In most years, there’s about five or six Melee players who are capable of winning a major. Nowadays, that number is much higher. Since 2021, we’ve seen eight different people win them. Within the rest of the competitive field, there are players who can defeat multiple of those eight major champions. Taking that into account, there may be a double-digit number of players who could win majors.
In theory, the best way to calculate someone’s chances of winning a major should be straightforward. You research their head-to-heads against every present opponent, calculate their win probabilities against each one, estimate the probability of running into each opponent, and put it all together to create one final number. Okay – on second thought, that’s not as straightforward as I initially thought. Or if it is, it takes too damn long to collect head-to-heads for hundreds of players.
Last night in Melee:
1. Plup won CEO 2023, beating Wizzrobe and Panda (FL)
2. Chem won The Nightclub VIP, beating Hax$ and Aklo (NY)
3. Fiction won Lawless, beating Kürv and cliché (SoCal)
4. KJH won BODIED, beating Ben and Kuyashi (MI)
— Melee Stats (@MeleeStatsPod) June 25, 2023
What can we do instead? One: be creative, and two: be efficient. In today’s column, I’m going to somehow lean into the “Melee Stats” part of my personality, as well as the “dungeon master” part of my personality to simulate 10 random majors. Before getting into the results, I will discuss which players’ chances I want to spend time analyzing and what led me to choose those players. I will talk about a couple key factors behind what’s led them to be strong contenders, and then I will explain to you how I’ve simulated our ten future majors before jumping into the actual results.
NOTE: For the rest of this column, I will be referencing majors by Liquipedia’s definition, as well as counting only offline tournaments to head-to-heads and tournament results alike. This column was also written before CEO 2023, so any discrepancies in projections come from that.
The Winners Quarters “Critical Point”
At most majors, Top 16 is usually when our top seeds run into their first notable Top 50 opponent. Now, technically it’s not impossible for major contenders to lose before that – we literally saw this happen last year. But it’s nonetheless exceedingly unlikely. We’re talking about the average major here, not necessarily 1 to 1 with Genesis.
You might be wondering why I’m only accounting for this from the winner’s side. You’d think there would be many examples of all-time greats making deep loser’s runs from outside Top 16 at a tournament However, when we’re talking about winning a major, these instances are incredibly rare. From what I could find in the last 15 years, only two majors out of 139 have been won from people who lost before winners quarters: Low Tier City 7 and Pound 3.
I suppose 1.4 percent is better than 0 percent. For now though, I’m going to treat losing before winners’ quarters as a death knell to an individual’s chances of winning a major. Keeping that in mind, we’re going to list out all the people who have made it that far at a major this year.
- 2saint
- Plup
- Moky
- Mango
- Cody Schwab
- N0ne
- Jmook
- Zain
- Leffen
- Rishi
- Axe
- Hungrybox
- Zuppy
- aMSa
- Soonsay
- lloD
- KoDoRiN
- Polish
- MOF
- Aklo
Not all of these players have a feasible chance of winning a major. As a result, we’re going to filter this group of 19 through additional criteria: has this player made multiple top eights this year or have they made it, bare minimum, to grand finals at a major? When we do that, we end up with the following group:
- Plup
- Moky
- Cody Schwab
- Jmook
- Zain
- Leffen
- Hungrybox
- aMSa
- lloD
- KoDoRiN
- Aklo
- Mango
That brings us 12 people whom I’d consider having nonzero chances of winning a major for the rest of the year. How reliable have those twelve been when it comes to reaching the aforementioned winners’ quarters critical point? I’ve gathered their success rates below, using each of their last ten majors attended.
| WQFs | Majors | WQF success rate | |
| Jmook | 9 | 10 | 0.90 |
| Cody | 9 | 10 | 0.90 |
| Zain | 10 | 10 | 1.00 |
| moky | 6 | 10 | 0.60 |
| Mango | 8 | 10 | 0.80 |
| aMSa | 8 | 10 | 0.80 |
| Hbox | 9 | 10 | 0.90 |
| Plup | 7 | 10 | 0.70 |
| Leffen | 9 | 10 | 0.90 |
| Aklo | 2 | 10 | 0.20 |
| KoDoRiN | 3 | 10 | 0.30 |
| lloD | 5 | 10 | 0.50 |
If we wanted to really drill down here, we could look at their head-to-heads vs. players outside their group for the whole year and use that as a frame of reference for their win-rates against the field. But, because we have limited time and I really don’t want to put all of that together, we’re going to use these rates instead as a ‘napkin math’ assumption for how reliable they are vs. everyone else.
