Nine years ago, Hungrybox won Evo 2016. Boasting 2,372 entrants, this was the largest offline Melee tournament of all-time, a community high that seemed impossible to achieve again. And yet last weekend, 2,422 people entered the Melee bracket at Supernova 2025 in Virginia, breaking the record for biggest major ever. It’s only fitting that the last person standing on Sunday was Zain: a home-grown Virginia native once again slowly making his case to be the best player in the world.
Tearing through a bracket of Kacey, Ben, moky, and Cody Schwab – the one player ranked above Zain on Summer SSBMRank – Zain has now tied Cody for number of majors won this year (three), while also taking the lead in the head-to-head count, 5-3. It’s additionally his twentieth major ever, just two below Armada, who currently sits third of all-time. Furthermore, by virtue of Supernova’s supermajor status, the victory moves Zain to seven supermajor first places for his career, one below Hungrybox, four below Armada, and five below Mang0.
Although Cody and Hungrybox placed to their seeds (second and third respectively), neither were the biggest story underneath Zain in the top four. That honor would go to Ginger, a previously retired top Falco player who achieved a career-high major fourth place following his sudden return to competition this year. After a year away from competing, as well as a quiet return to 17th place at GOML, Ginger defeated each of Grab, Aura, Sirmeris, Jmook, Axe, and Aklo at Supernova. He also became one of only 33 people to ever finish top four at a supermajor.
Beneath the top four are a series of other wild storylines from Supernova. Consider this column my recap of what I found – and continue to find – most interesting.
MD/VA is on the map
By Sunday, five of the tournament’s top sixteen placing players were from MD/VA: Zain, Khryke, Junebug, Kevin Maples, and RapMonster. For reference of this feat’s impressiveness, the previous major, GOML, had only one Ontario representative in moky at that same part of the bracket. Meanwhile, at last year’s Collision, New Jersey had two people there: Wally and 404Cray.
In the case of Khryke, this is his first ever top eight. Entering this event as the 27 seed, he had an epic losers bracket run where he followed up a loss to Aklo by beating Joey Donuts, Frostbyte, Sirmeris, Junebug, and SDJ – the seventh seed – before eventually falling to Hungrybox for seventh place. Meanwhile, RapMonster’s thirteenth place finish came by virtue of defeating Krudo again, and Kevin Maples ended up defeating Spark and Nicki on his journey there too.
It’s worth noting the number of strong performances from other MD/VA players too. For example, Frostbye ended up being not one, not two, but three strong Fox players in Dimension (who had beaten Bekvin last year), Rikzz, and Kevbot on the way to 25th place. Juicebox had a quiet run to 49th place, but he scored another win over a Top 50 player in mvlvchi. Seven defeated 2024 Top 100 player Nanami. I almost forgot that Daniel, the Yoshi player who defeated Graves, is technically a representative, in a way. On the whole, this event was a W for MDVA. WMDVA.
Axe’s neat loser’s run to top eight
Upon reaching the highest position on a rankings he’s had since 2019, Axe had fallen to the losers bracket at the hands of Sirmeris. Funnily enough, these two have a neat little history: though Axe had previously won all their last sets, they had a notorious barn-burner at Super Smash Con 2019, where well before Sirmeris broke out to the Top 100, Axe had to make a three-stock comeback to beat him and advance in the bracket after having already lost to PudgyPanda. Coming six years later, Sirmeris’ victory ended up being the biggest upset of the tournament by Upset Factor (7), by virtue of taking down the fifth seed as the 28th seed.
Obviously, given his wealth of time in the scene, Axe is no stranger to losers bracket runs. After losing to Sirmeris, he proceeded to defeat Inngenn (who beat fitzy and Gahtzu in coin flip matchups and a mini-losers bracket run of his own), Aura, n0ne, Magi, and Ben before falling to Ginger for seventh place. Speaking of Ben…
Ben’s big run to ninth
Not many players have routinely finished in the Top 50 in each year since the return of SSBMRank. Ben is one of them, but heading into Supernova 2025, he had a pretty tough bracket. Making it to Top 64 alone was nowhere close to guaranteed, as his projected opponent was mvlvchi, who beat Ben three times over the course of one weekend last year. Ben did end up beating him in a revenge set, yet his Top 64 path still looked rough, with daunting opponents in Jmook and Krudo coming up ahead.
