Mel loses her reflect in patch 26.22: Riot treated the symptom and left the disease

Jhin's fourth shot left the barrel, touched Mel's reflect field, and snapped back into Jhin's chest. One bullet, two deaths. I have seen that scene hundreds of times, and every single time the crowd in the livestream made the same sound — the sound of people who just watched a scam too beautiful to be angry about. I take notes on every game I watch. Not the score, the timing. Which second Mel held her W, what she was waiting for, whether the opponent realised they were being baited. After a few hundred games, a pattern emerged that was uncomfortable in how clear it was: Mel's value on the map was never in the damage she dealt. It was in the fact that she made other people afraid to press a key. Riot Games has just announced the removal of exactly that mechanism. Patch 26.22, scheduled to hit live servers roughly three weeks after the announcement, converts Mel's W into an entirely different tool: destroy incoming projectiles, grant a shield, and grant movement speed to nearby allies. Passive, Q, E and R are all adjusted alongside it. Five ability slots, five touches. This is the scale of a rework, not a numbers pass. And Riot describes it as a way to keep Mel more balanced while preserving her unique identity. I read that sentence three times. What identity is being preserved? For years, Mel was defined by one sentence: she punishes. Not sustains, not heals, not roots. Punishes. You shoot her, your shot turns around and kills you. That mechanism is provocative by design, and it forces opponents to play differently — not by dodging skillshots, but by thinking about whether to press a key at all. That kind of pressure is extraordinarily hard to balance. It is binary: either the reflect lands and you win the exchange, or it does nothing and you have wasted an ability. There is no grey area to tune. Riot says it tried multiple rounds of adjustments before giving up. A design team does not discard a signature mechanic after repeated tuning failures unless the mechanic is structurally untunable — and reflect is exactly that shape of mechanic. Silence is never a victory, only extra time before the collapse. Multiple rounds of adjustments, no announcement, then a sudden deletion. That is how design teams admit defeat without saying the word. What was removed and what was kept Read the change list closely and one detail stands out above all the others: the new W still destroys projectiles. Which means that if you fire an ability at Mel and she presses W, your ability still vanishes. You still lose the mana. You still lose the cast. The only difference is that this time your ability does not turn around and kill you, and Mel gains a shield plus a bit of movement speed. If Riot's goal was to limit the situations that feel too frustrating when playing against this champion, they have completed half the job. The frustration has two layers. The spectacular layer — my ability killed my own team — is gone. The structural layer — my ability does not exist — remains intact, and is arguably worse in one respect: previously, if Mel mistimed her W, she lost the ability and was punished. Now, pressing W grants a shield and movement speed regardless of whether you fired anything at all. The punish tool is dead. What replaced it is a low-variance survival tool — and in professional play, low variance always frustrates for longer than high variance. That is the part most commentary is missing. People look at the loss of reflect and conclude Mel got weaker. Mel did not get weaker. Mel changed profession. Mel is being pushed further back, and that is deliberate Two small changes in the patch notes tell the same story. First, E: root duration reduced at close range, increased at maximum range. That is an extremely clear signal. Riot is paying Mel players to stand far away. Close range is taxed, long range is rewarded. Second, passive: early-game damage and duration down, but health and armour scaling up. A mage with armour scaling on a passive is unusual. It does not say you will win duels. It says you will survive being dived. Combine those two with the new W — shield and movement speed for allies — and the picture is fairly clear. The entire kit is being redesigned around the posture of a support: stand behind, hold distance, peel, and hand tools to other people. Mel's mid-lane identity had been contested for a long time. It is now being replaced by a different identity, tidier, easier to balance, and less controversial. At the same time, Q is being redistributed: front-loaded burst damage is pushed into subsequent hits. In plain language: to reach her old damage, Mel has to stay in the fight longer. She is no longer the fire-one-shot-and-retreat champion. And R: cooldown up, per-level scaling down. That is the clearest cut at her strongest window. Mel used to peak in the early and mid game — using the ultimate to force fights, using W to break the opponent's engage. Both legs have been sawn off. If she is to remain relevant in professional play, her value must migrate to the late game, where shields and movement speed are what actually cost money. There is one cohort that loses in a way no statistic can measure: the players who spent hundreds of hours learning to read reflect timing. That skill does not transfer. You cannot use reflect reflexes to shield an ally. This is the harshest class of balance change: not a numbers nerf that players can adapt to, but a skill-obsolescence event. The entire accumulated stock is written off in a single patch. Riot's examples are beautiful examples, not common ones Riot offers two illustrations of the frustration: Mel reflecting Seraphine's ultimate and charming the entire enemy team, and Mel reflecting Jhin's fourth shot back into Jhin himself. Both are magnificent moments. Both are extremely rare moments. The balance report offers no figure whatsoever on how often such sequences occur per game. No win rate, no pick rate, no ban rate. Seventeen information points, and not a single performance statistic for Mel. That is the most worrying thing in this entire story — not because Riot lacks data, but because the decision was communicated through memorable moments rather than through frequency. The new meta lives where people are afraid of losing something, not in the strategy. What is Riot afraid of losing? The clips. The moment a player posts online asking why this game allows this. Riot is not afraid of the average number. I understand why. A reflect sequence is worth more than ten soppy analysis pieces, in reach terms. But balancing a game by reach always leads to overcorrection. Where I could be wrong I have been wrong often enough to know that a good argument is not the same as a correct one. So here is where I could get my face broken. First, I am assuming the new W is more frustrating than the old one. That assumption rests on my feel for exchanges, not on data. There is a real possibility that most players do not encounter Mel often enough to feel the difference, and that measured frustration actually falls because opponents are no longer killed by their own abilities. Second, I am assuming Riot has no frequency data. They almost certainly do. Not publishing it does not mean not using it. If internal data showed reflect produced one decisive sequence in thousands, then removing it for emotional-experience reasons is still defensible, and my fixing-by-clip argument collapses. Third, my support-pivot read could be entirely wrong in direction. Shield and movement speed for allies is not necessarily a support tool — it could serve a mid laner who enables a diving jungler. I am reading the kit through the most familiar template, and the most familiar template is not always the correct one. Fourth, for professional play this could actually be good news. A champion who can be drafted mid or support is an extremely valuable flex piece, because it defers the role reveal until after the opponent has committed. Losing solo-lane power while gaining role ambiguity is a trade many teams will happily take. Fifth, timing. Patch 26.22 lands in the late-season window, most likely after the World Championship patch lock and before the following spring split. If so, professional teams get an entire preseason to adapt. The competitive shock of a rework-level change is spread out, and spread out usually does not hurt. What I will be counting I do not trust predictions that cannot be checked. So here are three things I will count, and you can count them with me. One, how often Mel appears in professional drafts as a support versus as a mid laner during the preseason. If support outpaces mid, the new identity has taken hold and Riot wins. Two, how often Mel's new W saves an ally from a lethal blow, per game. If that number is far higher than the number of times the old reflect produced a swing, then Riot traded a rare mechanic for a common one — a good deal, even if it is not a pretty one. Three, the community reaction after two weeks. If the complaints persist, my structural-frustration hypothesis holds. If there is silence, I am wrong, and I will say so. In the first half people laughed at me; in the second half I laughed at the whole match. But there is no second half here yet. There is one patch, one champion stripped of the exact thing that made people remember her, and a game company betting that memory fades faster than habit. Mel no longer reflects. She has a shield, movement speed, a stronger root at range, and a weaker ultimate. That is a different champion. The only question left is whether players will invest in learning her from scratch, or simply pick someone who was never called the most frustrating champion in the game — and that is the number Riot is actually waiting on.