Is it time to make a new matchup-chart?

[Ringedge]

There has been about 5 or so years or advancement in the metagame since the last one was made, or is it not different enough to matter? It also takes a big effort from the whole community to get accurate results.

 

[victra♥]

The meta has shifted significantly since the last tier list and even more so since the last match up chart update.

I'm in full support of this.

 

[Spak]

The current tier list came out July 26, 2013, so I don't think the meta has shifted that significantly... Maybe another year or two?

http://www.ssbwiki.com/List_of_SSBM_tier_lists_(NTSC)#Eleventh_tier_list_.28July_26th.2C_2013.29

 

[Ringedge]

I think the tier list is pretty accurate for right now, the only real significant changes would be Yoshi moving up to mid-tier, but besides that it's about right. But this thread isn't about the tier list, it's about the match-up chart.

 

[Spak]

Oh, sorry, I was talking about the wrong thing. I agree that the matchups need to be updated.

 

[Walbytamer11]

I'm all for a new matchup chart! A new tier list isn't exactly shouting out to us but I think that a few character shifts should be considered.

 

[Massive]

The problem with match-up charts is that it's difficult to empirically collect evidence for them.

Unless you're able to somehow get a large sample of match outcomes between players of relatively even, high levels of skill, the chart isn't really going to be any more accurate than previous charts are now.

Especially considering how underrepresented most lower tier characters are and how hard it is to judge player skill relatively to one another.

 

[Max?]

Every matchup chart made in the last 5 years has failed. People don't agree on numbers, or what they mean, and it ends up a huge mess.

If one was to be made, I think it needs to happen with the "OT Matchup System", which is simple and streamlined. There are three matchup ratios

Evenish
Advantage
90-10

Outside of that, good luck agreeing on whether someting is 60/40, or 65/35

 

[Ringedge]

What if we did something like people saying what they think the match-up is, and having to show a match that supports their theory? And I think a format of even, plus/negative 1, plus/negative 2, plus/negative 3 could work, although it's not as accurate as people would probably want. Or maybe have different ones for what people who main the character say and what others say? I just think a new match-up chart could be really useful for people starting out to get a general idea of how hard match-ups are.

 

[Laudandus]

Knowing how hard a matchup is is the opposite of useful. You want to be thinking about how to beat your opponent, not whether it's ok that you are losing to them.

A tier list is more useful because players know to select a character who's high enough on the tier list to do well, but I think that the current tier list is accurate enough for that save a few maybe-too-low characters like Pikachu/Samus/Yoshi.

 

[zuloon]

Knowing how hard a matchup is is the opposite of useful. You want to be thinking about how to beat your opponent, not whether it's ok that you are losing to them.

The way I look at it is you want to figure out the greatest degree to which you can humiliate your opponent. That could involve choosing the worst matchup and beating him anyway.

 

[BTmoney]

The problem with match-up charts is that it's difficult to empirically collect evidence for them.

Unless you're able to somehow get a large sample of match outcomes between players of relatively even, high levels of skill, the chart isn't really going to be any more accurate than previous charts are now.

Especially considering how underrepresented most lower tier characters are and how hard it is to judge player skill relatively to one another.

Every matchup chart made in the last 5 years has failed. People don't agree on numbers, or what they mean, and it ends up a huge mess.

If one was to be made, I think it needs to happen with the "OT Matchup System", which is simple and streamlined. There are three matchup ratios

Evenish
Advantage
90-10

Outside of that, good luck agreeing on whether someting is 60/40, or 65/35

I kind of agree with these posts. But something I've figured out is that MUs are almost irrelevant among good characters (top 8) and good players. They still aren't overly important when you get down to characters like Luigi and Pikachu (Yoshi is a bit different since FD hard counters him and simple good shielding and not jumping into DJ armor makes you really safe lol. The MU is so easy for spacies if you abuse dair/lasers (both), shield, and avoid getting shield grabbed which is super slow.). In general I think MU charts are kind of pointless unless you're learning the game cause idk the last time I even thought to look at it. Hardly ever in relevant MUs do you get to a point where you feel your character stopped you from winning and you feasibly couldn't have played much better. Tier Lists are a bit more fun and a little less trivial imo.

How about Even(ish), Slight, Advantage, Skewed for high level play (not necessarily super-top)

Fox-Falco = evenish
Fox-Marth = evenish
Falco-Marth = Slight Falco (I stand by this 1010% lol, Falco has the edge in the MU)
Fox-Sheik = Slight Fox
Falco-Sheik = Evenish/Slight Falco
Fox-Puff = Advantage Fox
Falco-Puff = Slight Falco
Fox-Peach = TBH in this meta I say slight Fox. You could argue advantage
Falco-Peach = Evenish
Fox-Falcon = Advantage Fox
Falco-Falcon = Advantage Falco (borderline skewed IMO but it's hard to tell with player skill gaps. I.e. S2J can beat any Falco except Dr. pp/mango but they are also just a lot better than everyone else so you can't see the MU play out)

Stuff like that.