Whatever happened this weekend, it apparently gave Ben the power to vanquish anyone in the Sheik ditto. He ended up 3-0’ing Jmook and then reverse 3-0’ing Krudo before falling to Zain and Axe for ninth place. This is Ben’s best performance at a major since he finished in fifth place at Riptide 2023 – a great turnaround from about a year and a half’s worth of obtaining national results well below what he’s capable of achieving.
The return of Fudge
In 2024, I predicted Fudge, then a power ranked Falco from Philadelphia, to break out and make the Top 100. Although that did not happen, he made the ballot for the end of year before then largely disappearing on a national level. I was a little saddened; Fudge was one of the coolest Falco players on the East Coast and someone who I believed in to break out for quite some time. It seemed like his story had ended right when it began, and when Fudge showed up Creed IV, he drowned in pools.
Thankfully, he showed up in form at Supernova. Attending his first major in several months, Fudge ended up defeating Freezus in a coin flip matchup before then losing to lloD and going on one of the most underrated runs of the tournament en route to 17th place. In it, Fudge would beat Vintage, Juicebox, Epoodle, and, most shockingly, OkayP, before finally being eliminated by Junebug. I don’t know if he plans on competing at Collision, but I would love August to be the month of Fudge. Month of Fudge. That sounds so weird to say out loud. Let’s run with it.
Other storylines
This section is for a group of players that I didn’t prepare full write-ups for, yet felt necessary to include. I’ll start with the breakout of three lower-seeded Falco players in Hawaii top dog Dova, 2024 Colorado No. 9, Loam, and Minnesota No. 2 Joey Donuts, Dova ended up making an upset over Maelstrom (which a very handsome man predicted), while Loam beat each of Electroman, Louis, and bonn. Meanwhile Joey Donuts upset each of eve and Inngenn to finish in 33rd place. I guess on a less obscure note, I should mention the fact that Aklo beat Hungrybox and made top eight here too; though not entirely unsurprising, it’s a good return to form for him after a brief slump.
I was also excited to see null, the ex-Top 25 SoCal Fox, return to form here. Though he didn’t beat JChu like I predicted, he ended up defeating ChuDat and upsetting Zanya in the losers bracket. On the note of ex-Top 100 players comes another good performance from Boyd here too. At Supernova, Boyd ended up beating Maelstrom to make the final bracket, adding to a year in which he’s already beaten SluG and Ben. While Boyd’s had a few bad showings this year, his top line resume looks pretty good for a potential return to Top 100 status.
Lastly, I want to mention Inngenn, a personal favorite rising player. Despite dominating Japan, beating Skerzo every week, and even starting to take sets from aMSa, Inngenn has had an unfortunate history of underperforming or merely placing to seed at North American majors. I’m happy to see that change last weekend; upon being sent to the losers bracket early, he proceeded to beat each of bambi, fitzy, and Gahtzu to also finish in 33rd place – barely enough to escape his fate as the 50 seed.
Okay – I lied because I actually have one final note here. I was surprised to see Massachusetts Sheik player 22K upset Mono and finish in 65th place. Although 22K has good results in his region, he’s not ranked in his state, and was more known as a community leader in the Worcester scene. This year, he dialed back his activity for a bit, so it’s nice to see him both score a notable win and represent his community on the big stage.
Was Supernova 2025 really the biggest Melee tournament ever?
Earlier this year, I talked about Ludwig’s gambit: paying for every single Melee bracket attendee in order to put his name next to the record of biggest offline Melee tournament ever as well as the largest online tournament. While he technically has the record, leading to the event, it was natural to question its meaningfulness. After all, there’s no way that all 2,422 people would be inside the venue itself to play Melee matches. Is this record fraudulent?
Heading into the event, I polled different Melee Stats community members about how many DQs (in this context, double DQs from the event) there would be at the tournament – the guesses ranged from 450 to 1,443. For what it’s worth, my original guess was 700, but I moved it up to 1,100 after experiencing a large amount of doubt. It turns out that I should have trusted my gut. The final number of DQs, per PracticalTAS, was 676, giving the event a still unprecedented high, though not entirely discrediting, 27.9 percent double DQ rate. This means that the real count of Melee players at this tournament was 1,746. This number on its own would make this the biggest Melee tournament since Evo 2016.
