Over the last couple of weeks, I’ve been thinking about the greatest major underdog runs in the history of Melee. I’ve also been trying to find the best ways – or if there’s any way, really – to empirically measure the best major underdog runs ever. In the last column, I covered the raw difference between placement and seed. As we saw, this methodology is not 1:1 with the greatness of an underdog run. Today, I’m taking another incremental step by measuring the difference between actual and expected placement for all the surprise major top eight runs in Melee history.
Actual placement is self-explanatory. When I say expected placement, however, I’m going by seed and assigning groups of seeds with multiple players to their expected placements. In other words, if someone is the 23rd seed and finishes in 17th place, they effectively placed their seed, even if there is technically a difference between their placement number and their seed number (their seed performance rating). In order to combat this, I’ve effectively assigned each placement a value, which you can see below.
| Placement | Placement Value |
| 1st | 1 |
| 2nd | 2 |
| 3rd | 3 |
| 4th | 4 |
| 5th | 5 |
| 7th | 6 |
| 9th | 7 |
| 13th | 8 |
| 17th | 9 |
| 25th | 10 |
| 33rd | 11 |
| 49th | 12 |
Once again, I would not necessarily take this statistic as a 1:1 measure of the surprise of an underdog run. For all the complaints you may have about seeding today, we’re not in an era where tournament organizers ran formats that required manual reseeding of final brackets following round robin pools. We also have far more data today to confidently seed players at our largest events. Additionally, I’ve decided to maintain our data set to just events from 2008 and onward, though I did initially want to expand our list to the MLG era. Ultimately though, the lack of publicly available data, as well as difference in bracket structures, prove too much to compare to other performances now, although maybe I’ll do a write-up in the future on my less confident historical ‘estimates.’
Seed Performance Rating of 5
- Mang0 at Pound 3
- Hungrybox at Genesis
- Amsah at Pound 4
- Taj at Genesis 2
- Nintendude at The Big House 3
- Wobbles at Evo 2013
- Darrell at Revival of Melee 7
- Abate at The Big House 5
- Druggedfox at HTC Throwdown
- Swedish Delight at Pound 2016
- Wobbles at Battle of Five Gods
- Ice at The Big House 6
- Swedish Delight at Smash ‘N’ Splash 2
- Zain at Super Smash Con 2017
- The Moon at GOML 2017
- Duck at Full Bloom 3
- Zain at Smash Summit 6
- lloD at Super Smash Con 2019
- Axe at Smash Summit 8
- Hax at Genesis 7
- Polish at SWT NA East Melee Finals
- SluG at Ludwig Smash Invitational
- n0ne at Genesis 8
- aMSa at The Big House 10
- Zuppy at Collision 2023
- moky at Genesis 9
- Junebug at Supernova 2024
All in all, we have 27 performances. While I won’t be reviewing all of these runs, I do want to mention the ones I’ve bolded and italicized so far because they seem of especially big prominence. We’ll begin with Mang0’s legendary win at Pound 3: from what I can gather from the recorded pools over at Liquipedia, he entered the final bracket as the 19 seed, but his spot in round robin pools seems to indicate that he was seeded decently higher before the event. Ergo, I looked at Cort’s spot – the player who effectively took the original planned place of Mang0, due to Vist, Wife, and Cort, in order of a championship-belt-series-of-events related to Mang0’s loss to Vist. In my opinion, this specific statistic captures a roughly accurate estimate of the disparity between Mang0’s expected placement by seed and his actual placement, but I will also say that if I were to truly make a biggest surprise runs list, this would probably be my number 1 pick because of its longer-lasting impact on the history of the game.
By the way, you might have noticed Armada’s Genesis breakout missing here. That’s because of the inherent limitations of this metric; he finished in second place as the sixth seed (an SPR of 3). Clearly, his seeding doesn’t fairly capture how truly shocking it was that Armada effectively beat the three best players in the world on his first try. However, I do think it accurately captures another breakout performance: Hungrybox’s. It’s easy to forget, but Hungrybox was originally seeded 16th at the original Genesis. When he finished in third place, that was a surprise – not the expected result. Mang0 and Armada understandably have their legacies forever tied to Genesis, but the same could honestly be said for Hungrybox.
On another note, any greatest underdog runs list would have to include Wobbles’ heroic Evo run. It’s underrated by our metric here solely due to the strict criteria: Wobbles was the ninth seed of this tournament, and by ending in second place, he “only” outperformed his seed by five expected placements. In reality, this undersells his showing even on a placement level. By entering grand finals from the winners side, he had a fundamentally different context to his placement. In the future, I may try to adjust our “seed performance rating” stat to be more about “sets away from winning the tournament.”
Seed Performance Rating of 6
- Soonsay at Big House 10
- Bananas at Big House 6
- Laudandus at Pound 2016
- S2J at Shine 2017
- Fly Amanita at Press Start
I talked about Soonsay, Bananas, and Laudandus last week, so no use repeating myself there. Meanwhile, S2J’s Shine run being here should be no surprise. Today, we know that he’s a contender for the backhanded title of “GOAT without a major,” but back at Shine 2017, S2J was only seeded 18th and hadn’t done too well earlier in the summer either. Lo and behold, he goes on to beat each of Duck, Shroomed, Plup’s Samus, and Mew2King, also beating Ice and having one of the most legendary post-set pop-offs ever.
Fly Amanita’s Press Start run, by this metric, is accurately captured as one of the most out-of-nowhere runs ever. While he was technically off a year where he finished at number 11, back then, the gap between players like him and the top echelon was seen as fairly large. A series of wild circumstances had to occur for this number 16 seed to end up in grand finals from the winners side. In no order: HugS had to beat Leffen in pools, Mang0 and Hax had to DQ, Leffen (with a lower seed in the final bracket) had to beat Mew2King, SFAT had to beat Leffen, Lucky had to beat Hungrybox, and Axe had to beat each of Westballz, HugS, and Lucky. For his part, Fly Amanita had to upset Shroomed, SFAT, and Axe, but these are obviously extreme circumstances. Honestly, forget any of Leffen’s groundbreaking victories. Forget about when Plup won Genesis 5. Press Start was the real end of the Five Gods Era. All three of the “gods” and Leffen had gone to a tournament and were in the losers bracket before the top eight. Okay, maybe that’s hyperbole. In fact, it totally is. Oh well. You get the point.
Seed Performance Rating of 7
- Javi at Apex 2012
- Nicki at DPOTG 2024
- Jmook at Genesis 8
Surprise, surprise – we have three of the greatest breakout runs in Melee history all here. I recapped them in my previous article, so I won’t go into detail here. But with that in mind, none of them have the top spot of our performances. That, my friends, goes to someone else.
Seed Performance Rating of 8
- ChuDat at Evo 2015
I miss Evo. I do not miss late bracket sign-ups that lead to players like ChuDat being practically unseeded. I do miss Evo though. Screw it – let’s beg for them to come back. If we have to, we’ll accept Zain vs. Cody in round one.
Takeaways
Have we finally reached the limitations of putting numbers to a greatest underdog list? No, obviously not. In next week’s column, I’m going to try my best to find a way to quantify “quality of wins” for underdogs. After that, I’ll then put this all together in one final graph that should hopefully be a starting point for anyone who wants to review this topic in detail.