Mel and Patch 26.22: When Riot Removes the Reflect Mechanic to Save the Player Experience

In a solo queue ranked match on the Korean server at the end of October, I witnessed a teamfight that, if clipped for a highlight reel, would have made the entire arena erupt. The opponent locked in Seraphine, cast her ultimate Encore across our whole formation, and Mel — Riot's newest mage — raised her W to reflect the entire thing. Seraphine's strongest ultimate bounced back, charming her own teammates. Within thirty seconds, the game flipped, and in the chat window, both sides typed lines that cannot be repeated. That was the moment that forced Riot to act. According to information from the PBE test server, Riot Games has decided to rework Mel's W ability — described by the design team itself as the most frustrating ability in League of Legends — in patch 26.22. The reflect mechanic is removed entirely, replaced by the ability to destroy incoming projectiles along with a shield and movement speed for nearby allies. Not only the W, but Mel's entire kit is being adjusted: passive, Q, E and R. My first reaction to the news was simple. This is not an ordinary damage nerf. This is an identity migration. Context: From Signature Mechanic to an Unbalanceable Problem Mel launched with a W designed as a punishment tool. She raised it, and every incoming projectile was reflected back at its source. In theory, this was a beautiful mechanic: it rewarded reading the situation, demanded precise timing, and created moments where the strong shot themselves in the foot. In practice, it was a balancing nightmare. The problem was that this mechanic is binary. Either Mel raised W at the right moment and reflected successfully — in which case her value spiked exponentially based on the skill of the strongest opponent. Or she mistimed it and W was just a small shield. There was no middle ground for Riot to tune. You cannot reduce reflected damage by ten percent, cannot increase cooldown by two seconds, cannot limit the reflect window to three-tenths of a second. The mechanic has only two states: on or off. Riot admitted in the announcement that the design team tried to preserve this mechanic through multiple rounds of adjustment before giving up. This is the single most important detail of the entire announcement, and it was almost entirely overlooked in the short news pieces. When a design team abandons a signature mechanic after multiple attempts to balance it, they do not do so out of laziness. They do so because the mechanic is structurally unbalanceable. You can balance damage. You can balance cooldown. You can balance range. You cannot balance a mechanic whose power depends entirely on whether the opponent chooses to throw an attack. This made Mel an uncontrollable variable in both solo queue and competitive play. In solo queue, she created games where players felt helpless: you played well, you locked the right target, you cast the perfect ability, and you were punished for it. In competitive play, she was a counter-pick too powerful to appear often, and too dangerous to ignore entirely. Data tells the story that media does not have the patience to hear. And in this case, Riot themselves provided no data. Core Analysis: The Structure of the Rework Patch 26.22 changes Mel across three layers, and I will go through each one. The first layer is the W — a rework-level change. The reflect mechanic is removed. In its place, W now destroys incoming projectiles, meaning the basic defensive function is retained, and grants a shield plus movement speed to Mel and nearby allies. This is a change of category, not of number. Mel shifts from a punishment tool to a protection tool. She no longer makes opponents afraid to cast. She makes allies safer when opponents cast. The second layer is the passive. Early-game damage and effect duration are reduced, but health and armor scaling per level is increased. This is the signal I want you to pay closest attention to: a mage with an armor-scaling passive is extremely unusual in League of Legends. It does not point toward winning duels. It points toward surviving dives. This is the mark of a unit designed to endure, not to dominate. The third layer is Q, E and R. Q shifts damage from the first hit to subsequent hits, reducing instantaneous burst. E reduces root duration at close range but increases it at maximum range. R increases cooldown and reduces per-level scaling. I want to pause on the E change, because it is the clearest signal of design intent. By rewarding longer root duration at range and taxing it up close, Riot is paying players to maintain distance. This is a mechanic that encourages playing Mel from the back line, and that is the posture of an enchanter or support, not of a mid-lane duelist. Combining the three layers, I read a clear direction: Riot is pushing Mel toward the support role. The shield plus movement speed for allies is the signature of the enchanter class. The armor-scaling passive is a defensive signal. The E rewards long range. All of it points in the same direction. This is not a nerf to make Mel weaker. This is a role shift to make Mel survive. This is a smart move from a design standpoint. Mel retains the ability to destroy projectiles, so she remains a useful pick against compositions built on skillshots. She gains team value. And she is no longer a binary variable in teamfights. In theory, this is how you handle an unbalanceable mechanic: stop trying to balance it, and replace it with something that can be balanced. Contrarian Angle: The Problem Is Still There But this is where I want to push back on Riot's decision, and I will state my confidence level clearly: about seventy percent. Riot states the goal is to limit situations that feel too frustrating when playing against this champion. But Mel's W still destroys incoming