By LUDWIG VON KOOPA - Did Sora clinch it?
Just two weeks ago, KoopaTV published the very last possible look at Super Smash Bros. Ultimate's online metagame BEFORE the last downloadable content character, Sora, was added. Sora, with his... unique playstyle (along with version 13.0.0's other balance changes), would no doubt change the metagame and which characters people would use.
But how can that change be measured? Well, Nintendo provided the opportunity: their NintendoVS US Fall Open 2021 tournament was divided into two events: one on October 15–16 right before Sora released, and the second on October 29–30 right after Sora released. By comparing the character usage stats collected from the tournament before and after, we could gauge the effect that Sora—and the other balance patch changes—had on the metagame.
By the way, the finals were streamed. No one used Sora in those, though there were plenty of other DLC characters:
While all eight of the players featured on that stream are getting some kind of prize, congratulations are in order for the four GRAND PRIZE winners: Jake (Enderman / Mario) of Region 8, Xerzal of Region 6 (Min Min), MarkSDKing (R.O.B.) of Region 1, and Avarice (Ridley) of Region 3. Nintendo promises to mail out your prizes within sixteen weeks.
As a result of the thousands of games and hundreds of players for the tournament, I have aggregated the fighter usage stats for Event 2 of the Super Smash Bros. Ultimate NintendoVS US Fall Open 2021. Below is a chart, along with a text-based list and my commentary—commentary on both the stats and the tournament itself (which is also commentary on the flaws of the data):
- Sora 648 4.28% N/A
- Pyra & Mythra 645 4.26% -1
- Joker 463 3.06% 4
- Bowser 443 2.92% 0
- Terry 436 2.88% 12
- Little Mac 427 2.82% 12
- Mario 427 2.82% 4
- King K. Rool 422 2.79% -5
- Banjo & Kazooie 409 2.70% -1
- Cloud 407 2.69% -4
- Steve / Alex 394 2.60% 22
- Kirby 394 2.60% -10
- Ness 380 2.51% 14
- Sephiroth 377 2.49% -4
- Byleth 354 2.34% -10
- Link 351 2.32% -4
- Kazuya 330 2.18% -3
- Samus 290 1.91% -5
- Incineroar 289 1.91% 0
- Captain Falcon 265 1.75% 1
- Piranha Plant 250 1.65% 3
- Yoshi 250 1.65% 4
- Sonic 244 1.61% -8
- Ganondorf 244 1.61% 4
- Luigi 229 1.51% 10
- Donkey Kong 214 1.41% -3
- King Dedede 212 1.40% 13
- Hero 201 1.33% -12
- R.O.B. 191 1.26% -20
- Pikachu 182 1.20% -8
- Min Min 181 1.19% -6
- Inkling 179 1.18% 9
- Pokémon Trainer 161 1.06% -13
- Marth 159 1.05% 23
- Ridley 158 1.04% 12
- Bowser Jr. 158 1.04% 42
- Roy 154 1.02% -1
- Bayonetta 147 0.97% 0
- Young Link 146 0.96% 12
- Palutena 140 0.92% -8
- Mewtwo 124 0.82% 19
- Toon Link 123 0.81% 2
- Mii Gunner 114 0.75% 10
- Corrin 113 0.75% -2
- Lucas 110 0.73% -16
- Mii Swordfighter 108 0.71% 0
- Lucina 107 0.71% -2
- Falco 106 0.70% 4
- Isabelle 103 0.68% -6
- Wii Fit Trainer 102 0.67% 19
- Greninja 100 0.66% -2
- Pit 100 0.66% 10
- Shulk 98 0.65% 6
- Dark Samus 97 0.64% 7
- Dark Pit 93 0.61% -25
- Jigglypuff 89 0.59% 12
- Mr. Game & Watch 86 0.57% -1
- Mega Man 85 0.56% 0
- Simon 81 0.53% 18
- Ike 78 0.51% -23
- Ryu 78 0.51% -7
- Chrom 72 0.48% 2
- Snake 67 0.44% -29
- Mii Brawler 67 0.44% -25
- Ken 66 0.44% -17
- Zelda 63 0.42% -32
- Wolf 61 0.40% -12
- Robin 57 0.38% -3
- Rosalina & Luma 56 0.37% -3
- Peach 51 0.34% 0
- PAC-MAN 50 0.33% -4
- Wario 49 0.32% 12
- Lucario 48 0.32% 6
- Meta Knight 48 0.32% 1
- Villager 48 0.32% -1
- Ice Climbers 45 0.30% 7
- Duck Hunt 45 0.30% -14
- Fox 43 0.28% -28
- Richter 41 0.27% -3
- Daisy 41 0.27% 5
- Dr. Mario 29 0.19% 0
- Diddy Kong 23 0.15% 0
- Zero Suit Samus 22 0.15% -12
- Olimar 8 0.05% -4
- Sheik 4 0.03% -13
- Pichu 4 0.03% -13
I highlighted all of the characters that were directly impacted—buffed—by the version 13.0.0 patch. As you can figure out, the patch didn't seem to have any logical impact. Unused characters were still unused. No one's using Lucario just because his down tilt is faster, or Dr. Mario because his Dr. Tornado is way safer to use. Maybe the only notable and statistically relevant change is Little Mac's. And maybe that's more because Little Mac can do some interesting things against Sora (like Straight Lunge armour through Sora's low-power moves) than the specific buffs he got.
