Is the time to start serious tier list discussion soon?

[Scarlet Jile]

Wtf?
Bowser J.r that high?
Bowser that low?
DDD after Ganon (0_-)
HuH??

It's time to face it, man. Despite everything, Bowser remains a mediocre character.

The upshot is, there is no E or F tier as far as I'm concerned. Everyone is viable, some are just more viable than others.

 

[clydeaker]

...the hell it's just the players. Metaknight didn't move NOWHERE.

many people considered meta knight and snake to be fairly even during the first tier list. it wasn't until Mew2king showed how metaknight could be played that the gap started to grow between metaknight and everyone else.

 

[DarkBlueSpark]

http://smashboards.com/threads/swf-...version-1-0-finished-time-for-a-break.379736/

Thought this was our temporary tier list?

 

[Sinister Slush]

The best you can do is a league of legends like feel one as long as the patches keep rolling out.

That really is the only comparison I got for a tier list of new age games.

And even then, nerfed top tiers remain up there for a good while until they overly buff champs or buff/create items.

In Smash terms, there can only be nerfs/buffs to characters unless they keep touching custom moves too for the "items" part.

 

[zpxociv]

many people considered meta knight and snake to be fairly even during the first tier list. it wasn't until Mew2king showed how metaknight could be played that the gap started to grow between metaknight and everyone else.

MK is proof that players don't change anything if something's too broken because there's no way to metagame your way around that.

 

[WritersBlah]

There were plans for post-apex that were basically snuffed out by the balance patch. If there is a balance patch for Lucas we're kinda paddling up **** creek.
An alternative could be iterative lists that are just "feels" rather than official (and are based off older/previous patch rather than the current live one), but it's hard to do things under that paradigm, most people I've personally brought up this idea too seem skeptical/disinterested.

Darn, that totally slipped my mind. But that's a really good point; creating a tier list for a game that gets constant balance patches is almost gonna seem worthless, unless Sakurai's team officially announces there will not be any more balance patches for the game...which I don't see happening for another four or five years.

 

[Wintermelon43]

Whoever creates the tier list should start making a tier list as soon as EVO Ends. Maybe more if there's another big tournament close after that.

 

[DarkBlueSpark]

Again, no one seems to be talking about this at all http://smashboards.com/threads/swf-...version-1-0-finished-time-for-a-break.379736/

Does that not count as a temporary tier list? Sure, it doesn't have Mewtwo in it yet, but more research still needs to be done, plus there's a healthy amount of discussion in that thread. So "serious" discussion started awhile ago. Unless you guys mean something different and I misunderstood?

 

[warionumbah2]

That's a popularity list not a tier list. You could vote down Sonic for being too blue.

There's no reason to talk about it at all.

 

[DarkBlueSpark]

That's a popularity list not a tier list. You could vote down Sonic for being too blue.

There's no reason to talk about it at all.

Except I really doubt a large number of people would have done something like that. If anyone did something like that, that'd be a really small minority. Plus, it's at least a place to start and it seems at least somewhat accurate, at least in the top tiers, since most people seem to agree on who's above average.

 

[clydeaker]

That's a popularity list not a tier list. You could vote down Sonic for being too blue.

There's no reason to talk about it at all.

Except I really doubt a large number of people would have done something like that. If anyone did something like that, that'd be a really small minority. Plus, it's at least a place to start and it seems at least somewhat accurate, at least in the top tiers, since most people seem to agree on who's above average.

This is the same argument people have about the event hub tier list: http://www.eventhubs.com/tiers/ssb4/.
and that tier list allows you to vote on the exact match ups instead of the tier list itself. because of that I see it as a better tool that the smashboards community voted tiers list. The thing is the if we did organize a back room it would still be a vote, the vote would just be based more on tournament results and the people who vote would supposedly be more educated on the subject of competitive smash.

 

[DarkBlueSpark]

This is the same argument people have about the event hub tier list: http://www.eventhubs.com/tiers/ssb4/.
and that tier list allows you to vote on the exact match ups instead of the tier list itself. because of that I see it as a better tool that the smashboards community voted tiers list. The thing is the if we did organize a back room it would still be a vote, the vote would just be based more on tournament results and the people who vote would supposedly be more educated on the subject of competitive smash.