Sheik-Peach = Slight Sheik
Sheik-Marth = Slight Sheik / could argue Advantage but I'm talking about good players who can play MUs correctly
Sheik-Falcon = Advantage/Slight Sheik
Sheik-Bowser = Skewed Sheik
Samus-Fox = Slight Fox
Samus-Falco = Evenish (imo)
Peach-ICS = Skewed Peach
etc.

Mus aren't an excuse to lose. It's more like a general measurement of how much better you have to play and effort you have to exert.

 

[Laudandus]

The way I look at it is you want to figure out the greatest degree to which you can humiliate your opponent. That could involve choosing the worst matchup and beating him anyway.

Sure if you're Borp, but the majority of newer players care more about getting good at the game than they do about humiliating opponents they are better than anyway.

 

[SAUS]

I like the idea of using terms instead of numbers. The numbers are kind of dumb anyway. How can you be sure that it is exactly 60-40? What's the difference between 80-20 and 90-10? Why are there only multiples of 5?

Hardly ever in relevant MUs do you get to a point where you feel your character stopped you from winning and you feasibly couldn't have played much better.

Also this.

 

[Fortress | Sveet]

Like Max? said, every attempt to make a matchup chart in the last 5+ years has failed, hard. Before you even attempt to start the project new, you'd be wise to look into the past attempts and appreciate why they failed. As someone who contributed to them, I still remember pretty freshly. Too freshly to get behind a revival.

 

[victra♥]

Strong Advantage
Advantage
Even
Disadvantage
Strong Disadvantage

That's all we need.

 

[theMagnumDragon]

I feel like at the top-level the match up chart is very accurate.

 

[1MachGO]

I feel like you'll never find a true consensus on MUs. They change depending on the skill level, they change depending on the state of the game, they change depending on who you talk to, etc.

MUs the majority of people agree on:

-Falco, Fox, and Sheik have the advantage over Falcon to some degree
-Peach hard counters Ice Climbers
-Fox counters Puff
-Sheik has the advantage over Marth to some degree
-Sheik has the advantage over most mid and low tiers

Everything else is probably fair game

 

[xbombr]

I personally don't see the value in a match up chart at this point. The only value of one at this point, imo, is for tiering purposes. It's not nearly as important to know HOW WELL one character does against another, it's important to know IF they can and/or HOW they can handle the match up. It's pretty common knowledge by this point what characters you can choose to counter specific match ups, and generally if you don't have an idea of who to choose, just pick up Fox or Falco (or Sheik if it's some semi-floaty low tier character that she can chaingrab to 60%).

I agree with Massive that collecting empirical evidence for these things is impossible, you have the actual advantages of each character to take into account, but the differing skill levels between players makes it more or less impossible to do so in any sort of objective way. There are too many variables to do it effectively.

Sveet has a valid point too. In the past, the match up chart threads tend to devolve into ****posts and unhelpful discussion anyway. I'll be the first to admit that I contributed to the stupidity of one of the old MU threads, but I was far from the only one doing so. Most people who participated in the discussions had too little real experience with a wide variety of players and skill levels. Even the people who actually played low tier characters extensively, lacked the experience with certain match ups that the community has collectively about matchups like Fox Marth or Fox Falco. These threads were simply 2-3 pages of people arguing with each other over pointless crap broken up by someone with actual experience with a character being weighed too heavily by their supporters or ignored entirely by people stuck in their ways.

Let's be real here, who really cares if Mewtwo's match up with Pichu is 60-40 or 70-30? The only people who care are the people who have a personal attachment to that character and think it's better than a character someone else has a personal attachment to. It's too much theorysmash because 70% of the match ups never occur (at least often enough) in a tournament setting at a high level anyway. It's more or less pointless. A more useful project would be constructing match up charts in each of the character subforums. That way if Samus players think Fox is a 60-40 match up, they can share strategies and reasons as to why, and if Fox players think it's a 70-30, then they can do the same there. There is no point in having these two groups of players come together and argue over pointless **** like Samus's percent chance of winning against Fox when both players have an even skill level. This situation rarely, if ever, actually happens and a match up chart based on this has no application to actually playing the game.

For the above reasons, I don't support making a new match up thread.

 

[victra♥]

70% of the match ups never occur (at least often enough) in a tournament setting at a high level anyway. It's more or less pointless.

