics, but this one is the biggest one that seems to be strange like this.

Why are we discussing whether rolls, counters, the blastzones, airdodges, shields, or whatever else you might want to name are well designed game elements? I mean, I think all of these things (definitely including rolls) are fine, but even if they weren't, what are we accomplishing here by deliberating over whether they're bad mechanics? If rolling is overpowered, the only real competitive discussion I'd expect would be discussion of how to optimally roll a lot to win, about which characters can roll the best, and other such things, but that doesn't seem to be the tilt at all. We can't change the game so it doesn't really make sense to discuss how the game could be different or how we'd prefer the game to be. We're best off just accepting gameplay elements for what they are and asking ourselves how we can use them to our advantage.

What I really fear is that I really do think a lot of new players read threads like this. They're coming onto this forum looking for competitive information about the game, and in fact I think a lot of them are having trouble punishing rolls since rolls can be pretty strong in this game and new players often have trouble making reads (even in Melee where rolls were awful, a lot of newer players found them difficult to handle). I really fear a lot of them will see a topic like this and conclude "rolls are overpowered, I'm not bad, the game is just broken" instead of learning how to play better to handle rolls or even better how to use rolls well to allow themselves to benefit from them instead of suffer by them. That seems way worse as an outcome to me than whatever good (if any?) we could realize if we came to some kind of hypothetical agreement about whether rolls make this game better ore worse.

To specifically address the OP's concerns, I'll point out that online lag makes rolls seem WAY better than they are since the lag is the difference between your ability to react and punish versus not in a lot of real match situations. When I play online, I spam roll liberally and just get away with it; offline I'm a lot more careful (still roll, just a lot less) and am looking more and more to punish rolls (you can just dash attack them so much, and it's so easy). Then remember that even offline smash 3ds lags some (yeah, really), and imagine how much more you'll be able to hit on wii u when you can finally play this game with other humans with zero lag slowing things down and making stuff harder to punish. Of course, even if rolls really were just overpowered, it would be better to frame things as "rolls seem strong, how should we be using this to our advantage", but I think this is a better picture of how strong rolls actually are in this game (in short, they're a lag tactic).

Disagree, if we think a mechanic is bad we should talk about it and theory-craft why we think it's bad ( or not bad). And how Smash in the future might benefit from these changes. The fact of the matter is a new Smash is probably going to be made in the future - so it makes sense to be looking at problems and suggesting corrections even if we can't change this particular game. Especially if it might lead to a more fun and robust Smash 5. For this reason I can't agree with sugar-coating our opinions for those hypothetical new players.

 

[Prawn]

Yknow I thought I might've been being a **** but I just read a lot of this again and I still think most of this thread is silly.

Carry on with the discussion, I'm gonna bow out gracefully. But any new players heed my words. Rolling, even if your characters roll is OP, is a dangerous game. And the more you get away with it the harder the habit will be to break when you need to

That being said plz roll on me all day

Bye

Disagree, if we think a mechanic is bad we should talk about it and theory-craft why we think it's bad ( or not bad). And how Smash in the future might benefit from these changes. The fact of the matter is a new Smash is probably going to be made in the future - so it makes sense to be looking at problems and suggesting corrections even if we can't change this particular game. Especially if it might lead to a more fun and robust Smash 5. For this reason I can't agree with sugar-coating our opinions for those hypothetical new players.

Nintendo is obviously paying close attention to smashboards

 

[PizzaWenisaur]

Yknow I thought I might've been being a **** but I just read a lot of this again and I still think most of this thread is silly.

Carry on with the discussion, I'm gonna bow out gracefully. But any new players heed my words. Rolling, even if your characters roll is OP, is a dangerous game. And the more you get away with it the harder the habit will be to break when you need to

That being said plz roll on me all day

Bye


Nintendo is obviously paying close attention to smashboards

Lol - ok then, PM Smash 4 Edition

Just joking, but seriously they might...even if they don't I find these discussions interesting.

As for the rolls, even though I think they are a little too strong. I'll consider that it could be the effects of me being a scrub and/or wireless lag. I hope you guys are right when you say it won't be that bad in Wii U version with a better controller and less lag, I just don't know if I have that much faith.

 

[Signia]

@ Amazing Ampharos This thread is in part a criticism of game mechanics. Of course we're going to consider alternatives. By showing there's a better way, we can decide that the current way is not good. And by closely examining them, we can better judge them. All that has nothing to do with whether they can be changed. It's about whether we should approve of the game's design. It can also point to what about the game we should be focusing on, and consider best ways of adapting to mechanics that are causing difficulty. Right now it looks like we should just avoid playing low tier characters that struggle in this game's environment.

 

[JamietheAuraUser]

To @ Prawn
I'm going to make the points clearer - not that I agree or support them fully and think they are true, but because it is the correct hypothetical view being presented in this thread.

The point being made is that, if you cannot use a kill move to punish a roll, then theoretically you can roll to avoid having a kill move landed on you. If this was a stamina/hp game, then you could whittle down someone's HP and win by using jabs or throws to punish rolls. But in Smash 4, those do not lead to kill moves on many characters. Thus, top level players who are using a character that cannot punish rolls via a kill move will be unable to land kill moves against skilled top-level players who know how to roll to limit their opponent's options.

This will likely be a large factor in tiers a year or two from now - if you can roll dodge intelligently to literally and mathematically prevent a character from being able to land a kill move, then that will make those characters less competitive as they can be countered by a strategy.

It's the same as how characters who can be chaingrabbed uniquely - like Fox and Ness in Brawl - end up being low in their tiers because of their mathematical shortcomings. This will be the same case, if things shape up the way we are hypothesizing.

