Mel's W No Longer Reflects: Patch 26.22 Rewrites a Champion's Identity While the Numbers Stay Missing

Jhin's fourth shot was in the air. Crit damage, the last trade of the fight, the exact moment a marksman has finished doing his arithmetic — and the bullet came back into his own face. Mel's W. Seraphine's ultimate charm went the same way. The player who timed their combo correctly was the one who got punished. In Patch 26.22, Riot Games has removed that ability. W no longer returns enemy projectiles; it destroys them, and in exchange grants Mel and nearby allies a shield and movement speed.

Riot itself admitted reflect was Mel's 'signature characteristic.' They tried to preserve it, failed, and removed it anyway. That sentence is the most honest part of the patch notes and the most troubling.

I have not a single number to measure this change against. No win rate. No pick rate. No ban rate. Three thousand eight hundred shots in, the model finally opened its eyes — here the model was given nothing to open.

Context: W was the ability Mel was built around

Mel is a mid-lane mage, and her entire draft value rested on one question: is the enemy combo projectile-based? If yes, she was a counter-pick you could not afford to ignore. The reflect mechanic flipped the moment an opponent committed their strongest asset. It was not a defensive tool. It was retaliation built out of the enemy's own damage.

Riot has a long habit of pruning mechanics of this type. Old Poppy, old Yorick — anything that punished the opponent for playing their own champion correctly became a focal point of solo-queue resentment and eventually got cut. Mel's W belongs in that lineage. Riot used the same language it uses in every rework: reducing frustration, improving the experience.

Patch 26.22 is not confined to W. There is a simultaneous numerical pass across passive, Q, E and the ultimate. The passive loses early damage and duration but gains health and armour scaling. Q's damage is shifted from the first explosion into subsequent hits, with a slower projectile. E's root is now range-dependent — shorter at close range, longer at maximum range. The ultimate is cut as well. Launch is expected roughly three weeks after the announcement.

There is only one way to read this package: total damage is down, utility is up, and the consequence is that Mel's role is changing — from burst counter-pick to sustain/enable mid. This is not number tuning. It is an identity transfer.

W: reaction replaced by assistance

The new W does three things — destroy the projectile, shield Mel, and grant movement speed to Mel and nearby allies. None of those three damages the opponent. The old W returned enemy damage to them; the new W simply stops that damage and pushes your own team forward. An ability that was an answer is now a preparation.

The ally movement-speed buff is the part patch notes explain least and that will matter most. In solo queue it does little on its own. In a coordinated draft it can rewrite an entire team's engage-disengage rhythm. Mel's ceiling has dropped, but her floor has risen — a change that is bad for solo-carry specialists and good for team-oriented mid players.

Q: a tax on guaranteed burst

Moving damage off the first explosion into later hits, and slowing the projectile, both point the same way. Damage that was previously unconditional now depends on tracking and aim. Calling this a nerf is half the truth. It is skill-indexing — it decides which Mel players are still valuable.

E: the range tax

Tying root duration to range is the cleanest design statement in the patch. A panicked close-range E now roots less; a measured max-range E roots more than before. This is Riot saying directly that Mel players will have to learn spacing, or they will feel nerfed.

Passive: the build path moved

Cutting early damage and duration while raising health and armour scaling is not cosmetic. It quietly points at a build path — tankier items are no longer wasted on Mel. And where a build path shifts, the first patch is almost always mis-tuned. My bet: at least one hotfix follows 26.22.

Who gains, who loses

The winners are obvious — Seraphine, Jhin, and by extension the projectile-reliant poke-crit marksmen in the Ezreal and Caitlyn class. Teams that previously had to bend their draft around the threat of Mel now have one constraint removed. That is not a small effect: when a counter-pick disappears from the market, the price of the entire mid-lane pool gets re-set.