This is an important point.

I think something that would be useful is having advantage and disadvantage with tips for each character (if you follow league of legends, this is a good example of this idea: here) and keep it with just relevant characters and outlier characters (YLink, Luigi, Yoshi, etc) and only when it's significant enough to note.

So for example (no need to discuss the details, just throwing it together roughly):

FALCON

Weak Against:
Strong Disadvantage
Disadvantage
Disadvantage

Strong Against:
Strong Advantage
Advantage
Advantage
Advantage

Even With:
Even (Non Final Destination)

Notable:
Disadvantage (Final Destination)

---

FOX

Weak Against:

Strong Against
Strong Advantage
Strong Advantage
Strong Advantage
Advantage
Advantage

Even Against
Even
Even (Non Final Destination)

Notable
Strong Disadvantage (Final Destination)
Above Average
Above Average
Above Average

Something to that effect.

 

[xbombr]

Honestly Victra, the mention of stages in yours makes it more useful than what was done in the past. At least that gives you an idea of where the match up is better or worse. It can give players an idea of what they can counterpick or what they shouldn't counterpick.

The counterplay idea like league of legends is nice. At least there won't be a ton of awful debates over that stuff and it can be based on practical knowledge of the match ups rather than ridiculous theory smash involving tournament results.

I'm still not entirely convinced something like that is a full community project though. It seems more suitable for something like this to exist character subforums (I'm pretty sure something like this exists, or used to, in a few of them already, at least in some fragmented sort of way). That way the community can decide on what match ups are relevant to their own character.

I do think that if we're going to do match ups as a community, there needs to be a major change from what was going on before. Cutting down the number of match ups discussed is a pretty good idea since many (even most) of them aren't seen in practice. However, we need to ask ourselves: how many of the matchups of top characters are already common knowledge, and what purpose does the chart serve if they are?

 

[1MachGO]

@ victra♥ I completely agree with @ xbombr , MUs are too subjective and even getting the list down to common knowledge wouldn't be particularly helpful.

To elaborate a bit further, the Marth/Fox MU is the embodiment of subjectivity. I've heard its even, I've heard its slightly in Fox's favor, I've heard its slightly in Marth's favor, etc. Factor in the polarity of FD, and then you could argue that the MU changes from Bo3s to Bo5s, its a headache and too complex to sum up in a few words.

 

[Acryte]

Sheik-Bowser = Skewed Sheik

This is where I stopped reading. This is why we need to have only ranked players allowed to vote.

 

[BTmoney]

This is where I stopped reading. This is why we need to have only ranked players allowed to vote.

so you basically read the whole thing?

and it was just an example to give context to what the words mean. maybe that MU isn't skewed in sheik's favor? Idk, can't she CG bowser, I'm asking? Isn't bowser garbage (yes)? What were you getting at cause I'm actually confused.

sorry I don't know the bowser meta like the ranked players do

 

[EddyBearr]

I would take the same format as t3h Icy used and change the following:

-One chart, not one per stage, but one overall.^[1]
-Only 5 scores: Big Advantage, Small Advantage, Even, Small Disadvantage, Big Disadvantage.^[2]
-If a matchup isn't known, have an "unknown" icon to exemplify it.^[3]
-Each claimed matchup should have a piece of exemplary evidence behind it.^[4]

[1]: It's not about who wins on FD: It's about who wins the entire set. If FD makes it so the set could go either way, then the matchup should be even. If PS makes it so Sheik definitively has a Small Disadvantage to Fox, then so be it.
[2]: It doesn't matter if Sheik +2's DK or +3's Ness. Sheik wins them both (Big Advantage), and if Sheik loses, it's Sheik's fault, not the matchup's possibilities.
[3]: EX: Jigglypuff vs Ness is (afaik) unknown, not "+3 advantage". If you can't find a recent Young Link vs Pikachu set video between two capable and comparably skilled players, say the matchup is unknown.
[4]: If Samus goes even with Fox, show us a video of Plup vs Colbol to verify; if Sheik has a Big Advantage on Ness, show us the Chain Grab info.

 

[xbombr]

This is where I stopped reading. This is why we need to have only ranked players allowed to vote.

This is the problem with these threads too. Only letting high ranked players vote on the match ups is sensible, but it severely limits what characters are represented.

I'm sorry, but I also hardly trust even top ranked players to know the ins and outs of every match up with how underrepresented low, and even mid to some extent, tier characters are at a high level. Even the super pros have probably only played a few serious Bowser or Kirby players (very few or none of which are actually on their skill level) and have experience that is primarily limited to messing around at smashfests or during downtime at tourneys.