Thus, only once the meta develops will we see if every character can, in fact, mathematically punish anyone's roll with a kill move. If not, then it will matter for that characters' matchups. The only way this will not be the case is if every character can punish rolls with a kill move (which is likely, but also not confirmed yet).

Yes, but can you punish a roll with a launcher and follow up? If yes, you can land a KO move off of a roll punish. It might take a few more reads than you'd like when compared to, for example, the Raptor Boost, but it seems to be a manageable tactic from this casual's point of view.

 

[Conda]

Yes, but can you punish a roll with a launcher and follow up? If yes, you can land a KO move off of a roll punish. It might take a few more reads than you'd like when compared to, for example, the Raptor Boost, but it seems to be a manageable tactic from this casual's point of view.

The point, if you need to depend on a tilt+followup (not everyone can combo to kill moves, by far), then that's proving that rolls are much more powerful and usable at high-level play to avoid kill moves. Because then it depends on read skill and chain-escaping to dodge the followup, which is more work than you'd have to do as a character you can punish rolls with a kill move. Which means it affects matchups greatly.

I understand your point, but you're reducing it to "rolls aren't a big deal, then, if there's some way to punish them", which is reductive. What matters is the applicability and requirement of the punish - if the effective punish for some characters/matchups depends on a tilt chained to a potential kill move via reads and mindgames - then it's supporting the "rolls are powerful and will affect the meta differently in smash 4" hypothesis. Rolling may be essential to how some matchups are made more polarizing.


Hypothetical example: If Kirby has no way - mathematically - to punish skilled greninja roll chains with a kill move, then he has to resort to punishing with a non-kill move. If that non-kill move does not chain/combo into a kill move, then he will end up racking up % but, if Greninja continues to utilize his rolls (which provide him a huge advantage in his matchup vs Kirby), Greninja can still avoid kill moves being landed on him. If Kirby's non-kill punish can chain loosely into a kill move, then a skilled Greninja will be able to escape it and sometimes he will fumble and be read - which leads to a kill for Kirby. But that is the most difficult way for anybody to get kills in Smash - if that's the only dependable way to get kills in some matchups, due to certain character rolls offering an advantage versus some movesets, then that matters.


We've yet to see if is at that point as we haven't crunched the numbers, but that's what matchups entail. Any advantage and disadvantage is weighed. In Smash 4, rolls are a huge advantage for some characters, and other characters may not be able to deal with them as well as others. Which will affect matchups and the metagame. This is unavoidable, and the most logical hypothesis currently.

 

[JamietheAuraUser]

The point, if you need to depend on a tilt+followup (not everyone can combo to kill moves, by far), then that's proving that rolls are much more powerful and usable at high-level play to avoid kill moves. Because then it depends on read skill and chain-escaping to dodge the followup, which is more work than you'd have to do as a character you can punish rolls with a kill move. Which means it affects matchups greatly.

I understand your point, but you're reducing it to "rolls aren't a big deal, then, if there's some way to punish them", which is reductive. What matters is the applicability and requirement of the punish - if the effective punish depends on a tilt, chained to a potential kill move via reads - then it's supporting the "rolls are powerful and will affect the meta differently in smash 4" hypothesis.

I do agree that the apparent general inability to punish an empty roll with a kill move is a drastic change from Brawl, and that they will most likely greatly affect the meta as a result, as compared to their relatively rare use in Brawl competitive. I just think that the affect on characters' viability won't be as drastic as you suggest. Of course, all of this is still mostly conjecture, but landing lag on an air dodge really does make multiple-read roll punishes a legitimate thing.

 

[Conda]

I do agree that the apparent general inability to punish an empty roll with a kill move is a drastic change from Brawl, and that they will most likely greatly affect the meta as a result, as compared to their relatively rare use in Brawl competitive. I just think that the affect on characters' viability won't be as drastic as you suggest. Of course, all of this is still mostly conjecture, but landing lag on an air dodge really does make multiple-read roll punishes a legitimate thing.

It's not possible to say a character's viability won't be affected, until we discover what matchups are affected.

Hypothetical:
If Kirby ends up being low tier, then a high-tier Marth having a weakness vs Kirby due to his rolls won't matter too much.
But if Rosalina is high tier and has awesome rolls, and mid-tier Mario has trouble punishing Rosalina's rolls, then Mario will be much less viable as a result.

Having a bad matchup versus MK or Falco in brawl meant way more than having a bad matchup versus Mario or Ness.


It's not about viability, it's about matchups. Dedede could be S tier, but if we later on discover that a ZSS that uses rolls more strategically can avoid being punished by Dedede's kill-moves - then that's a big deal.

Normally in Smash, rolls wouldn't come into much consideration when matchup difficulty/ease is discussed. But in Smash 4, rolls seem to have a larger strategic effect during a match, and are a much more powerful tool. Rolls could be a big element in Smash 4 that helps or hurts some matchups.

 

[Tagxy]

Conda youre being more reasonable then the rest of this thread, lol. But your posts should serve as an example of the right way to talk about rollings potential improvement.

That being said everyone ignored my previous frame test . It seems like rolls are about the same as they were in melee/brawl, and peeps are getting into their own heads (like they did with falcos roll in Brawl).

@ Signia Obviously in any game low tiers dont succeed in the games environment.

If the point was whether or not you should approve of the game's design then this thread should be locked and closed.
Competitive Discussion is not about stating personal preferences for game mechanics leave that for your personal blog. You might as well be telling us your favorite color. I gave OP the benefit of the doubt and assumed he thought this would be degenerative when he was talking about whether it was good or bad in terms of design choice, even if it seems far from degenerative.