The losers are equally obvious — Mel one-tricks and Mel specialists. For players whose rank rested on that single pick, this is a champion-level ageing cliff. The ability was not deleted, it was altered, but the tool they picked Mel for is gone from their hands.

One thing the patch notes do not say, and I will. This was not driven by pro play. Mel is not named in any professional context. This is a solo-queue frustration-metrics decision — ban rate and 'playing against this feels pointless' complaints, the classic precondition for any mechanic removal. Acknowledging that clarifies both the purpose of the patch and its limits.

The part nobody is writing: a new frustration

A shield and movement-speed aura can produce a different kind of irritation. People were annoyed by reflect because their own damage came back at them. People may be annoyed by the new kit because Mel saves her team and they escape. The first is 'I was punished.' The second is 'my work was wasted.' Those are different emotions, and Riot has only measured the first one so far.

The contrarian read: 'identity preserved' is the weakest sentence in the report

The framing is that these changes preserve Mel's identity. That is the least-supported claim on the page. Riot itself wrote that reflect was her signature characteristic. When you remove a champion's signature thing and hand back a shield, you have not preserved the identity — you have replaced it. Calling that preservation is like changing every parameter in a model and insisting it is the same model because the filename did not move.

The second soft spot is the relief narrative. There is no win rate, no ban rate, no pick rate in the notes. We are told the game is less frustrating now, without the number that would prove it. That is data-light relief. Emotion-led design decisions are often right in direction and wrong in magnitude — and magnitude is exactly what the next patch decides.

Third, rework churn. A patch that moves four or five abilities at once almost never lands correctly on the first pass. My pre-registered call: within two patches of 26.22 going live, Mel receives at least one adjustment, and there is a real chance a lighter version of reflect returns — perhaps only on the ultimate, perhaps only for a limited window. Riot has removed identity mechanics and half-restored them before.

I should also file the tournament impact as unconfirmed. Patch 26.22 arrives roughly three weeks after announcement, and if it lands near a major competitive window it creates a practice-patch versus tournament-patch divergence. Riot generally keeps rework-level changes out of the active tournament patch, so professional impact will lag live-server impact by a cycle. That is an inference, not a fact, and I will label it as one.

One more thing, from experience. In 2026, inside Bengaluru FC's three-person analytics unit, I hand-tagged 3,800 shot events across two ISL seasons. Two days before the final I showed the coaching staff that our press had collapsed from 8.4 to 12.1 across the last six games. I was told to stop chasing ghosts. The thing was correct in my spreadsheet and useless in their week. Mel is the same lesson: a number you do not have is easy to win an argument with and impossible to change. Riot has won the argument because no number is on the pitch.

Which is why I do not read this as a balance story. It is a design-governance story. The publisher publicly acknowledged it is removing a champion's core identity, and that decision was weighted toward the opponent's experience rather than the player's attachment. That stance may even be correct — but selling it as 'identity preservation' is where the notes broke their own clarity.

And a detail that is easy to skip: this coverage comes from a Vietnamese source, with LMHT branding and a localised name for W. It is not a regional meta story. It is global patch notes, localised. Anyone claiming Mel is prioritised more in one region has no data, and neither do I.

A model is a promise you renew weekly, not a chart you publish once. I trust the shot map over the highlight reel and I can show you why — this time the shot map is blank, so I will say the true thing: the direction is clear, the magnitude is unknown.

What to watch next

I will track three numbers and grade myself publicly after two patches. First, Mel's win-rate delta post-26.22 — if it falls more than two percentage points, the rework overshot. Second, ban rate — if ban rate does not fall, reflect was not the problem, her whole kit was, and Riot cut the wrong limb. Third, the real value of the shield aura in teamfights — because if that value runs high, the next complaint will be 'Mel just saves her team,' and Riot will have to make the same decision it made today, again.

The ban never touched my models. It touched my invoices. The same is happening here — removing reflect did not break Mel's models, it broke her players' confidence. And confidence is the one metric that never shows up in a patch note.