That's true. I'm not against the idea of a backroom, in fact, I support it, I'm just saying that the community did already make a temporary tier list so we could start there. The title of the thread just gave me the impression that no discussion about a tier list has happened yet, when that's not true.

 

[LightLV]

"As opposed to relying on what?" You cannot be serious. Let me ask you something... Why did you choose Link and Shulk as your mains over maining Little Mac, Sheik, Rosalina or Diddy (Pre-Patch)? At that time, they will give you the best mileage over Link and Shulk. Heck, Shulk was "low enough" to get a buff when the Wii U came out.
Do you remember in the Demo when everyone thought Megaman/Pikachu were terrible?

I've probably played Shulk a grand total of 6 times since this game launched. I have no clue why I put those as my mains.

There's a difference between playing who you want to play, and playing who you think has the best chance of winning. Tier lists are useful in that regard.

There are precious few, if any, facts to do with Smash Bros tier lists. Yes, they can observe data and extrapolate tiers based on tournament results, but that's a results-table, not a tier list.

The absolute best we can hope for in a backroom tier list is the same as any other community tier list: that they are as educated and informed as possible in their guesses.

Results table = Tier list. It's like that for Smash 4, just like any other fighter, and just like every smash iteration before it, and they are accurate. They are always accurate at their job BECAUSE they are results lists. This is why I cannot understand why the smash bros community is so commonly torn when people mention Tiers. Play whoever you want, get as good as you want, but tier lists are not born of opinions. You may be able to excel in For Glory or whatever, but in the competitive scene, characters simply are not created equal. Sure, the best Zelda player in the world COULD beat the best Mario in the world, but on paper, but it would take a massive amount of human error to happen, which isn't a very "competitive" way to approach winning.

It does NOT take a year of gameplay for this to happen, it only takes enough collective data, and people have been playing the hell out of this game. Making customs legal only expanded the amount of options, but it's not going to change anything. The most effective customs will be weeded out relatively quickly, because competitive players play to win, and will start winning with them.

Someone may find some glitch, exploit or "advanced technique" or whatever that allows some previously overlooked character to become competitive, but that is not going to knock any of the current top-tiers down or move anyone near them up. Samus and Zelda aren't shooting up any spots without massive buffs. No amount of playing Lucina will ever make her a better character than Marth. People just gotta accept things sometimes.

 

[san.]

Tier lists represents top level play, but most top level players have their own ideas about MUs and often times conflict. Tier lists won't even be very accurate for even your average competitive player, since they aren't top level. From my experience in Brawl, general MU mechanics sometimes vastly differed from top level options and mixups, mostly because of power shielding and SDI.

Smash 4 seems to be simpler in that regard without strong SDI and the like screwing with things, but I still think that line of thinking somewhat holds. Most people already have an idea who's good, who's limited, and who's unexplored, so anything extra seems superfluous.

 

[COLINBG]

Ok, so tiers can't be made by just comparing raw move statistics? I think they can because all it really boils down to is mathematics. Everything is numbers and the only objective way to determine tiers is to define what the human limits are and that way you can sort out the potential actual human performance of each character. It could be as simple as TAS testing using rules that simulate human controller input and reaction limits to determine the outcome. Why not try the direct experimental approach? Just try it out with the two suspected best characters as a smaller-scale experiment.

TAS testing wouldn't work. Imagine a chess game where two perfect computers play against each other. Every single game would end up in a draw. It's kind of the same thing here. You know these times when two smash players take dozens of seconds to actually get something going from the neutral, because no one want to try an approach? You would just get that everytime.

Smash relies a lot on mindgames. If you get hit, it's because you made an error, as simple as that. It means you either ****ed up yourself by making an unnecessary/dangerous maneuver, or the opponent outplayed/baited you and forced you to make an unnecessary/dangerous maneuver. Since AIs don't think like we do, TAS perfect AIs would always take the safest option, and no one would commit to an option that can be punished, thus making the match endless. Deciding when a ''worse'' option is actually ''better'' not to become predictable is what differenciates human players from AIs.