Late edit: Heres Frame data on the standard rolls from each smash game + PM
64: 4-20/32
Melee: 4-19/31
Brawl: 4-19/31
PM: 4-19/33
*Smash 4: ?-?/ 29-31

*Based on the test done on the previous page. Invincibility was difficult to determine, the length was roughly between 29-31 for sheiks back roll.

 

[SamSun]

If any of you read the OP's following posts, or even the original post, his intent should be obvious.

 

[Amazing Ampharos]

@ Amazing Ampharos This thread is in part a criticism of game mechanics. Of course we're going to consider alternatives. By showing there's a better way, we can decide that the current way is not good. And by closely examining them, we can better judge them. All that has nothing to do with whether they can be changed. It's about whether we should approve of the game's design. It can also point to what about the game we should be focusing on, and consider best ways of adapting to mechanics that are causing difficulty. Right now it looks like we should just avoid playing low tier characters that struggle in this game's environment.

My point is that we don't gain anything by criticizing the game mechanics. It's not necessary to decide if the game mechanics could be "better" (whatever that means) to decide how to optimally use what the game mechanics are. I do agree we should understand how rolling will be useful in the metagame and how it will affect the tactics and characters we use in the future. We aren't doing any of that here.

For instance, you say we should avoid the low tier characters who struggle in this game's environment. Who are those characters, and specifically how are they unable to deal with rolls or utilize them effectively? From my perspective, Charizard seems pretty low tier all around, but he's the very best character at punishing rolls (Dragon Rush is such an easy and heavy punish, and even the default Flare Blitz, while a limited move in general, is pretty much specifically made to do huge damage to roll away). Yoshi seems like one of the best characters, and his roll seems slower than most of the others. I've observed no correlation between apparent character quality and their ability to play the roll game. If anything I've said is untrue, we're not doing anything to discover a different truth in this thread. I have not seen any discussion of which characters roll particularly well or which characters struggle to punish roll, of what kinds of situations rolling is especially powerful, or how a roll spammer can effectively deal with common anti-roll tactics (like just dsmash rolls in and dash attack rolls away). If we were talking very specifically about gameplay dynamics and some of the users happened to find them distasteful while providing a thorough analysis of the gameplay or were at least making very specific gameplay claims (not things like "rolls are too good", things like "if Little Mac rolls away in close quarters neutral, King Dedede is seldom able to punish" that we could meaningfully discuss), I wouldn't mind. Instead the majority of the thread does not seem to be reaching any conclusions about how to play this game better, and we really are projecting to everyone who is not already a part of this community when we make a lot of threads like this and they're the highest traffic ones that we aren't taking the game seriously.

To provide my $0.02 on dealing with rolls, here's the generic formula for beating them with any arbitrary character and a weak read. If you have a hard read or superior character specific options, punish heavier for sure.

-If they roll toward you, down smash unless you react really slowly or just have a trash dsmash in which case turn around ftilt or grab (or SH nair if you're like... Ness and maybe a few others).
-If they roll away, dash attack (or start throwing projectiles if you have good ones).

Rolls are useful when your opponent is already hitting buttons (and thus can't act freely to punish you) and you need to move in a way that the roll's invincibility will protect you in that movement. They're also situationally useful to position on the extreme edge of platforms when your opponent is really far away. As a pure neutral movement option they're pretty bad, and while I do strongly feel it's correct to roll far more often than "never", it is the case that once rolls are integrated into your game at all it is usually best to use them less and less instead of more and more. I have not seen any gameplay that convinces me the above formula does not still work.

 

[Signia]

@ Amazing Ampharos I already laid out who might struggle against rolls in an earlier post. I also limited the discussion to beating rolling away. Everyone should be able to punish someone who rolls and ends up near you.

Basically, most characters have either a fast dash, good dash attack, projectiles, or aerial mobility. Not everyone, though.

What characters neither have a fast dash, good dash attack, a spammable projectile, fast horizontal attack, or good air speed?

Problem characters seems to be G&W, Shulk, Luigi, Kirby, Ike, Lucario, Falco, Zelda, Ganondorf, Megaman (added characters people are having trouble punishing rolls with).

G&W is right in the middle on the dash speed rankings so he might have issues punishing the really fast rolls. Otherwise he's fine, and can probably even combo off of a late dash attack.

Shulk is in the same boat, but he has Speed Monado fast rollers.

Luigi at least has a projectile that he can zone with to discourage rolling, instead being able to punish it outright. Or go for a misfire, lol.

Kirby has a multi-hitting dash-attack which makes things easier if you want to predict a roll instead of punishing it on reaction.

Ike... Side-B? His dash and dash attack is too slow to punish with on reaction. I'm kinda stumped on this one. Maybe you could corner them by walking toward them and reacting to things, then and force a roll-toward by the edge, and get a big punish off a preemptive Fsmash, but his walk is very slow he requires a huge commitment to get anything but a jab combo.

Lucario? Idk, he's right below Ike on the dash speed rankings. But, he has a projectile. Use it.

Falco can side-B or laser, I guess? Not the greatest options, but maybe you have use them.

Zelda has good projectile and a multi-hitting dash attack. Stop complaining.

Ganondorf has side-B and down-B as risky hard reads. That's how the character is with everything.

Megaman is a projectile character... sorry, but you can't punish things on reaction. Zone. Throw projectiles to control space and discourage them from rolling away in the first place. Besides, he also has a multi-hitting dash attack and skid cancel dtilt.