Top-Level Head-to-Heads
Within 2023 alone, we just don’t have enough head-to-head data to make assumptions about players’ long-term trajectories. For the sake of collecting information that could point us in a general direction, I’ve decided to gather a cross-table head-to-head for each of our twelve players, as well as how they’ve performed against each other, offline, since Smash Summit 11.
| Jmook | Cody | Zain | moky | Mango | aMSa | Hbox | Plup | Leffen | Aklo | KoDoRiN | lloD | |
| Jmook | N/A | 4-5 | 7-4 | 3-0 | 3-1 | 2-8 | 1-11 | 2-1 | 3-1 | 1-0 | 6-0 | 1-0 |
| Cody | 5-4 | N/A | 8-7 | 1-2 | 2-8 | 10-2 | 8-2 | 7-1 | 2-2 | 0-3 | 4-4 | 0-0 |
| Zain | 4-7 | 7-8 | N/A | 4-0 | 6-5 | 3-4 | 11-2 | 2-1 | 3-2 | 3-0 | 5-0 | 2-1 |
| moky | 0-3 | 2-1 | 0-4 | N/A | 0-4 | 5-1 | 2-5 | 1-0 | 1-1 | 3-1 | 1-3 | 1-0 |
| Mango | 1-3 | 8-2 | 5-6 | 4-0 | N/A | 2-6 | 7-5 | 3-1 | 3-0 | 2-0 | 5-1 | 2-1 |
| aMSa | 8-2 | 2-10 | 4-3 | 1-5 | 6-2 | N/A | 7-5 | 0-3 | 3-1 | 5-0 | 5-0 | 2-1 |
| Hbox | 11-1 | 2-8 | 2-11 | 5-2 | 5-7 | 5-7 | N/A | 3-6 | 2-5 | 2-0 | 9-2 | 5-0 |
| Plup | 1-2 | 1-7 | 1-2 | 0-1 | 1-3 | 3-0 | 6-3 | N/A | 0-1 | 1-0 | 5-0 | 3-0 |
| Leffen | 1-3 | 2-2 | 2-3 | 1-1 | 0-3 | 1-3 | 5-2 | 1-0 | N/A | 1-0 | 4-0 | 1-1 |
| Aklo | 0-1 | 3-0 | 0-3 | 1-3 | 0-2 | 0-5 | 0-2 | 0-1 | 0-1 | N/A | 2-5 | 0-2 |
| KoDoRiN | 0-6 | 4-4 | 0-5 | 3-1 | 1-5 | 0-5 | 2-9 | 0-5 | 0-4 | 5-2 | N/A | 1-5 |
| lloD | 0-1 | 0-0 | 1-2 | 0-1 | 1-2 | 1-2 | 0-5 | 0-3 | 1-1 | 2-0 | 5-1 | N/A |
Before any of you yell at me for including sets from two years ago – I understand that not all of these head-to-heads may carry over exactly the same amount of weight. As you all know, they don’t account for recent trends, such as moky’s two consecutive victories over Hungrybox in spite of losing the first five head-to-heads. However, I still think that they can point us in a direction of what could happen when two players run into each other.