Making humans play frame by frame in a TAS setting, even if it simulates human input, wouldn't work either because of two things. Not everyone has the same reaction time. Where do you set the limit? Do you take airforce pilots' average reaction time, or the reaction time of semi-competitive players? It would be too hard to set a line because there is no universal human reaction/input time. And also, it wouldn't work because that's not how we play the game. Yes, in theory, that's how we should make tiers, but playing in an environment where you can and will make frame perfect (or the best you're capable of) decisions and inputs simply will never happen in a real competitive environment, so it's kind of pointless. Yes, you get an accurate tier list of what the ''real', tiers are supposed to look like, but it wouldn't be representative of how we really play smash, which I think is important. That brings me to my next point.

----

There are different things that determine where a character is placed on a tier list.
1. Frame data. Moves with less startup/endlag offer less reaction time to the opponent and are safer. Generally, a character with good frame data will be doing better than a character with bad frame data.
2. Options. The more options a character has, the more he will be able to mix things up and become less predictable. If a character has amazing frame data but only has one option for each situation, he will be outplayed hard because you will always know his next move.
3. Results. I know some people will not like this point, but I believe it is essential. A character might be the best on paper, but if nobody is ever winning with it, I don't care, it's not the best character. Theorycrafting is good to an extend, but without some concrete results and proofs it doesn't mean anything. There can always be some aspects that we forgot to cover tha make the character worse, and if he actually is any good, it's only a matter of time before we get the confirmation with real results. TAS testing will never give us ''real'' results, and neither will theorycrafting.

There are some other things to consider, though. Range, for example, is important. This is often overlooked, but if a character has phenomenal range, it doesn't matter if his options or frame data are not that good. He can simply abuse his range to get significant results. This is why disjointed hitboxes are so good. You can't hit the opponent, while he can hit you. He has little to lose and a lot to gain. The other way around, if a character can't hit you at all without getting hit in return, he's bad. This is a little extreme, but it's to illustrate my point. You could make similar arguments about weight, speed, and other characteristics like these.

I think these are the three main things to consider when making a tier list: frame data, options, and results. Yes Yoshi has really good frame data, but he doesn't have too many approach options. Yes Pacman has amazing options, but he lacks in the frame data. Yes Villager seems amazing, yet he doesn't have too many significant results. Don't kill me for these examples, please. I know they aren't 100% accurate, but you get the idea. A character will (almost, it's debatable) excel in every category, that's why it's so damn hard to make a tier list. Is frame data more important than options? How much should we factor results in the tier list? Is being fast and really light better than being slow but super heavy? That's why we shoud not seek to make perfect tier lists, but to make them as close to what the majority of players can agree with, and that's also why they are changing all the time.

Woooah sorry if it's TL;DR. I tried to make it short, but I really wanted to express myself on this and didn't want to leave things out.

 

[zpxociv]

Slow-mode duels could help determine tiers as long as you keep in mind what's reasonable human input in normal speed. TAS is probably too much work but is still the most precise way to determine which character has overall the most advantages in each situation. Human limits are only a factor that determines what input timings you should apply when doing TAS. I say don't knock it until you've done it right.
Screw tourney results. They're not a cause but merely an effect of the real causes of what makes the tiers what they truly are.

 

[COLINBG]

Yes, but with TAS or slowed downed, you'd have much more time to make a decision. You play a lot less instinctively and analize more. It results in a different playstyle that has nothing to do with normal matches since you have an easier time making reads and punishing. (Fast) characters who can punish easily are at an advantage, and (larger/heavier) characters who have a lot of lag at a disadvantage. The gap between curent "tiers" would become more apparent. It also rewards characters like Marth and Jiggly who can time and space their attacks more effectively (Tipper, Rest, etc).

The results wouldn't be similar to what we do in real time, and while you would get more "perfect" results, they don't matter if they don't reflect what we see in tournaments.

Testing the slowing down/TAS with AIs (who don't need time to think more) instead of human players doesn't work.

I'm all for the idea of testing it if anyone is willing to do it seriously, but I'm pretty confident that it would not give us the tier list that we want to have.

 

[Scarlet Jile]

Results table = Tier list. It's like that for Smash 4, just like any other fighter, and just like every smash iteration before it, and they are accurate. They are always accurate at their job BECAUSE they are results lists.

They are accurate results lists, sure. But tier lists have to take more than that into consideration, or else really good characters without high-level rep become bottom tier.

Seeya at bottom tier again, Yoshi. So long, Pikachu. You're bad now, sorry.