Though a problem with a system like this is that you can't just pile on damage to eventually get a kill. You eventually have to connect with something that will finish them. In this game, it seems there is little to stop people from just rolling around everywhere, and that players can easily play in such a way to avoid KO moves while still being able to rack up damage themselves. The best you can do is corner them and patiently wait for them to do something risky, or force mixups during edgeguards after weak hits. But those plans can easily fall apart or even backfire in this chaotic game.

 

[DeLux]

I just want to clarify that if you're referencing the "frame data" from my thread (looking at you @ Signia ), I actually didn't test it and just put numbers down to illustrate a hypothetical way to explain the phenomenon the thread was documenting. I used the numbers that I thought I remembered from the frame data from I think Robin, so you could see how that might be lost in translation in terms of accuracy, but I figured it was okay to just throw out there since the point of the thread wasn't the frame data but the ramifications of having an IASA before autcancel windows.

I think this thread probably needs an objective analysis of not only the frame data on rolls, but probably the frame advantage on roll forced whiff, against the general damage/knockback of moves that exist within the parameters of the frame advantage.

As to whether we should determine if rolls are "problematic", by pretty much game theory definition we'd have to see if the game degenerates to rolling for a strictly dominant strategy. Given the roll has no hitbox and finite distance / frame data, intuitively there is a counter to them thereby making them unlikely to ever develop into a strictly dominant strategy.

I think people simply are operating under the paradigm of, "In Smash Games, Rolls are pretty Ass," and them being actually fairly powerful against that context could seem shocking. However, from an objective depth perception, I'd argue it probably deepens the game, since you have an option that isn't Ass. Since options don't exist in a vacuum that might obviously weaken other options, but the other options are necessary in order to win so they obviously have use within game theory structure.

That is of course, unless the frame data analysis suggests otherwise. Which I doubt is that case. But Sakurai makes some real head scratcher decisions sometimes (see Meta Knight's frame data) so I don't want to fully rule that out prior to said analysis.

But once we have that analysis, our role as the player is to abuse the strategy to win. Anything else doesn't really matter.

 

[Strong Badam]

My point is that we don't gain anything by criticizing the game mechanics. It's not necessary to decide if the game mechanics could be "better" (whatever that means) to decide how to optimally use what the game mechanics are. I do agree we should understand how rolling will be useful in the metagame and how it will affect the tactics and characters we use in the future. We aren't doing any of that here.

Your statement assumes that everyone is here to exclusively discuss how to optimally play the game. This is obviously not true, as there are threads here to describe newfound gameplay mechanics, rulesets, etc. etc.. It's easy to apply "We don't gain anything from criticizing X" to anything in which we don't have control over the creation (television, music, the film industry, etc.) but criticism keeps media creation honest. Although it's obvious Sakurai doesn't care what we want in a general sense, it certainly wouldn't help if we simply rolled over and accepted each game without question.

Make a new thread called "How to defeat roll spamming" if you'd like a thread dedicated exclusively to developing counterplay. It's perfectly acceptable for people to post in this thread criticizing the effectiveness of rolls, and what their potency means for the competitive metagame (positive or negative)

 

[Tagxy]

Your statement assumes that everyone is here to exclusively discuss how to optimally play the game. This is obviously not true, as there are threads here to describe newfound gameplay mechanics, rulesets, etc. etc.. It's easy to apply "We don't gain anything from criticizing X" to anything in which we don't have control over the creation (television, music, the film industry, etc.) but criticism keeps media creation honest. Although it's obvious Sakurai doesn't care what we want, it certainly wouldn't help if we simply rolled over and accepted each game without question.

Make a new thread called "How to defeat roll spamming?" if you'd like a thread dedicated exclusively to developing counterplay. It's perfectly acceptable for people to post in this thread criticizing the effectiveness of rolls, and what their potency means for the competitive metagame (positive or negative)

The point is this forum isnt a place for everyone to state their personal tastes and feelings on mechanics. Itd devolve into the poor CD's on other forums. Its about as helpful as talking about a girl you like at school, that stuff belongs in personal blogs. If the mechanic threatens to make the game degenerate thats something different.

Discussions in this forum should differentiate from the general boards. Posts should give positive and relevant contribution and threads should be presented well with a clear purpose and have a solid introduction to them.

 

[Strong Badam]

That description doesn't conflict with people posting their opinions on gameplay mechanics in a competitive context in any way.

 

[DeLux]

Your statement assumes that everyone is here to exclusively discuss how to optimally play the game. This is obviously not true, as there are threads here to describe newfound gameplay mechanics, rulesets, etc. etc.. It's easy to apply "We don't gain anything from criticizing X" to anything in which we don't have control over the creation (television, music, the film industry, etc.) but criticism keeps media creation honest. Although it's obvious Sakurai doesn't care what we want in a general sense, it certainly wouldn't help if we simply rolled over and accepted each game without question.

In terms of nomenclature, I'd assume the purpose of examining rolls in the competitive context in "competitive discussion" would be how to use them/beat them aka optimal play. As such, I figure presenting them as "problematic" ought to open discussion on how to beat them, how they aren't problematic, etc.

If the objective is to say if rolling is "good or bad" for the game, perhaps having this thread be in general discussion would be more appropriate?

Although I'm skeptical of someone having a logically backed coherent assessment of such, in the same I'm skeptical on my speculation that having more powerful rolls might deepen the game.

 

[Tagxy]

Id agree its better for general discussion, but as I said earlier I gave the OP benefit of the doubt and assumed he thought this was degenerate.

That description doesn't conflict with people posting their opinions on gameplay mechanics in a competitive context in any way.

Personal opinions on gameplay mechanics are not relevant contributions. Obviously its not a big deal if it happens but it shouldnt be the highlight of a post, the thread, or the typical posts in a thread.