Sora took the #1 spot from Pyra and Mythra, who only just got that spot from former perennial #1 Joker (who inched back up this time to #3). While it looks very close in the stats, Sora is in fact far more popular than what these stats show. That's because Battlefy, the eSports tournament platform that Nintendo sponsored the event with, screwed up the character selector that tourney entrants have to go through with all of their matches. Sora wasn't available to choose from in the first 2/3s of Regions 5–8's ladder round. While he was available for all of Regions 1–4's ladder (because those regions went after the fix happened), keep in mind that Regions 5–8 (which represents Texas and everything east of it) had over 71% of the population of players, with really the only notable state in terms of the number of people who live there in Regions 1–4 being California.
Sora being absent for such an extended duration of time means that contestants put... whatever other character in his place on the character selector. I've heard people put in characters as diverse as Pit (because they look kind of similar), Cloud (because they share a common character), and Mii Gunner (...because they wanted Mii Gunner's usage stats to be higher, I guess?). I manually adjusted for Sora's numbers and against those other characters in the very few instances where people told me, but Sora's numbers remain dramatically overcounted, and other characters likely overcounted as a result. It's a testament to how popular Sora was that he's still at #1 despite that setback. At least one player testified to me that he experienced a sort of chilling effect in that he wasn't sure if Sora was actually allowed in the tournament because you couldn't pick him, so he didn't use him as much as he otherwise would have. That means there's definitely pent up Sora fandom not represented above.
Though I must admit... in the eleven sets I did in the ladder, I encountered three different Ness players, but zero Sora players. I don't have any personal evidence that Sora was that popular, but that's why we look at aggregate stats. Well, there might be a reason for that: Sora players tended to lose their matches, collectively having a win rate in the 40%s. Since the ladder round matched players against opponents of a similar skill (win record), and since I did pretty well in the tournament, that meant I wasn't matched against the low-skilled Sora masses.
Still, this whole debacle taints the data, which is quite upsetting. One of the Battlefy tournament organisers personally messaged me, “sorry about your data” during the tournament. (And I didn't directly mention the data to him before he apologised for it, though he and everyone else involved with the tournaments knows I collect and publish it, so he knew what I was thinking.)
The number of tournament participants is so low that a couple of dedicated character mains taking the whole weekend off to celebrate Halloween or something would be enough to drastically affect a character's relative placing, so I wouldn't look too much into this. While Event 1 had 1,320 registrants, Event 2 actually had only 1,319... but also an over 5% decrease in participation rate for people actually showing up, resulting in less data overall that's even more prone to outliers. I blame Nintendo for this because they did an awful job in promoting the events. You'd think the average fan would jump at the chance to use their new toy, Sora, but most people didn't know about these events were happening at all. Instead, Nintendo literally chose to promote the NintendoVS US Fall Open 2021 two to three days after Event 2 already had ended. And it wasn't a dedicated message, either—it was alongside a long list of other, allegedly “competitive events” like the Animal Crossing: Pocket Camp four-year anniversary. Click that link if you don't believe me. It's baffling.
Anyway, I'm still hoping Nintendo and Battlefy do more of these. If only because I want to see Sora off that #1 spot. And who knows... I might win one of these some day. Maybe I'd need to main a pay-to-win character first, though. At least King Bowser's results are steady.
Ludwig dropped from 9th place to 17th place in the Top 32 bracket between Event 1 and Event 2, though he was still good enough to get into the Top 32 bracket to begin with. One of his problems is that he keeps getting matched up against characters he doesn't know how to even fight, like Sonic the Hedgehog. Anyway, feel free to comment about... the character stats, how Bowser Jr. actually had the biggest movement of any character on the list, Nintendo's lack of commitment to competitive gaming, or anything else.
The next available fighter usage stats are for April 2022.
Sora being too popular will make me avoid this game for an indefinite period of time.
ReplyDeleteI'd say he might drop off, but... all 12 DLC characters are in the Top 31 (with Min Min at the 31 spot), and half the Top 10 are DLC characters, so I guess they never actually do fall off in popularity.
Delete