 

[LightLV]

Slow-mode duels could help determine tiers as long as you keep in mind what's reasonable human input in normal speed. TAS is probably too much work but is still the most precise way to determine which character has overall the most advantages in each situation. Human limits are only a factor that determines what input timings you should apply when doing TAS. I say don't knock it until you've done it right.
Screw tourney results. They're not a cause but merely an effect of the real causes of what makes the tiers what they truly are.

Slow-motion could possibly help, if only for making execution easier. But TAS is completely useless in determining anything in fighting games, because it takes execution (the human element) out of the equation. People make mistakes, bad calls, panic, screw up execution, become stressed, and do a bunch of other things. Technically, anything done in TAS is indeed possible, but no match takes place under perfect conditions and execution, and games are designed for humans to play them like humans, not frame-perfect simulations. (Which is why in many games, TAS speedruns are able to just break the game.)

So it wouldn't really be worth it to know that one move beats another, or one move can combo into another, but the execution requirements are is so specific and tight that it would only ever happen by random chance in a real match.

They are accurate results lists, sure. But tier lists have to take more than that into consideration, or else really good characters without high-level rep become bottom tier.

Seeya at bottom tier again, Yoshi. So long, Pikachu. You're bad now, sorry.

But tier lists don't only take results into consideration. Within weeks of Smash 4's release, veteran players were able to determine who would likely be good characters. It's the traits and capabilities that make characters strong, and when you have strong attributes, you're more likely to win tournaments.

Yes, this kind of disregards characters that people don't play often, but they can only fly under the radar for so long. Yes, it also kind of disregards characters that are difficult to use...but that's just another human execution problem. To put it simply, if a character was theoretically the best character in the game mathematically, but is too difficult to use with any consistency, they aren't the best character in the game, because you'd get more consistency out of an easier but weaker character. They may be placed on a higher tier based soley on their possibilities, but if the best of the community can't squeeze results out of them, it's unlikely they'll be winning any competitive tournaments.

 

[Favid]

Bump. I felt like this should have been established a month ago. Aside from the diddy nerfs, the "god" tier list hasn't changed much.

 

[Scarlet Jile]

Results table = Tier list.

[Tier lists] are always accurate at their job BECAUSE they are results lists.

But tier lists don't only take results into consideration.

Welp, okay then. As long as we're on the same page.

 

[LightLV]

Welp, okay then. As long as we're on the same page.

If those three statements confused you somehow, or sounded like contradictions, then I don't know how to help you.

Edit: Oh, yes i do. Someone already did it for me.

I think these are the three main things to consider when making a tier list: frame data, options, and results. Yes Yoshi has really good frame data, but he doesn't have too many approach options. Yes Pacman has amazing options, but he lacks in the frame data. Yes Villager seems amazing, yet he doesn't have too many significant results. Don't kill me for these examples, please. I know they aren't 100% accurate, but you get the idea.

These 3 points aren't necessarily weighted 1:1, but a tier list is pretty much this. Regardless of frame data, options, hitboxes, ect, the end result of these variables should produce somewhat consistent results. Regardless of any of that other stuff, if the character in question can't beat other characters near it, then it's not going to be sitting at the top of a tier list.

You made a distinction between a "results list" and a tier list as if they are somehow fundamentally different. One just incorporates more information than the other. You also suggested that there is no "fact" that goes into making a tier list. We can dance around the definition of the word "fact" in this context, that is also just not true.

 

[Scarlet Jile]

These 3 points aren't necessarily weighted 1:1, but a tier list is pretty much this. Regardless of frame data, options, hitboxes, ect, the end result of these variables should produce somewhat consistent results. Regardless of any of that other stuff, if the character in question can't beat other characters near it, then it's not going to be sitting at the top of a tier list.

You made a distinction between a "results list" and a tier list as if they are somehow fundamentally different. One just incorporates more information than the other. You also suggested that there is no "fact" that goes into making a tier list. We can dance around the definition of the word "fact" in this context, that is also just not true.

It just sounds to me like you're taking a long time to get around to saying what I said at the very beginning, which I then described as being incomplete without arbitrary or subjective prioritizing. No matter how much data you try to incorporate, how you prioritize that data still comes down to a series of votes and opinions. Just because it happens behind closed doors doesn't make it any less subjective.