 

[DeLux]

Even if it's degenerate, it's not like we can discretely ban rolling to my knowledge.

Which I think is where Ampharos is drawing his line.

 

[Tagxy]

Well if it really was degenerate then we'd definitely need to figure something out. IMO people toss around the word (or use of) degenerate too freely though.

 

[DeLux]

Well if it really was degenerate then we'd definitely need to figure something out. IMO people toss around the word (or use of) degenerate too freely though.

A couple of non-sequitors given the direction of this discussion:

https://twitter.com/ChaseTheLux/status/522970183522996224
https://twitter.com/ChaseTheLux/status/512511334244548608

 

[Tagxy]

Funny as that is, I do think its important to set a good example and keep threads and posts in this forum relevant. At least so other people dont get the wrong idea. I know the posts in this thread including these last few havent been the best example.

 

[DeLux]

This whole rolling thing reminds me of ROB in brawl having a dsmash with 19 frames of invincibility on startup. limited tangibility from frame 20-25, and a hitbox that came out on frame 29.

People learned to deal with it.

 

[JamietheAuraUser]

Hello, mods and other people! Rather than having a meta-discussion on whether this discussion should be in this board, why don't we get back on the topic of how to punish/optimally use rolls and how to land a KO move on a defensive opponent abusing rolls and sidesteps? (Note: No offence is intended to anyone. Just sort of making a point/noticing something seemingly ironic.)

This whole rolling thing reminds me of ROB in brawl having a dsmash with 19 frames of invincibility on startup. limited tangibility from frame 20-25, and a hitbox that came out on frame 29.

People learned to deal with it.

Wait, really? No wonder that move was so ridiculously abused by CPUs!. And no wonder my Smash Attacks (and jabs, and tilts...) seemed to so commonly whiff for absolutely no visible reason, only for me to eat a DSmash immediately afterwards.

 

[DeLux]

Outside of the generic, charge a smash attack or soft read and/or turn around and softer punish with grab/tilt/jab/etc that STILL works -

The removal of jab cancelling makes the defensive options more powerfuls.
The addition of dash > pivot tilts mitgates them.

I still think we need objective frame data analysis on the game as a whole to really paint an accurate picture.

 

[Tagxy]

Delux beat me to the punch. The biggest problem with finding solutions right now (as some pointed out earlier) is lack of frame data for rolls and attacks, etc.. Granted we have some obvious solutions that exist already, but it doesnt allow for a super deep discussion.

 

[JamietheAuraUser]

Outside of the generic, charge a smash attack or soft read and/or turn around and softer punish with grab/tilt/jab/etc that STILL works -

The removal of jab cancelling makes the defensive options more powerfuls.
The addition of dash > pivot tilts mitgates them.

I still think we need objective frame data analysis on the game as a whole to really paint an accurate picture.

I think a few characters still have jab cancelling. Mii Swordfighter definitely seems to.

But yes, lack of frame data is a definite problem.

 

[DeLux]

I think a few characters still have jab cancelling. Mii Swordfighter definitely seems to.

But yes, lack of frame data is a definite problem.

The more you know! Mii Swordfighter top tier. I know so little about this game it's amazing.

[collapse="Food for thought"]It makes me so happy. To be at the beginning again, knowing almost nothing. People were talking about the end of physics. Relativity and quantum mechanics looked as if they were going to clean out the whole problem between them. A theory of everything. But they only explained the very big and the very small. The universe, the elementary particles. The ordinary-sized stuff which is our lives, the things people write poetry about - clouds - daffodils - waterfalls - ... these things are full of mystery, as mysterious to us as the heavens were to the Greeks. Because the problem turns out to be different. We can't even predict the next drop from a dripping tap when it gets irregular. Each drip sets up the conditions for the next, the smallest variation blows prediction apart, and the weather in unpredictable the same way, will always be unpredictable. When you push the numbers through the computer you can see it on the screen. The future is disorder. A door like this has cracked open five or six times since we got up on our hind legs. It's the best possible time to be alive, when almost everything you thought you knew is wrong. - Arcadia, Tom Stoppard [/collapse]

 

[JamietheAuraUser]

The more you know! Mii Swordfighter top tier

Maybe the Swordfighter would be if it weren't for the fact that you basically have to land everything or get owned. On the ground, it's sort of like Marth except instead of spacing perfectly for extra damage, you use perfect spacing in order to not get stuffed or punished hard. Oh, and you actually have a usable jab combo coupled with the ability to string any tilt of your choosing out of jab 1 (string, not combo). (Although, Forward Smash does deal more damage the closer you get to the tip of the blade, sort of like Link's.)

I thoroughly expect the Swordfighter to be much more "viable" on Wii U, where players will have the ability to easily distinguish between tilt and Smash inputs rather than screwing up and getting punished. Even the New 3DS won't solve the problem completely as you still might accidentally FSmash when you wanted an FTilt, or FTilt when you try for a stutter-step FSmash.

 

[theunabletable]

Alright bit of a post incoming.

So lots of people were looking for some frame data, so I decided to try and collect some. Looked at Little Mac's roll because his appears to be the fastest, so if it could be easily dealt with, then no roll would be a "problem" (the word problem being left intentionally vague.)