The old tier threads did, in fact, weigh the advice of high ranked players or at least players who were notably good with their characters (i.e. could get into brackets/place at large and/or national tournaments) more heavily than the random smashboarder's contribution (in fact there were times where match up numbers were changed just based on the contribution of relatively few top level players). The problem is that even back then, when there were far more active high ranked players visiting the melee GD on a regular basis, there were too few that contributed or could contribute to the thread. I imagine nowadays it would be even harder to get input from enough players that the project would be worth tackling.

Another issue we have with making these threads is that we have to judge skill levels when using empirical evidence (i.e. videos and/or tournament results). The point of the chart is to show match ups, but we cannot truly judge skill level since that is independent from the character one chooses, yet their ability to win/compete is affected by it. For example, a highly skilled Bowser player might be just as skilled at the game as one of the top 5 smash gods, but since he uses Bowser, his ability to win tournaments is limited by his character. However, empirical evidence would indicate that he is not actually the same skill level as the top 5 because he loses to them consistently based on the match up. What we're left doing is judging how much of the player losing is their character and how much of them losing is on them. This is a fruitless process because you have 2 independent variables affecting your result and there is no scientific way to go about remedying the situation.

Where this leads is that the threads all devolve into impractical theorysmash because people are too set in their opinions discussion is necessary. Even the top players didn't completely agree on match ups the last time we tried this, and as a result, the issues inherent in the process weren't solved.

 

[Acryte]

I feel you man but if you allow non-ranked to vote it should be for a mid-level vote, because that's really what we are contributing, mid-level understanding and theory-crafting which has little relevance when addressing top-level metagaming.

The dude said that the Sheik vs Bowser matchup is skewed slightly in Sheik's favor. Uuugh yea, skewed... if you mean Bowser being completely skewered and getting slow-roasted over the flames. Talk about one of the most god-awful matchups in the game. Bowser has to commit to everything and a defensive Sheik has to take no risk at all.

 

[Acryte]

Every matchup chart made in the last 5 years has failed. People don't agree on numbers, or what they mean, and it ends up a huge mess.

If one was to be made, I think it needs to happen with the "OT Matchup System", which is simple and streamlined. There are three matchup ratios

Evenish
Advantage
90-10

Outside of that, good luck agreeing on whether someting is 60/40, or 65/35

The perfect example of people working out 50-50 or 60-40:
https://www.youtube.com/watch?v=xtLePUJgg78;t=3m57s

 

[Fortress | Sveet]

If you're making a matchup chart, its really unimportant what that matchup between bowser and kirby is for their placements. A matchups value to the chart overall is equal to how often it is played.

You could shorten the process and make a chart for only the top X characters. If you want to include the low tiers, only provide their matchup ratios against the top X characters, that way there is no pointless speculation.

 

[xbombr]

If you're making a matchup chart, its really unimportant what that matchup between bowser and kirby is for their placements. A matchups value to the chart overall is equal to how often it is played.

You could shorten the process and make a chart for only the top X characters. If you want to include the low tiers, only provide their matchup ratios against the top X characters, that way there is no pointless speculation.

I believe Victra suggested something similar above. It really would streamline the process. That way you wouldn't see a page of good top tier discussion interrupted by "Hey guys, let's spend 5 pages talking about Mewtwo's nair in his match up with Ness"

In fact, I think it would be cool to have it set up to where only one match up would be discussed at a time.

 

[Fortress | Sveet]

No his idea is different.

Instead of 26x26 grid, have 16x16 grid. We are then ignoring the bottom 10 vs bottom 10 matchups that we agree aren't played enough to be accurate. If we do make matchup values for the bottom 10, only make them vs the top 16, since those are the matches they would be against 99.9% of the time.

16/10 split is just an example.

 

[Kaiza]

Yeah a new matchup chart is definitely needed
I also think stages specific stages deserve to be taken in consideration when making it though. Maybe a top 8 matchup chart for every stage to keep it simple?

 

[Kidney Thief]

Let's make a Match up chart that only includes the characters that are used by at least 0.5% of smashers. So we would not include : DK, Game and watch, Kirby, Zelda, Bowser, Ness, Mewtwo and Pichu. For the rest of the cast we could make a poll where people would vote only about the match-ups they feel like they know that way they don't just put random numbers because they don't know. They would also need to give an explanation or else their vote would not count that way we don't have people spamming ridiculous matchups that make no sense. I'm already working on a thread, but if no one does this I'll go for it in a few months once my guide is complete. I'd rather have someone else do it tho

 

[CriticalRibbon]

Check it out.

http://www.reddit.com/r/SSBM/comments/2ke82v/rssbm_matchup_chart/

It's not a perfect system, but it's something.