You can't tell me that a tier list and a results list are the same and then describe to me how they are different in the next sentence. "One incorporates more data than the other." Not exactly, because one is data, and the other incorporates that data as one component of a larger whole. Do they overlap? Of course. But to say they're the same is like saying your state tax is the same thing as the sum of all your taxes.

You have continued to quote me and claim that you disagree, but aside from trying to tell me very mundane quarter-truths like a tier list and a results list are the same thing, I don't really see what you are getting at. I hope that's not the only thing.

 

[LightLV]

Alright then, here's my real problem with your post.

There are precious few, if any, facts to do with Smash Bros tier lists. Yes, they can observe data and extrapolate tiers based on tournament results, but that's a results-table, not a tier list. You can observe character data and matchup data and try to measure out the strengths and weaknesses objectively, but we have seen many times that data and practical application often don't coincide.

And then you do this:

We can also try to somehow splice character data, matchup data and tournament results, but --

Yeah, yeah we can, it's exactly what majority of people will agree on what a tier list is. Majority of the time it works. Fluctuations happen, sure. But the process still works.

There are precious few, if any, facts to do with Smash Bros tier lists.

the decision-making process for how to prioritize that data will ultimately be subjective.

Yeah, but no? You can call it subjective, or non-factual, but with the exception of a few in the middle and who belongs at the very bottom, a general consensus can be made within a community on where everyone should be placed.

It's not like the highest level players in the community are going to provide wildly different tier lists every time you ask them, and it's hiiiiiiiiiiiiiighly unlikely some middle-low tier is going to magically shoot to the top because of...reasons, especially not when people are actively using them. So trying to make a distinction between a "results" and "tier list" in this fashion just comes off as grasping.

It's like you're arguing that, no, you can't have tier lists because they're made by people, who have their own opinions, and we'll never have enough solid, raw data to make a definitive one. Well, duh, of course not. But that doesn't make SSB64's tier list, or Melee's tier list, or Brawl's tier list any less effective at its job at determining who's likely about to get bodied in a matchup.

tl;dr It being "subjective" doesn't change a thing. Consensus with enough data works well enough.

 

[Scarlet Jile]

Yeah, yeah we can, it's exactly what majority of people will agree on what a tier list is. Majority of the time it works. Fluctuations happen, sure. But the process still works.

What's your point? I never said anything to contradict this. Tier lists are fine, but as even you seem to be aware, they are not divinely inspired nor the result of flawless calculation.

Yeah, but no? You can call it subjective, or non-factual, but with the exception of a few in the middle and who belongs at the very bottom, a general consensus can be made within a community on where everyone should be placed.

Okay? Cool. Doesn't apply to anything I've said.

It's not like the highest level players in the community are going to provide wildly different tier lists every time you ask them, and it's hiiiiiiiiiiiiiighly unlikely some middle-low tier is going to magically shoot to the top because of...reasons, especially not when people are actively using them.

This is very likely true. Still not applicable or contradictory to any statements I've made.

So trying to make a distinction between a "results" and "tier list" in this fashion just comes off as grasping.

I'm not trying to do anything. At this stage, you might be the only person in the world to whom the distinction isn't obvious. There is no ulterior motive to that statement except to make known the truth in the statement itself.

It's like you're arguing that, no, you can't have tier lists because they're made by people, who have their own opinions, and we'll never have enough solid, raw data to make a definitive one.

Nope! Never said any of this. Never implied it. In fact, I made a tier list of my own in this very thread, and I happen to like it despite the high probability that it is wildly inaccurate.

tl;dr It being "subjective" doesn't change a thing. Consensus with enough data works well enough.

Works well enough? It probably does, for all intents and purposes. I never said it wouldn't, so once again, I can't even begin to fathom why any of this is relevant. Maybe it's time to take a break from the internet. Get some air, come back with a clean head and read back through the thread to see exactly what you are arguing. Because as far as I can tell, you are imagining about 80% of this conversation.

 

[Pyr]

Ya.... Not going to get into your guy's argument here. My opinion in this is we'll be ready to make the Smash 4 list when the average smasher doesn't get a say in it.

 

[Arturito_Burrito]

The answer to this is actually NO.

Tierlist in smash games take 2 years to come out, welcome to the community, please be patient while we all try to figure out this meta.