I used the in-game timer in a match vs a CPU, and pause buffering using the home button. As I got more comfortable at it, the results got more consistent, and that's visible in the small amount of data I gave. All my attempts that went visibly poorly I discarded.

https://docs.google.com/spreadsheets/d/1r7lFnEFHDGNVDkbdgiozghuniRJv2G3LAmbncddupQA/edit?usp=sharing

The first big chunk is forward roll, the last chunk is back roll. I separated each into 4 parts. I recorded the time at which the roll started, when the invincibility became visible*. when the invincibility stopped blinking, and when the sound of the shield occurred. All those times are in in-game time, so it doesn't perfectly translate to frames, which I believe is how the game deals with everything. 100 units of in game time per 60 frames, so every number, multiply by 60, and divide by 100 for the speed in frames.

Quick observations:
Forward roll and back roll for Little Mac appear to have the exact same frame data, which is interesting but feels about right. A-D for the length of the roll in seconds, multiplied by 60/100 and you get it in frames. I stopped gathering data when each attempt was predicted, but obviously the sample size is small, and I could be off by a tad.

So it appears that both rolls last (40/100)*60=24 frames, which feels about right. Invincibility starts on what seemed to be frame 3, but it may have been frame 4, and lasts until (10/100)*60=6 frames before the move ends. So invincibility on frames 3/4-18.

It appears to be the same speed, or 1 frame slower than MK's forward roll in Brawl, which is surprising to me, but it has twice as much invincibility, which may account for the difference in "feel". (EDIT: Actually I misinterpreted it. If the shield comes out on the 24th frame, then the roll lasts 23 frames, which would make it the speed of MK's forward roll)

*I'm not sure how the game translates the invincibility visually. I figured it was generous to record the exact frame that the character flashes, and then to record the exact frame after the last "blink" ends, the moment there's no more white. There's a chance that there is invincibility on the frame before the white flash appears, or on the frame after it disappears, but it's hard to test, and so I figure the safer/more charitable bet is to estimate down.

EDIT: Summary:
Startup frames 1-2
Invincibilty frames 3-18
Cooldown frames 19-23

Appears to be the same for both back and forward roll.


Also one other tangential observation. Shields may come out in more than one frame in this game? The sound of the shield was instant, 1 frame. It would make the sound of the shield coming out on the frame after it was inputted, but with no change in animation. Then there'd be the animation of Little Mac blocking with his arms (but no shield bubble at all,), and then on the third frame, the shield bubble would appear.

It's hard to test when you're actually hittable without another person, but I could probably use some item later in training room to figure it out. It isn't obvious to me that the frame where the shield sound first happens is one in which your shield is actually active. Maybe other people should take a look, your character doesn't make any animation changes by that point.



Anyways, that's the frame data I found. My reaction time hovers between 200 and 240 ms or so, or 12-15 frames, without any noticeable delay from hardware or software, and on a simple click test taken from here: http://www.humanbenchmark.com/tests/reactiontime

Suppose I'm acting on the slow end, around 15 frames. I have about 7-8 frames of wiggle room. Anything slower than 7-8 frames will hit a shield. I don't know for sure, but it seems reasonable to guess that grabs are still 6 frames max in this game (and slower with a dash grab.) As well, with grabbing, I believe you'd get an extra frame of error because buffered spot dodge gets invincibility one (perhaps two in this game, but if Brawl is any indication) frame slower than shield.

Given that stuff, it would look like jabbing or grabbing an opponent who rolls to the spot you're facing is within human reaction time, although it's tight, and probably harder in practice than in theory, especially during a game where you have to think about anything other than punishing rolls. It's not clear that there's reasonable time to turn around grab, or turn around jab. The visual cue for punishing should be to hit your punish the instant you see the roll, because your reaction time+move startup will be slower than the time it takes for his invincibility to fully run out.

Take all that how you will.

Gonna make another post as a response to some people in a minute.

 

[theunabletable]

If the double post is a problem, I'll feel free to delete and edit this into my last post. This post was on such a different type of discussion, though, that formatting into two separate posts seemed much more useful, both as a writer, and for the readers.

As to whether we should determine if rolls are "problematic", by pretty much game theory definition we'd have to see if the game degenerates to rolling for a strictly dominant strategy. Given the roll has no hitbox and finite distance / frame data, intuitively there is a counter to them thereby making them unlikely to ever develop into a strictly dominant strategy.

I think people simply are operating under the paradigm of, "In Smash Games, Rolls are pretty ***," and them being actually fairly powerful against that context could seem shocking. However, from an objective depth perception, I'd argue it probably deepens the game, since you have an option that isn't ***. Since options don't exist in a vacuum that might obviously weaken other options, but the other options are necessary in order to win so they obviously have use within game theory structure.

I don't believe it's entirely reasonable to consider a mechanic "problematic" if, and only if, the entire metagame devolves to everybody using that mechanic.

There is an inherent degree of subjectivity in "problematic", and there's nothing wrong with that, either. We can comment on the competitive aesthetics of a game without making broad generalizations about the entire game, or whether it's "worth" playing.. Of course any game can be played, and whatever it takes to win is, roughly, the capacity for skill within that game.

We don't just discuss strategy broadly and flaccidly as if we're a bunch of robots who will be playing the game for eternity where the metagame will always develop. If there's any lesson to learn from Brawl, there's not much to discuss about competitive play if nobody plays the game. The "competitive aesthetics" of the game are a relevant part of the game, just as much as the "square" theoretical side.

I was very, very careful with my choice of words. I said "problematic" because I did NOT want to be interpreted as making any prediction about the progression of the metagame, nor about whether rolling is "unfair" or "broken". If rolling is strong, the metagame will adapt, and it'll just be part of the game (and probably not the only part of the game, as you noted.) I wasn't intending to, say, put rolling on the same level as planking in Brawl.

When I say "problematic", I'm not referring to anything even close to the game-theory based definition you brought up, and so dragging the discussion back in that direction avoids the points I'm trying to explore.