To partially account for which opponent they run into in winners’ quarters, I’ve combined the above stats into another chart that entails each of these players’ chances against fellow contenders. More often than not, many of these sets have happened this late or later into the bracket. You can consider the following table for each player’s win-rate vs. the average fellow contender.
| Total Wins | Total Losses | Total Sets | WinRate | |
| Jmook | 33 | 31 | 64 | 0.52 |
| Cody | 47 | 35 | 82 | 0.57 |
| Zain | 50 | 30 | 80 | 0.63 |
| moky | 14 | 23 | 37 | 0.38 |
| Mango | 42 | 25 | 67 | 0.63 |
| aMSa | 43 | 33 | 76 | 0.57 |
| Hbox | 51 | 49 | 100 | 0.51 |
| Plup | 22 | 19 | 41 | 0.54 |
| Leffen | 22 | 18 | 40 | 0.55 |
| Aklo | 6 | 25 | 31 | 0.19 |
| KoDoRiN | 16 | 51 | 67 | 0.24 |
| lloD | 11 | 18 | 29 | 0.38 |
NOTE: Ironically, some sets between these twelve players happened in prior majors before winners quarterfinals. It’s worth noting this just to acknowledge the differences in time span, as well as player trajectories at majors. I do not think these players would run into each other at a ‘typical’ major before winners’ quarters, but it’s certainly possible if they all attended something like Genesis, Big House, or a Mogul Moves event.
There is, of course, one final thing to consider. It’s so obvious that you might have forgotten about it.
Likelihood of Attendance
A major doesn’t have to technically feature everyone. More or less, it just has to have enough prominent people and prestige. A major can be anything from an installment of Genesis to arguably something as small as Lost Tech City. Keeping that in mind, let’s examine each of those twelve players and how present they have been at each of the last ten majors.
| TO14 | BOBC | MU | Collision | G9 | SWT | Mainstage | Apex | SS14 | LSI | Total | |
| Jmook | Yes | Yes | No | Yes | Yes | Yes | No | Yes | Yes | Yes | 8 |
| Cody | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | 10 |
| Zain | Yes | Yes | No | Yes | Yes | No | No | Yes | Yes | Yes | 7 |
| moky | Yes | Yes | No | Yes | Yes | No | Yes | Yes | No | Yes | 7 |
| Mango | Yes | Yes | No | No | Yes | Yes | Yes | No | Yes | Yes | 7 |
| aMSa | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | 10 |
| Hbox | No | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | 9 |
| Plup | No | No | Yes | No | Yes | Yes | No | No | Yes | No | 4 |
| Leffen | No | Yes | No | No | Yes | Yes | Yes | No | Yes | Yes | 6 |
| Aklo | Yes | Yes | No | Yes | Yes | No | Yes | Yes | No | Yes | 7 |
| KoDoRiN | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | Yes | Yes | 9 |
| lloD | No | No | Yes | Yes | No | Yes | No | No | No | Yes | 4 |
You’ll notice a few things: Cody and aMSa basically attend everything. Behind them, Hbox and KoDoRiN are similarly active. Zain, moky, Mango (just in terms of technically showing up), and Aklo will usually be present when they’re not on breaks. At the bottom are three people who balance Melee with other obligations: Leffen, who’s now playing two other fighting games, Plup, whose future in the game I have zero clue about, and lloD, a doctor.
All in all, there’s three rough factors we’re going to measure for each player from the categories we described above: how they do vs. the field, how they perform vs. each other, and how likely they are to be present at a major. And now, with all this build up out of the way, we can get into the fun stuff. What we’ll be doing is some combination of the following for each hypothetical major:
- Simulating the attendance of each of those twelve aforementioned players with a D11, where a 6 or higher leads to the player being in attendance. Anyone with a major attendance rate at 90 percent or above will receive a +2 advantage to their roll. Anyone with a major attendance rate at 70 percent or above will receive a +1 advantage to their roll. Anyone with a major attendance rate in between 50 percent and 70 percent will receive no advantage. Anyone with a major attendance rate beneath 50% will receive a -1 disadvantage.
- Simulating the likelihood of each of those twelve aforementioned players to make it to winners’ quarters. This will be based on their respective winners’ quarters success rates and determined with a D11, as well as a similar advantage/disadvantage dynamic as what I described above. A 6 or higher is needed to make it this far in the bracket.