 

[LightLV]

Works well enough? It probably does, for all intents and purposes. I never said it wouldn't, so once again, I can't even begin to fathom why any of this is relevant. Maybe it's time to take a break from the internet. Get some air, come back with a clean head and read back through the thread to see exactly what you are arguing. Because as far as I can tell, you are imagining about 80% of this conversation.

Wellllllll sir, about 85% of the time i'm posting from work, so as much as i'd love to take that advice, unfortunately i cannot. Sadface.

And oh yes, i probably am imagining 80% of this conversation, hell possibly 99.8% judging from that breakdown, so we should just drop it, since we clearly aren't even in the same thread right now. Apparently. In fact, i don't even know how I got here, or where my underwear went, and someone should probably call the cops.

 

[Charey]

The answer to this is actually NO.

Tierlist in smash games take 2 years to come out, welcome to the community, please be patient while we all try to figure out this meta.

Actually both Brawl and Melee had tier lists out in about a year, Melee took longer then Brawl but it had to make it's own scene before a tier list could be made.

The first list will not be perfect, but it will be better then any list made by one pro or voted on by people who don't know enough for their votes to have anything to do with the real meta game. (To be fair I would count myself in that second category.)

 

[vegeta18]

until we get something better, im taking zero`s tier list that he is making as the official one. I hate reddit tier lists and other community voted tier lists because they are often based off the thoughts of a lot of casual players. The community tier list on this site isnt too bad cuz its at least a site designed for competitive smash players and has some really good opinions, but its extremely out dated. Ceo and EVO will probably give us some good idea of where characters stand though and who knows what will change after the lucas patch.

 

[Sleek Media]

How hard is it to figure out which characters are placing high in tournaments, and which aren't?

S - NERF. HER. RANGE.
A -
Junk -

Nearly everyone else is either around the same viability, or isn't played enough to evaluate.

 

[Gawain]

It's time to face it, man. Despite everything, Bowser remains a mediocre character.

The upshot is, there is no E or F tier as far as I'm concerned. Everyone is viable, some are just more viable than others.

I don't know about that. Characters like Sheik and Rosalina pretty much completely shut down some characters. Especially Rosalina I'd argue. A well played Rosalina can make it almost impossible for some characters to do anything.

 

[clydeaker]

WOMEN ARE TOP TIER!
-
-
-
-
-
-
-
-
-
-
-
-
-

 

[Sleek Media]

Rosa is way overrated. Nobody is winning big tournaments with her, even with Diddy out of the way. She benefits greatly from people not understanding her mechanics and playing against her incorrectly, but that isn't a real strength. Rushdown characters dance around her neutral game and keep Luma dead. Without customs, she can find herself in matchups where she must approach without any good options (Yoshi, Shiek, etc).

Customs definitely benefit her, but it isn't clear how much. Sure, shooting star bits force an approach, but they don't do anything to help with her hard matchups (rushdown characters poking at her and Luma). Luma warp is a great KO tool, but she loses considerable spacing options and many desync combos. Regardless, it says a lot when gaining a worse version of Shiek's standard special can bump even an already strong character like Rosalina up the tier list.

 

[Shaya]

Umm, seems like the latest few posts are actually talking about what the 'tier list' is/should be, which we do have a dedicated thread for.

Has the lifetime of this thread's purpose come by or?

 

[TheHypnotoad]

This game has over 50 characters. That is a LOT of characters. It would be virtually impossible to actually come up with a distinct order for every single character, when its very hard to tell for a lot of them which is better than the other. And due to the larger number of characters, a definitive tier list would be much too controversial BECAUSE there is so much ambiguity in how good the characters are. Tier list arguments for Smash 4 would definitely be angrier than for Melee and Brawl.

tl;dr We don't need a tier list.

 

[Emblem Lord]

A patch is dropping in a week

Are you people ****ing serious?

 

[Charey]

I haven't heard anything about balance changes with the patch, so far it looks like it's only new content so it wouldn't affect anything. Besides if we wait until we know there will be no patches we will not have a tier list for 10 years.

 

[Emblem Lord]

New characters effect balance. That one char adds 40 something more match-ups.

How do you not think that affects anything?

It affects EVERYTHING!!!

 

[Stringsbatman]

Why not make a rotating tier list based on results? Make it every week based on all large tournaments and seperate for customs? Could be a stickied post with the title, "This weeks tier list."