Riot removes Mel's reflect mechanic and pushes her toward the support role

Jhin's fourth shot leaves the barrel, buries itself in Mel's chest, bounces back and punches through the marksman's own skull. In another game, Seraphine casts her ultimate into a teamfight, the orb sails through the enemy front line, comes home, and charms her own team. Both moments have lived in the League of Legends community's memory long enough to become memes, and long enough to become a reason people ban Mel in ranked queue. The day Riot Games announced the W rework, I reopened every note I had taken during nights watching Korean server games. Nothing surprised me. Only one thing is worth saying: Riot is deleting exactly the thing that made Mel hateful, and keeping exactly the thing that still makes her annoying. Mel shipped with a kit built around reflection. Her W did not block projectiles, it caught them and threw them back. As design, it is one of the most ambitious mechanics Riot has released in years: it rewards players who read tempo and anticipate a cast before it happens. As experience, it is a disaster. Jhin players learned their fourth shot could kill them. Seraphine players learned their engage ultimate could become a tool of self-destruction. Riot states plainly in the announcement that the team went through multiple rounds of adjustments trying to preserve the mechanic before concluding it could not be saved. That is the part most readers skim past. A mechanic abandoned after several tuning passes usually does not die because it is too strong, but because it cannot be subdivided any further. Reflection is binary: you catch the projectile or you do not. There is no grey zone for a five percent nerf, no coefficient to shave ten points of damage from. Riot had two options: leave it and accept that every game with Mel risks producing an absurd moment, or delete it. They chose the second. Patch 26.22 is where the change lands. According to the announcement, the build reaches live servers roughly three weeks after publication, a buffer long enough for Riot to collect test-server feedback and long enough for pro teams to prepare if they are still competing. Mel's new W does three things: it destroys incoming projectiles, it grants a shield to Mel and nearby allies, and it grants movement speed to those shielded. The reflect mechanic is gone. That is the largest change, and it alters the champion's nature rather than her power level. Reflection is a punishment tool. While Mel holds W, opponents must hesitate: cast and eat the return, or hold and lose tempo. She creates a psychological no-go zone around herself. A shield plus movement speed is a protection tool. It punishes nobody; it buys an ally two more seconds. The two serve different player types in different roles. The rest of the tuning pass makes the direction obvious. Mel's passive loses early-game damage and duration but gains health and armor scaling. Armor scaling on a mage passive is a strange signal: it does not help you win duels, it helps you survive a dive. Q is redistributed, with front-loaded burst reduced and later hits increased. The ultimate gets a longer cooldown and lower per-level scaling. E changes in the most interesting direction: root duration drops at close range and rises at maximum range. Add those four together and the message is clear: playing Mel up close is taxed, playing her far back is rewarded. That is the posture of a support or a long-range mage, not the posture of a mid-lane duelist. In my commentary work I have watched hundreds of games featuring Mel on Korean and Chinese servers. The good Mel players I met were not fast casters. They were the ones who held W and counted. They knew Jhin had two bullets left, knew Seraphine had just hit level six, knew how long the opponent would wait before pressing a button. That is a hard-read skill, and it does not transfer to a shield. The Mel player who was good yesterday starts over today. This is the harshest class of balance change. A numbers nerf leaves good players good, they just recalculate damage. A mechanic deletion vaporizes accumulated mastery. Riot did not take five percent of Mel's power; they took her entire signature skill. There is precedent: when Aatrox lost his revive in the major rework, the one-trick population lost the thing they had drilled for months and needed a long time to find footing again. But there is a point most online analysis gets wrong. Riot's stated goal is to limit situations that are too frustrating when playing against this champion. The new W still destroys projectiles. The structural frustration, the feeling that my spell did nothing, remains intact. What was removed is the spectacular frustration: my spell came back and killed my team. Riot is targeting the frustration that gets clipped, not the frustration that gets felt every game. And here is where data should appear but does not. The announcement cites no figures at all: no win rate, no pick rate, no ban rate, no frequency for how often reflect swings actually occur in a game. The examples given, Seraphine charming an entire team or Jhin eating his own shot, are rare and beautiful. They are excellent highlight material. They are not frequency data. A balance decision driven by clip salience rather than observed frequency carries a systematic risk of over-correction. I am not saying Riot is wrong. I am saying we are accepting a major design decision with no published quantitative evidence, and that is worth remembering when Mel shows up on ban lists again three months from now. People say I write to shock, but I only describe what they look away from. There is another layer few mention: timing. Riot normally freezes a patch for major international events, so a rework-level change announced roughly three weeks before it hits live most likely falls outside the active competitive window. If so, the competitive shock is deferred and every team gets a full preseason to test. That is the healthiest way to absorb a large change. Now the part where I might be wrong. There is a reverse reading, and it is not weak. Mel loses mid-lane power but may gain something more valuable in professional play: role ambiguity. A champion who can play both mid and support stops the opponent from locking their composition until you have committed. In pro draft, that ambiguity is worth more than a few dozen points of passive damage. An ally-facing shield plus movement speed is the defining signature of the enchanter class, and an E that pays for max range is an invitation to stand further back. If Riot genuinely wants Mel in the support role, this is a successful redesign, whether mid laners like it or not. I might also be wrong to weight frequency so heavily. Some mechanics appear once in twenty games and are enough to break faith with the game, and emotional balance is not a meaningless concept. Riot has internal data I do not. That fire taught me something: telling the truth burns, but burning is how you get light. My testable prediction: the first time Mel appears in a professional match after patch 26.22, she will be fielded as a support, not a mid laner. If her support win rate is higher than her mid win rate after two patches, this rework is a successful identity migration. If not, Riot has a harder question to answer: they deleted a champion's signature skill to solve a problem the data never proved existed. The ESFP in me works like this: feel first, explain later, and always be emotionally right.