My point is that we don't gain anything by criticizing the game mechanics. It's not necessary to decide if the game mechanics could be "better" (whatever that means) to decide how to optimally use what the game mechanics are. I do agree we should understand how rolling will be useful in the metagame and how it will affect the tactics and characters we use in the future. We aren't doing any of that here.

We don't have discussions purely for the purpose of advancing the metagame, even in avenues such as these. As if everyone who has discussions about the tier list is attempting to add to some "great utility", or the entire purpose of a power ranking is to provide some "useful thing".

We can have discussions about things which don't have a primary use, but are interesting in and of themselves, and that's what this is. Not every discussion has to have a normative, ought-based background.


There is an underlying criticism of the game's mechanics here, but it isn't the primary focus at all. In a discussion like this, you'd hardly be motivated to try and discuss aspects of a buffed or nerfed mechanic unless you really liked the change, or you really disliked it. It's a discussion about some type of subjective, but only descriptive, quality associated with rolling, with normative underpinnings based on the result of an accurate description. There's a bit of a distinction.

 

[Judo777]

To @ Prawn
I'm going to make the points clearer - not that I agree or support them fully and think they are true, but because it is the correct hypothetical view being presented in this thread.

The point being made is that, if you cannot use a kill move to punish a roll, then theoretically you can roll to avoid having a kill move landed on you. If this was a stamina/hp game, then you could whittle down someone's HP and win by using jabs or throws to punish rolls. But in Smash 4, those do not lead to kill moves on many characters. Thus, top level players who are using a character that cannot punish rolls via a kill move will be unable to land kill moves against skilled top-level players who know how to roll to limit their opponent's options.

This will likely be a large factor in tiers a year or two from now - if you can roll dodge intelligently to literally and mathematically prevent a character from being able to land a kill move, then that will make those characters less competitive as they can be countered by a strategy.

It's the same as how characters who can be chaingrabbed uniquely - like Fox and Ness in Brawl - end up being low in their tiers because of their mathematical shortcomings. This will be the same case, if things shape up the way we are hypothesizing.

Thus, only once the meta develops will we see if every character can, in fact, mathematically punish anyone's roll with a kill move. If not, then it will matter for that characters' matchups. The only way this will not be the case is if every character can punish rolls with a kill move (which is likely, but also not confirmed yet).

You CAN land kill moves on rolling opponents. As I'm sure has been stated, rolling is worse than walking because it is a commitment. Once you roll you forfeit control of your character until the roll ends. You may have to read the roll to punish it with a kill move (honestly how it should be) but if a roll is read many characters are guaranteed a kill move. If you read a walk, you are guaranteed nothing. I personally don't think you should be able to react to an option from a neutral position with a kill move, that's what IC's do in brawl and its BS.

 

[JamietheAuraUser]

You CAN land kill moves on rolling opponents. As I'm sure has been stated, rolling is worse than walking because it is a commitment. Once you roll you forfeit control of your character until the roll ends. You may have to read the roll to punish it with a kill move (honestly how it should be) but if a roll is read many characters are guaranteed a kill move. If you read a walk, you are guaranteed nothing. I personally don't think you should be able to react to an option from a neutral position with a kill move, that's what IC's do in brawl and its BS.

Actually, what the ICs do in Brawl that's so completely, utterly ridiculous is react to an option from a neutral position with an instant kill combo from any %. It's sort of like taking Street Fighter 4 and adding a new character with a grab that does enough damage to OHKO anyone in the cast.

 

[Conda]

You CAN land kill moves on rolling opponents. As I'm sure has been stated, rolling is worse than walking because it is a commitment. Once you roll you forfeit control of your character until the roll ends. You may have to read the roll to punish it with a kill move (honestly how it should be) but if a roll is read many characters are guaranteed a kill move. If you read a walk, you are guaranteed nothing. I personally don't think you should be able to react to an option from a neutral position with a kill move, that's what IC's do in brawl and its BS.

I'm speaking of all characters vs all other characters. If there are some matchups where (at a high-level once the meta has developed) rolling is an effective way to deny a certain character a kill-move opportunity the majority of the time, then it's a strategy that will be viable and used.
We don't know if that's the case yet, but if it were, then it'd be a big deal.All we know is rolls got a buff, and we have to accept that maybe down the road we'll see certain characters' roll speed/characteristics be a countering factor against some characters' roll-punishing abilities.

If it doesn't happen, then it doesn't. No biggie. I can't say what'll happen, but I recognize the potential that some things maybe could this time around.

 

[Judo777]

I'm speaking of all characters vs all other characters. If there are some matchups where (at a high-level once the meta has developed) rolling is an effective way to deny a certain character a kill-move opportunity the majority of the time, then it's a strategy that will be viable and used.
We don't know if that's the case yet, but if it were, then it'd be a big deal.All we know is rolls got a buff, and we have to accept that maybe down the road we'll see certain characters' roll speed/characteristics be a countering factor against some characters' roll-punishing abilities.

If it doesn't happen, then it doesn't. No biggie. I can't say what'll happen, but I recognize the potential that some things maybe could this time around.

That's my point tho, it doesn't deny characters a kill move any more than shielding does. It's literally the same thing as shielding except shielding has no timing required for the user and has no commitment. The only advantage rolling has to shielding is that rolling can't be grabbed unless you read where they are going. But so far that doesn't seem to be a big problem.

And even if you can't land a kill move on a rolling opponent, you can definitely land some kind of move on them, which can force them into the air or offstage, then you can DEFINITELY land a kill move after that.