- Accounting for what happens if 9 or more of our 12 players are present at the event. If 9 or more of those players are calculated as having made ‘winners’ quarters,’ each player who has to fight another fellow contender in the ‘winners round of sixteen’ matchup will have it calculated on its own merits, similar to what I describe in the next bullet point, and the metric will be functionally replaced with a ‘make winners round of 16’ roll.
- Simulating each of the Top 12’s matchups against each other if they run into each other in winners’ bracket. Each player in a head-to-head will be assigned a die based on how many wins they have vs. each other, with the advancing player being the one with the higher score between the two dice rolls. To ensure volatility, ties will favor the loser.
- Ensuring fairness, as well as accounting for competitive volatility. In our head-to-head rolls, the losing player will be assigned +1 advantage on their rolls and the winning player will be assigned a -1 disadvantage. For example, in the case of Cody Schwab vs. Aklo, where Aklo is up 3-0 in the head-to-head, I will not be treating it as an auto-win for Aklo. Instead, Cody will need Aklo to effectively roll a “one” on a four-sided die. Future matchups beyond the head-to-heads we have will not be considered, just to simplify the process.
- In a situation where a member of the Top 12 is playing someone outside of this group after winners’ quarters, we are going to roll a D11, where the contender player only needs to avoid a 1 on the roll to advance.
- In a situation where two Top 12 members have never played each other or have only one set each, I will roll two D10s and pick the player with a higher score. Tiebreakers are determined by the higher ranked player.
- The seeding order I have for each of these events is completely subjective, as are the matchups, but they are determined by previous results, with Major 1 being my subjective call on how I’d seed an event with any present players today, and future majors being based off how the prior one or two majors went.
- We are going to assume, just for the sake of simplicity, that anyone who makes winners’ side of grands will have to roll a D100 and score within the range of their expected win-rate vs. the Top 12. For example, if Mango were to make grand finals at a future major from winners’ side, he would need to roll a 38 or above to win the first set and take the tournament.
- In a case where a player in winners’ side of grand finals does not win the first set, the player would roll another die with the same win condition. If the player rolls another “lose,” we will assume that the winner of the major is the player who has scored the highest volume of wins vs. the person from winners’ side, as they are the most likely culprit.
Major 1
- Who attends: Jmook, Cody, moky, Mango, Hbox, Aklo, and KoDoRiN
- Who makes WQF: Jmook, Cody, Not moky, Mango, Not Hbox, Aklo, KoDoRiN, Not Top 12 Member
- Who makes WSF: Not Jmook, KoDoRiN, Aklo, Mango
- Who makes WF: Mango, Aklo
- Who makes winners side of grands: Mango
- Does this player win set one of grand finals?: Yes
- Major winner: Mango
Major 2
- Who attends: Zain, aMSa, Jmook, Aklo, Hbox
- Who makes WQF: Not Zain, aMSa, Jmook, Not Aklo, Hbox, and three non-Top 12 players
- Who makes WSF: A non-Top 12 player, aMSa, Jmook, Hbox
- Who makes WF: Hbox, Jmook
- Who makes grands: Hbox
- Does this player win set one of grands?: No
- Does this player win set two of grands?: No
- Major winner: aMSa
Major 3
- Who attends: aMSa, Mango, Hbox, Jmook, Zain, Cody, moky, Plup, Aklo, KoDoRiN
- Who makes WQF: Not aMSa, Mango, Not Hbox, Jmook, Zain, Cody, KoDoRiN, Plup
- Who makes WSF: Plup, Mango, Cody, Jmook
- Who makes WF: Jmook, Mango
- Who makes grands: Jmook
- Does this player win set one of grands?: Yes
- Major winner: Jmook