projectiles. This means the frustration of my ability did nothing at all remains fully intact. What has been removed is only the frustration of my ability killed my own teammates. Riot is targeting the spectacular frustration, not the structural one. Take the two examples Riot themselves gave to illustrate the frustration. First: reflecting Seraphine's ultimate to charm the entire enemy team. Second: reflecting Jhin's fourth shot, turning his critical strike into a critical strike against himself. Both are dramatic, rare and highly memorable situations. But Riot provided no data on how often such reflect moments actually occur per game. No percentages, no averages, no sample. Balance decisions driven by impressive clips rather than frequency data carry a systematic risk of over-correcting. This is the biggest blind spot of the entire process. I understand why Riot did this. A mechanic that produces memorable moments spreads further than one that produces quiet frustration. But player frustration usually comes from things that repeat game after game, not from edge moments. If you get your ultimate reflected once in fifty games, you will remember it, but you will not quit the game over it. If your abilities are blocked constantly in every teamfight, you will quit the game. And there is an internal contradiction in Riot's announcement that I need to point out. Riot says the changes will make Mel more balanced while still preserving her unique identity. But the unique identity Riot described is precisely the reflect mechanic. You cannot both delete a mechanic and preserve it. These are two conflicting statements, and I assess the confidence of this contradiction at about ninety percent. State never stands still, only the observer changes the viewing angle. Riot is changing their own viewing angle on Mel, and they are calling it identity preservation. Impact on Players and the Market For dedicated Mel players, this is bad news in its harshest form. Not a number nerf you can adapt to. This is a skill-obsolescence event. The entire skill of reading when to raise W, which the best players built over countless games, becomes worthless. That skill does not transfer to the new kit. You cannot use reflect reflexes to play a champion that hands out shields. When it comes to competitive adaptability, the picture is more complicated. On one hand, Mel loses her solo-lane power and early skirmish window, the phase where she was strongest before. The early passive is nerfed, the R is nerfed, the Q is redistributed. On the other hand, a shield plus movement speed for allies opens up a genuinely flexible support option. In professional draft structures, a champion who can play multiple roles has outsized value because it delays role reveal until the opponent has committed. Losing solo-lane power while gaining role ambiguity can be a net gain for professional viability. This is a counter-intuitive but structurally sound read, and I assess it at about sixty percent confidence. On timing, the patch is expected to arrive about three weeks after the announcement. If so, it lands in the late-season window, likely after the patch lock for the World Championship and before the following spring split. If that holds, professional teams gain an entire preseason to adapt, which substantially reduces the shock of a rework-level change. I assess confidence here at only about forty percent, because the patch date and the 2026 event calendar are unverified. What I can state with certainty is that this patch is a test of Riot's design philosophy. If Mel becomes a viable support and her pick rate stabilizes without generating a new wave of complaints, Riot will have more evidence that sometimes you must abandon a signature mechanic to save the overall experience. If she disappears from every rank tier after two weeks, Riot will face a harder question: whether they are optimizing for clip viewers instead of actual players. Success on the arena is written in highlight moments, but its cost is written in other numbers. Progressive Thought Looking back at this whole story, I see three layers of problem. The first layer is the specific story of Mel: a champion with a beautiful mechanic that could not be balanced, and Riot finally choosing to abandon it. The second layer is the question of design methodology: how do you balance a mechanic whose power depends on opponent behavior. And the third layer is the question of data: whether balance decisions should be driven by impressive moments or by actual frequency of occurrence. For Mel, I think her best chance is the support role. Not because that is where she is strongest, but because that is where she can exist without creating frustrating games. A champion who no longer produces lethal reflects can still be a good pick if she provides enough team value. But for that to happen, Mel players need to accept that they are playing a new champion with the same name. And perhaps that is the larger lesson. In a game run as a service, no identity is permanent. The reflect mechanic appeared, created miraculous moments, made players fall in love, and then was deleted when it proved it could not survive the balancing system. What remains is the question every game designer must face: do you build a champion's identity around a mechanic, or build the mechanic around an identity? Riot answered. Their answer is that identity can change. And I think that is the right answer, even if it forces a community of players to start over. Because a game cannot live on the miraculous moments of a few players; it must live on the stable experience of the many. The remaining question is not whether Mel keeps her reflect mechanic. The remaining question is whether Riot dares to abandon other mechanics they are still trying to balance with small numbers, while the real problem lies in the structure.