@Jamie yes I understand, but its still a kill move from a neutral position. Very few characters have that option in smash (fox is one that does, and everyone know his usmash is ridiculous in all previous versions of the game)

 

[PizzaWenisaur]

I feel like perhaps the best course of action is for us to play matches with each other, in which we see how much we can abuse rolls. I feel like that's more productive than walls of frame data.

Since @ Amazing Ampharos wants the discussion to be a little more specific than "Rolls seem OP," I'll say specific examples of why I think rolls are a bit too strong.

I play as Mario and Shulk ( so my opinion might be jaded with that in mind ). If someone rolls towards me or if they dodge in place, that's when rolls don't seem that bad. Most characters, as far as I've seen, can handle this. And can do decent punishes if you properly predict it. With both Shulk and Mario I can generally go for a grab, a tilt, or a Down-Smash.
* One time someone rolled into Shulk's Down-Smash and managed to roll out of it even though they should've been where my hit-box was.
* It's difficult to punish someone rolling towards me with a F-Smash ( which is stronger than a Down-Smash ) as either Mario or Shulk due to the lag of the move. Plus if I mis-time it they will punish me instead.

Now, when someone rolls away that is when rolling seems too strong.
Shulk, because of his dash attack speed and his defult run speed, is too slow to punish this. All of his aerials are pretty slow too. If he goes into Monado-Run he could probably get a grab. But Shulk doesn't have grabs that lead to combos or kills.
Mario could probably get a dash attack or a grab (perhaps). But neither leads to something significant. And the fact that this game doesn't have momentum conservation means that, in general, it's a lot harder to punish rolls with aerials.

In these cases, against someone rolling away from me the best I can hope for is a small punish (meaning small percent and no kill). Because shielding is so strong and because a lot of moves ( particularly Mario's dash attack ) are unsafe on shield, at worst I get punished.

Here's my point. Whoever said that any punish was sufficient ( or whatever ) is wrong. In this game more than others, it feels like a stock lead is a huge lead. I believe that is because once someone obtains a lead, most people logically go into that "deal as much damage as possible" mentality. That means they aren't playing to get a kill - just to tack on a little damage. So they start to play a little more conservatively. Now consider what I said before about only being able to challenge those rolls with a small punish. Because this game has a bigger divide between kill moves and non-kill moves, this scenerio intrinsically favors abusing back-rolls for the person in the lead.

Think about it, this game doesn't have momentum conservation or L-cancelling. Rolls have less punish frames. Moves have less kill potential. In previous games you could challenge a back-roll with an aerial that would either 1) straight-up kill them at high percentages or 2) lead to a combo that would kill them. Or even sometimes go for a grab that would kill, or set up for an edgeguard ( another thing that is less effective). I can't see a reason why someone wouldn't just use extremely safe moves followed by rolling away from the opponent. Especially if they are in the lead, seeing as this would most likely make the game devolve into a rock-'em-sock-'em robots type of gameplay. And since less moves kill, this would more heavily favor the person in the lead than in past games.

*Note that when I say back-roll I just mean someone rolling away from you. Not an actual back-roll.

I believe that we've experienced this also - an unnaturally snowball of the match after the first stock was taken...

 

[DeLux]

If the double post is a problem, I'll feel free to delete and edit this into my last post. This post was on such a different type of discussion, though, that formatting into two separate posts seemed much more useful, both as a writer, and for the readers.



I don't believe it's entirely reasonable to consider a mechanic "problematic" if, and only if, the entire metagame devolves to everybody using that mechanic.

There is an inherent degree of subjectivity in "problematic", and there's nothing wrong with that, either. We can comment on the competitive aesthetics of a game without making broad generalizations about the entire game, or whether it's "worth" playing.. Of course any game can be played, and whatever it takes to win is, roughly, the capacity for skill within that game.

We don't just discuss strategy broadly and flaccidly as if we're a bunch of robots who will be playing the game for eternity where the metagame will always develop. If there's any lesson to learn from Brawl, there's not much to discuss about competitive play if nobody plays the game. The "competitive aesthetics" of the game are a relevant part of the game, just as much as the "square" theoretical side.

I was very, very careful with my choice of words. I said "problematic" because I did NOT want to be interpreted as making any prediction about the progression of the metagame, nor about whether rolling is "unfair" or "broken". If rolling is strong, the metagame will adapt, and it'll just be part of the game (and probably not the only part of the game, as you noted.) I wasn't intending to, say, put rolling on the same level as planking in Brawl.

When I say "problematic", I'm not referring to anything even close to the game-theory based definition you brought up, and so dragging the discussion back in that direction avoids the points I'm trying to explore.

So you're talking about a situation where if they roll away from you and you don't react to it by overcommitting, they can't punish you because they gave you space. Further if you do react to it, you are able to tack on some finite punishment depending on the character you're using.

So the problem here is overcommitting as opposed to rolling away being "too powerful" from a risk/reward perspective, let's be real. Overcommitting is a player decision problem, not a game design problem. In theory, the heavier you decide and leave yourself open to being punished by something other than reverse roll, the heavier punish you'd be able to net. For example, if you waited to soft confirm the roll, you might be able to dash attack. If you went on a full hard read and ran preemptively expecting a roll, you'd net an usmash (aka a kill move), but you'd leave yourself open to getting hit preemptively for an aggressive dash. It's actually a pretty solid yomi-loop.

That doesn't even begin to get into the meta-theory of "if rolls are powerful, they are more likely to roll in that situation", therefore making "if they are more likely to roll, then I should more likely be able to make an aggressive read expecting that roll". Which leads into, "if they are more likely to make an aggressive movement action expecting a retreating roll, maybe I ought to try to stuff them with a prememptive option", etc.

Spoiler