Patch 26.22: Riot Games Removes Mel's Reflect Mechanic, Opening a Support Flex Path

In the 26.22 patch notes, Riot Games wrote that the team had gone through multiple rounds of adjustments to preserve Mel's reflect mechanic before deciding to remove it. That moment — a short announcement on the live client — marked the first time one of League of Legends' most frustrating abilities was replaced by a different mechanic rather than having its ratios reduced or cooldown increased. Mel's W no longer reflects; it destroys projectiles and grants a shield and movement speed to nearby allies. The story is this: Riot did not reduce Mel's power, they changed her profession. Mel was released with the identity of a mid-lane mage who could punish opponents based on reflexes. W was the centerpiece: it caught all projectiles for a duration, then fired them back. In theory, this is a high-skill ability: read the timing, hold W for the right moment. In practice, it created two kinds of frustration. The first: an opponent uses a spell on Mel and that spell comes back to kill them. The second: Mel reflects a major spell and the entire enemy team is hit by the effect. Riot cited two examples: reflecting Seraphine's ultimate to charm the entire enemy team, and reflecting Jhin's fourth shot back at Jhin. These are high-salience moments, easy to clip, easy to argue about. But they are not frequency data. The source article provides no win rate, pick rate, ban rate, or occurrence count for these situations. That is the largest gap in the entire story. Riot says the goal is to “limit situations that are too frustrating when playing against this champion.” But the new W still destroys projectiles. That means the feeling of “my spell did nothing” remains. Only the feeling of “my spell killed my team” disappears. This is the key point: Riot is fixing the loud part, not the structural part. The loud part is the clip of Seraphine turning an entire team around. The structural part is that a defensive ability can completely invalidate a class of champions. If you are Jhin, you cannot fire your fourth shot into Mel. If you are Seraphine, you cannot cast your ultimate into Mel. That is an imposition on opponent design, not a beautiful outplay. Riot removes the “reflect” part but keeps the “destroy” part, so pressure on projectile-reliant champions remains in a lighter form. This is why I read this patch as an identity migration, not a nerf. The new W shifts from a punish tool to a peel tool. Shield and movement speed for nearby allies is the signature of the support class. E reduces root duration at close range and increases it at max range. This is the clearest signal: Riot is paying players to maintain distance. You stand far, you are rewarded; you stand close, you are punished. That is the posture of a support mage, not a mid-lane duelist. Passive reduces early damage and duration but increases health and armor scaling. A mage with an armor-scaling passive is a defensive signal: she needs to survive dives, not win duels. Q redistributes front-loaded burst into subsequent hits. This reduces lane bullying and increases sustained damage. R increases cooldown and reduces per-level scaling. This reduces the value of using the ultimate to start fights or pick off targets. Taken together: Mel's power moves from early-mid to late, from burst damage to teamfight utility. She is no longer a champion who amplifies lane advantage; she becomes a champion who amplifies team composition. What does this mean for dedicated Mel players? This group built its skill on reading when to hold W. They did not learn to stand behind, they learned to stand in front. They did not learn to peel, they learned to punish. When W loses reflect, that skill does not transfer. This is the harshest kind of change: not a ratio nerf, but a skill-obsolescence event. You can adjust ratios and players keep the feel; you delete a mechanic and players lose the language. I have seen this in traditional sports: when an athlete builds a career on a technique that gets banned, no training drill compensates. In League of Legends, it is like a racer losing their home track. But there is another side: flex value in professional play. Shield and movement speed for allies, plus E being rewarded at max range, opens genuine support viability. A champion who can go mid or support is an asset in draft because the opponent must wait to see where you place her. In BO3/BO5, flex picks are more valuable than in BO1 because you can hide the role across games. Losing solo-lane power while gaining role ambiguity can be net-positive for pro viability. This is the counter-intuitive read: Riot may not be killing Mel, they are moving her to a role where she is less frustrating and easier to draft. However, there is no post-patch data to confirm. We can only say: if she appears in scrims as a support, the patch succeeded; if she disappears, Riot removed a frustration without replacing a function. Riot says it tried multiple rounds of adjustments to keep the reflect mechanic. This is