EDITOR’S NOTE: Shortly after this was published, I was informed of a different calculation made by a post here. I believe this won’t change too much of my column; just figured I’d link it.
Interestingly, past Evo iterations, like Evo 2016 and Evo 2015, did not always track DQs; they would sometimes put checkmarks next to advancing players in a bracket for both real victories and unplayed matches alike. Keep in mind that this doesn’t even account for a relic of past tournaments: when they used to confusingly DQ players from one side of the bracket and not the other. Today, this practice is largely frowned upon; for now, when I say DQ, I’m referencing double DQs (the only justifiable type of DQ).
Based on my background knowledge, as well as talking to major tournament organizers, the typical Melee major features roughly ten percent of its attendees double DQing. Now, I technically don’t have proof to say otherwise, but contextually speaking, ten percent seems gracious to grant as an assumption for smashers at Evo. This is an event with scale that reaches well beyond the Smash scene, features many potential newcomers, and runs 8:00 a.m. pools. Although most people I spoke to who have worked with Evo told me that the DQ rate for most fighting games is around twelve percent, the rule of thumb is that Smash and non-FGC games noticeably suffer from more DQs. One person told me that Pokken, for example, had roughly about thirty percent of its entrants drop out. Clearly applying the typical major DQ rate and the typical Evo tournament DQ rate doesn’t seem like a proper assumption to make. Yet neither does the Pokken tournament, which was announced before the game even came out and reportedly had children signing up for the event – only to then be presumably told “no” by disappointed parents.
Out of curiosity, I looked at Evo 2017’s Melee bracket; the first one I could find on start.gg. Using this tourney as a foundation, I noted the first twenty round one pools groups I could find with recorded instances of double DQs at Evo 2017 as a frame of reference. Again; I heavily doubt the meaningfulness of this number since many of the pools are recorded entirely with ambiguous checkmarks. Within these twenty groups, I found 46 DQs. Needless to say, 2.3 recorded DQs per pool, with the potential for more DQs when accounting for double-checkmarked results gives us a wide range of possibilities when it comes to extrapolating estimated DQ count for Evo 2016. There could be anywhere from one to three more DQs per pool; it’s genuinely impossible to say. At the very least, for this new “post-DQ” standard for evaluating entrant records, I think Evo and tournaments that don’t track DQs should be excluded.
What would be included, however, is Genesis 3, a major which tracked DQs, was held the same year, and had 1,828 registered entrants. To save time, I randomly chose 20 round one pools groups from this event and counted 29 DQs. After projecting this across all 128 groups, this would give us an estimated 185.6 DQs, bringing the estimated real count of Melee players at Genesis 3 to roughly 1,642 people, less than Supernova’s real count of Melee players.
EDITOR’S NOTE: As mentioned before, shortly after this was published, I was informed of a different calculation made by a post here, which also covers the same topic. For what it’s worth, the writer’s script had 1,734 confirmed attendees and those with null results at Supernova 2025, while Genesis 3 only had 1,619 confirmed attendees and those with null results.
The purpose of records is to be legitimate. The standard used for counting the number of people at a Melee tournament is by counting the number of registered entrants. Supernova’s record is legitimate by this standard. If you believe this standard is illegitimate, then when trying to account for DQs through an alternative standard, there still lacks significant evidence to definitively disprove Supernova’s claim. The one event in 24 years that could contest it did not consistently track DQs, is unique enough of a series to treat any estimation of DQs with a grain of salt, and would reasonably be excluded from any qualifying status. Therefore, I do not think there is enough evidence to confidently state that Supernova isn’t the biggest Melee tournament of all-time.
This tournament’s record-breaking status is a pleasant surprise, not a mere technicality or entirely a marketing gimmick. It’s also the culmination of hard work from people who love Smash and want to see it do well. The success of one series is obviously not necessarily one to one with the health of the entire scene, and yet it’s strictly good that grassroots events for a two-decade old game can continue to bring thousands of people together. Melee players, myself included, love to scrutinize and nitpick everything, but we should also celebrate cool things that happen. I think it’s pretty cool that after so many years, the Melee community can still break records.