Well, guess I'll put my money where my mouth is and collect some hard data, to serve as an example.
I.
Let's take a look at the Smashboards tournament ranking results (by character) for events from Feb 1st - Now:
Several interesting things jump out, but let's break it down.
II.
The percentages are the % of players (of that character) who placed at the corresponding levels. In
theory, this should be a similar spread for characters of
all skill levels, if we assume that that players are themselves automatically filtering for tiers and only playing their best character(s).
In more advanced theory, we'd expect to see a slight tilt for higher-tier characters having higher percentages, merely because the most competitive "spike" mindset players (who win more as a group) will gravitate towards them.
We see important exceptions to this trend:
We all know that
has a top-level problem, but according to the data
and
suffer similar fates in the
aggregate smash community. Yes, iStudying and Ranai have made their few wins pretty big ones, but the general trend within this period still stands.
and
are disproportionately played by people who perform
really well with them. ROB is actually just, really strongly performing period.
Shockingly (to me),
wins relatively
rarely for the pool of players that play her; she's more popular with the broader smash community than we (I?) assumed.
I've long said that feminine characters (including Jiggs), extreme heavyweights, and
specifically all have
abnormally high character loyalty such that they are
played disproportionately by experienced mains only.
I've also suggested in private that this has a radical warping effect on win-rates relative to tier strength--and we see that here. All of those characters exhibit abnormally high Top 16 or Top 8 placement rates relative to the number of players using them--ESPECIALLY
and
, who may be even more dominant at lower level play than we suspected.
Curiously
is included in this pattern while
is not, which is an important lesson about gender identity and assumptions.
III.
Returning to the top of the rankings, we see some surprising results. We can get a general placement score by combining the top 16/8/1 placements, weighted significantly towards the latter. We can see that 5/6ths of the entire cast falls within one standard deviation of the average score, with all the outliers being on the top.
The elephant in the room is that
is king, and
a whole boatload of people are both playing
and winning with
at all levels of play. Looking back historically, this was no February fluke; Mario has been popular (and unlike Falcon, winning) everywhere for a long time.
Zero might not be worried about them, but this is currently the most typical Grand Finals in Smash 4.
What's striking is just how much stronger performing these two characters were over this period, about an
entire standard deviation each above the others. In fact, if you were to turn this into a results list, it might look pretty shocking:
TOP LEVEL VIABLE:
SS:
S:
A+:
A-:
B+:
B-:
C+:
NOT TOP LEVEL VIABLE:
C-:
D+:
E+:
E-:
Note: I didn't include the Miis, because we (as a community) don't play with them much. (So there isn't enough data to submit for analysis.)
IV.
Now, I already here you protesting: Results don't automatically equal tiers.
Wait, who said anything about tier lists? I never said tier list. You said tier list!
Results are reality, tiers are theory--an
expectation of future reality at a given level of play. The problem is that defining that level of play is hard, and almost everyone ends up overshooting.
We like to say that a tier list is exclusively supposed to measure "top-level play", but what exactly is top level play? Is it
just Zero?
THE OFFICIAL TOP LEVEL SMASH 4 TIER LIST
Zero Tier:
Not-Zero Tier: Everyone Else.
A tier list may be top-level, but even top-level should be a considerable range.
What if we limit our results data to just the top 500 ranked players? That's a really small pool of the
millions of players that play Smash, but it turns out that the
overwhelming majority of the results in our rankings already come from those players. In fact, I originally intended to run two separate sets of data (all players and top 500 only), but there was no point since the results were the same. (Since it was
the same data!)
In other words, the temptation to attribute this list to the mouth-breathing For Glory masses doesn't fly. Smashboards' ranking methodology isn't perfect and the data is often incomplete, but it should be fairly representative for the questions we are asking.
Results indeed aren't the same as tiers, but whenever they are at odds, the burden of explanation always falls on theory rather than reality. Results are never wrong, and can never be truly dismissed.
V.
To anyone who has been around competitive gaming long enough, it shouldn't be a surprise that the reality of which characters are
actually winning the most has poor correlation with the echo chamber that is public opinion.
My shenanigans aside, this is a prime moment to reflect on both how limited the bounds of our own discussion are relative to the broader smash community and how we treat new users who offer us new perspectives. (Even if those new users are just a-hole moderators in thinly-veiled disguise.)
Why did a new user posting this information (albeit with minimal context) yield vitriol, reports, and zero likes? Why did the posts like mine poking fun at a new user get a whole bunch of likes?
Why is it that I can sit here typing all this out
in advance, knowing with
100% certainty that this is
exactly how it's going to go down?
Thanks for the internet points, suckas!
Yes, we're all cool Internet nerds who are great at video games, and God only knows I will take the prize for smug know-it-all who takes his wit too far. But we have
got to cultivate a community that is not merely tolerant, but willing to engage with fresh faces (and the often... unconventional ideas they bring to the table).
"Git gud" is not a viable path to the future of Smash, even in the survival-of-the-fittest jungle that is the CCI thread.
Btw, my lawyers want me to remind everyone that alternate accounts are against the Global Rules of Smashboards and are considered major Terms of Service violations and grounds for banning.
VI.
Anyway, where were we? Ah, right:
Please discuss anything you think is out of place!