Major 4
- Who attends: Jmook, Cody, Zain, Mango, aMSa, Hbox, Aklo, KoDoRiN, lloD
- Who makes WQF: Jmook, Cody, Zain, Mango, not aMSa, Hbox, Not Aklo, KoDoRiN
- Who makes WSF: Jmook, Cody, Zain, Mango
- Who makes WF: Jmook, Zain
- Who makes grands: Zain
- Does this player win set one of grands?: Yes
- Major winner: Zain
Major 5
- Who attends: Zain, Jmook, Mango, aMSa, Cody, Hbox, Plup
- Who makes WQF: Zain, Jmook, Mango, aMSa, Cody, Hbox, Plup, and a non-top 12 player
- Who makes WSF: Zain, Jmook, Hbox, Cody
- Who makes WF: Zain, Hbox
- Who makes grands: Zain
- Does this player win set one of grands?: No
- Does this player win set two of grands?: No
- Major winner: Cody
Major 6
- Who attends: Cody, Jmook, Mango, Leffen, Aklo, KoDoRiN, lloD
- Who makes WQF: Cody, Not Jmook, Mango, Leffen, Not Aklo, Not KoDoRiN, lloD, and a non-top 12 player
- Who makes WSF: Not Cody, lloD, Mango, Not Aklo
- Who makes WF: A non Top 12 player, Mango
- Who makes grands: Mango
- Does this player win set one of grands?: Yes
- Major winner: Mango
Major 7
- Who attends: Mango, Cody, Zain, moky, Leffen, KoDoRiN, lloD
- Who makes WQF: Mango, Cody, Zain, moky, Leffen, Not KoDoRiN, not lloD, and a non-Top 12 player
- Who makes WSF: Mango, Cody, Zain, moky
- Who makes WF: Mango, Zain,
- Who makes grands: Zain
- Does this player win set one of grands?: Yes
- Major winner: Zain
Major 8
- Who attends: Cody, Mango, aMSa, Hbox, moky, KoDoRiN
- Who makes WQF: Cody, Not Mango, aMSa, Hbox, moky, Not KoDoRiN, and two non-Top 12 players
- Who makes WSF: Cody, a non-Top 12 player, aMSa, Hbox
- Who makes WF: Cody, aMSa
- Who makes grands: Cody
- Does this player win set one of grands?: No
- Does this player win set two of grands?: Yes
- Major winner: Cody
Major 9
- Who attends: Cody, Zain, moky, aMSa, Plup, Aklo, KoDoRiN, lloD
- Who makes WQF: Cody, Zain, moky, aMSa, Not Plup, Not Aklo, KoDoRiN, lloD
- Who makes WSF: Cody, Zain, moky, aMSa
- Who makes WF: Cody, Zain
- Who makes grands: Zain
- Does this player win set one of grands?: Yes
- Major winner: Zain
Major 10
- Who attends: Zain, Cody, moky, Mango, Plup, Leffen, Aklo, KoDoRiN, lloD
- Who makes WQF: Zain, Cody, moky, Not Mango, Not Plup, Leffen, Aklo, lloD
- Who makes WSF: Zain, Aklo, moky, and a non-top 12 player
- Who makes WF: Zain, Aklo
- Who makes grands: Aklo
- Does this player win set one of grands?: No
- Does this player win set two of grands?: No
- Major winner: Zain
The End of Our Journey
There’s a few head-scratching results from this timeline. For starters, Leffen only attended three majors, which is way lower than what I anticipated. There’s also the fact that we saw Aklo – someone whom I just detailed had a bit of a rough time in getting to winners’ quarters – make winners finals twice, even taking a set from Zain in one of them.
Zain’s relative dominance, relative to the field, also has to be noted with a massive grain of salt. Within our timeline, Jmook, in a total twist of fate, didn’t go to any of the last four majors of the year. Perhaps influenced by the lack of his toughest opponent – although he did beat him once – Zain ended up winning majors in that time span where Jmook wasn’t around. Cody and Mango are tied at two apiece behind him, and Jmook and aMSa each have only one major. Ironically, just five people ended up winning majors of the 12 I decreed to be contenders with nonzero chances. Imagine that. Five people ruling the scene and one of them suddenly goes absent.
This is ultimately just one possibility etched out in creative fashion. I could do this another ten times and I’m sure I’d discover something different. I would not argue that my model is anything other than glorified Smash astrology, but stranger things have happened. In spite of a silly premise, I hope that this column tells a fun story of how things “could” turn out.