an important detail. Design teams do not discard a signature mechanic after multiple tuning passes unless that mechanic is structurally untunable. Reflect has binary power: either you reflect, or you do not. There is no middle scale. You cannot half-reflect. So every balance attempt falls into increasing or decreasing duration, increasing or decreasing reflected damage, or changing trigger conditions. Those changes do not solve the structural problem: the opponent still loses a spell. Riot concluded that this mechanic cannot be balanced in a game where projectiles are a large part of the ability system. This is my inference, not Riot's statement. But it fits with keeping “destroy” and removing “reflect.” Destroy can be balanced with cooldown and range; reflect cannot. On timing, the article says patch 26.22 will arrive about three weeks after the announcement. Riot typically ships patches every two weeks. If correct, 26.22 lands in the late-season window, possibly after the World Championship patch-lock period and before the following spring split. This reduces competitive shock: teams have the entire preseason to adapt. But this is a low-confidence inference because the 2026 event calendar is not verified in the article. If the patch lands mid-split, the impact is completely different. A rework-level change mid-season can distort a week of competition; a preseason change is just one variable in a training plan. I do not draw a conclusion on timing without an official calendar. The common read is that Mel was heavily nerfed and lost her identity. I disagree with that reading. First, Riot did not reduce overall power; they moved power from a reflex ability to utility. Second, Mel's identity is not only reflect. She also has range control, sustained damage, and lane pressure. Removing reflect does not erase the entire identity, it erases an extreme expression of it. Third, if you look at E and passive, you see a champion designed to survive and peel, not to dive in. This may be a more coherent direction for a long-range mage. There is, of course, a contradiction: Riot says it wants to “preserve her unique identity” while removing the very mechanic described as her “signature trait.” Those two statements cannot both be true. If identity is reflect, removing it removes identity. If identity is range control, reflect was only one expression. Riot chooses the second definition. Players have the right to disagree. But as design, the second definition is more flexible and easier to balance. Riot's examples are salience-selected, not frequency-selected. Seraphine reflecting an entire team and Jhin's fourth shot returning to Jhin are beautiful moments to tell stories about, but they do not represent thousands of games. If Riot balances based on clips, it risks over-correcting. This is why I always ask: where is the frequency data? The article has none. No win rate, no pick rate, no count of successful reflects per game. No data, no conclusion. I can say this patch changes identity, but I cannot say it makes Mel stronger or weaker. That is a limitation that must be stated clearly. I have followed League of Legends patches since 2026, when I was still a player and tournament organizer. My experience is that the most controversial changes are often not the largest ones, but the ones that delete a skill players have invested in. Players can accept damage reductions. They struggle to accept losing a mechanic. World Cup 2026 taught me that the scoreboard does not know how to play football; in League of Legends, the scoreboard does not know how to time W either. So when evaluating this patch, I do not look at ratios. I look at whether players still have a reason to learn Mel. If the only reason was reflect, Riot just removed that reason. If the reason is range control and peeling, Riot just opened a new one. I write documentaries to answer questions, not to confirm answers. With patch 26.22, the question is: can a champion trade identity for survival? Riot is betting yes. They are betting players will accept a quieter Mel, with fewer clips, but more stable in draft. They are betting professional teams will see the value of a support flex pick more than a frustrating counter-pick. And they are betting that the structural frustration — my spell did nothing — is acceptable if the loud frustration — my spell killed my team — disappears. That is a design gamble, not a nerf. The thing to watch is not patch day, but the first week of scrims. If Mel appears as a support, Riot won: they converted a frustrating champion into a functional one. If Mel disappears, they lost: they removed a frustration without replacing the value. Between those two outcomes is a data gap — where balance decisions are made from clips instead of frequency. The missing footage always contains something someone does not want us to know. Here, the missing footage is data on how often reflect actually happened. Without it, we can only say: Riot changed Mel's profession. Whether she does the new job well, the scoreboard will answer later.