I have played Apex Legends since the early days, and I have watched the ALGS meta cycle through the same faces more times than I can count. Wraith, Valkyrie, Gibraltar. Wraith, Valkyrie, Gibraltar. Caustic, Valkyrie, Gibraltar. If you have followed competitive Apex at all, you know the rhythm. So when I first saw a machine learning study claim that Fuse — yes, Fuse — was one of the best-performing Legends at the highest level, I laughed. Then I looked closer. In 2026, with the ALGS still evolving and balance patches constantly reshaping the battlefield, that old finding feels less like a joke and more like a warning.
The study came from an Apex player known as WavyKirk. They built something called the Kirk Algorithm, a modified version of the Colley Matrix — a system originally used for college football rankings — and applied it to ALGS tournament results. The goal was simple: figure out which Legends actually perform well when the stakes are highest. The results were strange. Gibraltar, despite near-100% pick rates in North American tournaments, offered poor value to teams. Bubble fights, where squads strafe in and out of his Dome Shield, were often not worth taking because third parties would clean up the winner more often than not. Meanwhile, Fuse — the moustachioed Australian bombardier — quietly posted strong numbers.
I know how that sounds. For years, Fuse was the fun pick, not the smart pick. His tactical, Knuckle Cluster, fires a handful of micro-nades that explode after a short delay. His passive lets him launch regular throwables farther and faster. His ultimate, Motherlode, surrounds enemies in a ring of fire that slows them, deals damage, and highlights them with Bloodhound-style wallhacks. On paper, that is a fantastic kit. In practice, the aiming was awkward, the ring was easy to escape, and most pros ignored him. He was a pub-stomping legend, not an ALGS legend.
But the Kirk Algorithm asked a different question. Instead of looking at pick rate, it looked at performance. One commenter on Reddit asked whether Gibraltar players simply had a low Elo average, and whether any character had a predominantly high Elo average. The answer was Fuse. That does not mean Fuse was secretly broken. It means the players who stuck with him — especially at higher levels — were getting more out of him than the community assumed.
The pro scene gave us a few hints. In ALGS Season 2, Elwin 'KSWINNIIE' Echeveria experimented with Fuse. Furia's Josh 'mrhackulo' Nguyen picked him regularly, won a match with 2,000 damage on the character, and helped his team qualify for Playoffs. Furia finished 15th out of 40 Pro League teams, which is not a trophy, but it is not nothing. Remember, that qualification came with a caveat: Furia would not have made the January tournament if it had been the planned in-person LAN. Still, the fact that Fuse was on the field at all was notable.
The study also found that the gap between controller and mouse-and-keyboard players was smaller than many people think. It crowned a surprising pro as the best of them all. Phillip 'ImperialHal' Dosen did not even make the top ten, and Konstantin 'Hardecki' Kozlov, despite his incredible Wraith plays, came eighth. That context matters. The Kirk Algorithm was not just a Fuse propaganda piece. It was a broader challenge to the assumptions that drive the ALGS meta.
| Legend | Common competitive reputation | Kirk Algorithm takeaway |
|---|---|---|
| Wraith | Elite, always safe | Still a top-tier pick, but not untouchable |
| Valkyrie | Rising star, great rotations | Strong, but not the only answer |
| Gibraltar | Must-pick defender | Poor value despite high pick rate |
| Fuse | Fun pub pick | Surprisingly high Elo average |
| Loba | Niche utility | Showed potential in EMEA Championship |
That table is a simplification, but it captures the tension. The ALGS meta can feel stale because teams are risk-averse. There is real money on the line, and no one wants to be the squad that throws a tournament because they tried a meme pick. I understand that. I have watched enough ALGS to know that consistency wins championships. But I also watch weekly tournaments and show matches, and that is where Fuse makes sense. Smaller events, early rounds, and experimental compositions are the perfect proving ground for an underdog like him.
In 2026, the conversation has shifted again. New Legends, reworks, and balance changes have moved the meta, but the core lesson from WavyKirk's study still holds: pick rate is not the same as value. A character can be popular and overrated. Another can be ignored and underrated. Fuse is not going to replace Wraith or Valkyrie overnight. His rise is unlikely to be as meteoric as Valkyrie's was. But I would love to see more teams test him in ALGS Split 2, weekly tournaments, or even a bold show match. If he is competitive, that is an added incentive. If he is fun, that is already enough for me.
I hope Fuse brings some chaos to the professional scene. Maybe one day we will see a team running Fuse, Loba, and Ash at the ALGS Championship. I will keep my fingers crossed. Until then, I will keep dropping Motherlode on unsuspecting squads, laughing like a maniac, and reminding my friends that the data was on my side all along.