Shedding Tiers: Building a graphical character comparison spanning all levels of play

[jimmyjoe]

If you've managed to land in this thread, you're more than likely already familiar with the Backroom's Official Tier List.
This is widely considered a fairly accurate character comparison at around the highest level of play, which may or may not be your level of play.
While most people on this forum are likely striving to reach the top level of play, The majority of us here are probably somewhere along the learning curve, with much to learn and master in this beautiful, timeless, masterpiece of competitive gaming.

Was there a time where you thought Ness was unstoppable?
Maybe you used to think Falcon was right there along side Link,
or any other idea you now disagree with thanks to your greater skill and knowledge of the game.
So the question is...
Where does each character stand at any given point in the Smash64 learning curve?
(given the BR ruleset)
That's exactly what this potential project would aim to determine.

I imagine it could be simply a graph where:
x = Tech skill and knowledge of the game
y = relative rank among the 12 characters(or something more precisely representing their strength.)

I remember a thread about the learning curve for smash, that could be a good starting point for determining x.
http://www.smashboards.com/showthread.php?t=50864&highlight=learning
It's definitely a bit dated, but a good starting point.

I imagine Kirby would remain high,
Samus would fall,
Falcon would rise,
DK would have a turbulent journey,
Link would hold down the bottom.
Personally I think it would be really interesting to see where these shifts occur.

Anyway, If you like where this is going I need a lot of help
This might be best handled by a backroom member,
but it should likely take into consideration opinions across all skill levels.
I'm not much of a graph maker, and I'm not yet sure how we could compile the data.

A$/Surri if this has been done feel free to delete it, or if someone with time/forum-skillz wants to spearhead this project idea, let me know

 

[asianaussie]

Are you suggesting 12 graphs, one for each character?

Interesting concept, but everybody learns differently. That's the main problem with tier lists anywhere except at a high level of play - people can be great in some areas but poor in others, and will be better suited to different characters accordingly.

 

[ciaza]

jj: Are you trying to show how someone who mains say Samus compared to someone who mains say Kirby, and show much better the Kirby player would be even though Samus and Kirby player have played for the same amount of time?

 

[cheeseball341]

I am a 5a.

In srslies though, Is this what you mean

 

[jimmyjoe]

Are you suggesting 12 graphs, one for each character?

Interesting concept, but everybody learns differently. That's the main problem with tier lists anywhere except at a high level of play - people can be great in some areas but poor in others, and will be better suited to different characters accordingly.

I was thinking one large graph, with a colored line for each character.
What you said does make a lot of sense, but I think it could be worthwhile to try. In a thread like the North American players tier list, we could take players generally grouped together (like all low B tier kaillera players) and survey how they felt there characters matched up with equally skilled players characters. Maybe we would find something unexpected, such as jiggly puff generally winning more than fox.

jj: Are you trying to show how someone who mains say Samus compared to someone who mains say Kirby, and show much better the Kirby player would be even though Samus and Kirby player have played for the same amount of time?

not exactly, but that certainly is related, and could definitely be worked out in the potential graph if the measurement used for Y was something precise.

 

[asianaussie]

Mm, this'd be an interesting project then. If anything, I'd suggest a few 'archetype' learners (everyone ends up in roughly the same place once they get online/into competitive local scenes). For instance, those who come from Brawl/Melee, those who come from traditional fighters, those who played when they were little kids, etc.

Could be a rather significant undertaking, and it'd be more for interest than anything, but the big problem with most projects, backroom and otherwise, is a lack of interest, so if this can garner some support then it could be great.

 

[kys]

This is absolutely impossible. People themselves don't even know "how much better they're getting" at a given time, much less remember correctly with each individual character. Add in all the different matchups, AND all the people they've ever played in those given matchups (which also needs to account for the growth in skill the people they play, at varying and unknown levels), and you have yourself an impractical, pointless, and inaccurate task.

Sorry to be a buzzkill, but it's the truth. It's a cool idea, but it simply can't be done accurately.

 

[Dingus]

This is absolutely impossible. People themselves don't even know "how much better they're getting" at a given time, much less remember correctly with each individual character. Add in all the different matchups, AND all the people they've ever played in those given matchups (which also needs to account for the growth in skill the people they play, at varying and unknown levels), and you have yourself an impractical, pointless, and inaccurate task.

Sorry to be a buzzkill, but it's the truth. It's a cool idea, but it simply can't be done accurately.

I dont think the intention is to create some perfect official graph, but rather a neat project that might produce an interesting perspective on how character tiers change with skill.

Quite frankly I think the graph that Cheers whipped together is about 5 seconds is already pretty accurate. I would critique it, but as it was obviously created as a hasty joke, I would feel pretty foolish doing so.

 

[Sangoku]

Funny, I had thought about such graphs a few months ago. I could try to help.

 

[jimmyjoe]

Mm, this'd be an interesting project then. If anything, I'd suggest a few 'archetype' learners (everyone ends up in roughly the same place once they get online/into competitive local scenes). For instance, those who come from Brawl/Melee, those who come from traditional fighters, those who played when they were little kids, etc.

Could be a rather significant undertaking, and it'd be more for interest than anything, but the big problem with most projects, backroom and otherwise, is a lack of interest, so if this can garner some support then it could be great.

Yeah I'm with you on all that, but it might make a single graph even more difficult with the 'archtype' system you describe. The easiest way, albeit maybe not truthfully the most accurate way of labeling levels of skill, IMHO would be in terms of Advanced Techs and combo ability. Spacing is up to the individual, but once a player is at a certain level of advanced techs(e.g. level 3- dash dancing, tech rolling and tech chasing, and short hops but not yet Z-canceling, level 4- all of that plus some inconsistent z-canceling, level 5 mostly always z-cancels, etc.) then where would each character potentially stand in terms of one another, given the players are equally adept at spacing.

I dont think the intention is to create some perfect official graph, but rather a neat project that might produce an interesting perspective on how character tiers change with skill.

Quite frankly I think the graph that Cheers whipped together is about 5 seconds is already pretty accurate. I would critique it, but as it was obviously created as a hasty joke, I would feel pretty foolish doing so.

Exactly. And agreed, Cheers has the right idea!

Funny, I had thought about such graphs a few months ago. I could try to help.

That would be awesome for this project seeing how nicely you've done with the shields thread.

I dunno if any body else would dig this thought, but I'm also interested to see where along the learning curve a character might an advantage over another character at a slightly higher level of play. For instance, a Luigi who understands z-canceling and does so most of the time, vs. a pikachu or Kirby that doesn't z-cancel at all, may still be at a disadvantage.

 

[The Star King]

I'm not sure this is a good idea. Like asianaussie said, people learn differently - there are many winding paths to the top. You can do an averagish sort of thing, but I don't think that would be too helpful. I won't stop you if you really want to do it, though.

 

[ciaza]

This project is over seeing as 1der's graph should be official.

.

 

[jimmyjoe]

This project is over seeing as 1der's graph should be official.

.

As long as he puts Australian just slightly below average where it belongs

 

[The Star King]

inb4ciazarage

 

[Battlecow]

Star King and Kys are right. You can say general things, like "Kirby's easy to learn" or whatever, but anything with numbers is already too official. Low-level players vary way too much for a valid comparison to be made.

Inb4 A$ "Ness rocks at low levels of play"

 

[sharksquail]

We should just take smasherlinks tier list and the official tier list and play connect the dots.