Mel's Reflect Is Gone in Patch 26.22: Riot Rewrote Her Identity and Published No Metrics

Hook: Two Seconds, Two Clips, and a Champion's Entire Draft Value

It was two in the morning in Seoul and I was running my ranked replay for the third time. Mel was on mid. The enemy Seraphine pressed ultimate — Encore — and the charm arc bent back into her own team. In the same teamfight, Jhin's fourth shot crit, and that crit landed on Jhin's own head. Inside two seconds, Mel had returned the enemy's two most expensive cooldowns to their owners. Those two seconds were her real draft value — the matchup people blind-picked her for.

I saved the clip. In the Patch 26.22 notes, that mechanic was removed. The clip is now the last record of an ability that Riot Games itself called Mel's "signature characteristic" — and then deleted anyway. I opened the replay file, and the headline collapsed in four minutes.

Context: What the Source Is, and What It Does Not Confirm

These patch notes came through a Vietnamese-language localisation — LMHT, League of Legends, with the reflect ability logged as "Phản Quang Chiết Kính." One thing is settled immediately: this is a global balance news item, not a regional competitive story. There is no team, no tournament, no pro player, no prize pool in it. Anyone converting this into a "the meta has changed" headline will be unable to say which meta, in which region, on what sample size.

Three things I flag up front, because my habit is to open the primary record before trusting a headline:

1. Patch number: 26.22. Riot's season-based numbering implies a 2026 patch, but the source carries no independent confirmation. The number is kept exactly as given.
2. Timeline: the patch goes live roughly three weeks after announcement. No exact date appears in the source.
3. Data: win rate, pick rate, ban rate — not one figure is present.

The third point carries the most weight, and the entire remainder of this piece rests on it. Three ability nerfs, one mechanic removal, and zero baseline. Without data, the magnitude of this change cannot be established — only its direction.

Mel is a mid-lane mage whose draft value rested primarily on the reflect. That mechanic naturally sits in the "anti-fun" category: the opponent's burst or crowd control returning to its sender. Riot's history is one of pruning that category — the old Poppy and old Yorick-style reworks followed the same logic. The source states Riot tried to preserve the reflect and failed, then removed the mechanic outright. This is a considered design decision, not a reactive hotfix — and the referenced "continuous adjustment rounds" indicate smaller nerfs had already failed to stabilise her.

Core: Five Abilities, One Direction

Looked at separately, the changes read as noise. Stacked together, the direction is unmistakable.

W — reflect out, utility in. Mel no longer returns enemy projectiles; she destroys them, and in exchange grants a shield and movement speed to herself and nearby allies. This is the structural change of the kit. Reflect was a solo-play tool — one that could flip a fight alone. A shield aura is a team tool. Within a single ability, the champion's centre of gravity has shifted from the player to the team.

Q — a tax on guaranteed burst. Damage moved out of the first explosion and into subsequent hits, with a slower projectile. The translation is direct: the first clean hit returns less, and the remainder demands tracking. This is not a nerf, it is a skill-expression tax.

E — a two-sided lever. Shorter root at close range, longer root at max range. The ability rewards players who manage spacing and punishes those who leaned on panic close-range lockdown. Reading it as a pure nerf is wrong — it is skill-indexing.

Passive — the change that turns build direction. Early damage and duration fell, but health and armour scaling rose. That is not cosmetic: the itemisation arithmetic itself has changed, and a health-armour stacking off-meta tank path may open — one Riot may not have fully stress-tested.

R — raw damage down. Direct and unambiguous.

Now add the five together. Raw damage is down in all four places. Utility and durability are up. Mel's carry ceiling has dropped while her enabling floor has risen. In practice she is sliding from "burst counter-pick" to "sustain/enable mid." That is a role-identity shift, not a number tweak — and writing both as one thing is exactly where patch coverage always goes wrong.

Who gains? Riot supplied two examples itself: Seraphine's ultimate reflect-charm, and Jhin's fourth-shot crit reflect. The projectile-reliant class Mel hard-countered — Seraphine, Jhin, and by extension Ezreal and Caitlyn-style poke-crit marksmen — has lost a permanent draft-phase threat.

Who loses? Mel one-tricks and Mel specialists. For them this is a champion-level aging cliff: the one pick their rank was built on now has to be relearned from zero. Across 18 years of watching this industry, the pattern repeats — the system changes, and the first casualty is always the player who took a single pick to mastery.

One thing is not clear and will stay unclear: the source names no player. So the price paid by a 19- or 20-year-old one-trick has no demo file behind it. I cannot measure it, and I will not pretend to.

Contrarian: "Her Identity Was Preserved" Is the Weakest Claim in This Coverage

Asterisk audit. There is not one metric in this report — no win rate, no pick rate, no ban rate. When data is missing, narrative fills the space. Here the filling was done by a relief frame: the frustration-reducing change, the champion fixed, justice served.

The problem sits in correlation versus cause. Players were frustrated by Mel — treated as fact. But was the frustration caused by the reflect, or by her numbers, or her cooldowns, or her place in the ban system? Answering that requires pick-ban data, which is not cited anywhere. Riot removed a mechanic — that is a fact. Riot removed the frustration — that is still a claim. A patch cycle sits between them, and the headline erased the gap.

The second claim — that the change preserves Mel's identity — is the weakest of all. Riot itself wrote that the reflect was her signature characteristic. Calling the removal of what the developer calls signature a preservation of identity is an optimistic inference, not neutral analysis. The identity has changed; the only open question is whether the new one holds.

A third risk nobody is writing: rework churn. Touching five abilities at once is the classic overshoot recipe. First passes miss, and two patches of hotfix churn follow. Treating the 26.22 state as final would be a mistake.

One more thing I have to repeat from a Seoul desk. LCK and KeSPA's disciplined patch management is one institutional route — and only one. Treating it as the universal endpoint would be my own convenience talking. The fact that these notes arrive via a Vietnamese localisation is itself a reminder that this content circulates across many markets simultaneously, and where a large share of the player base is mobile-first, the effect lands at a completely different ratio. I am reading from documents, not from presence — that stays on the record too.

Takeaway: Three Things to Watch in the Next Two Weeks

First, where Mel's win rate settles in the first fourteen days on live servers. If it drops below 45 percent, there is a live chance part of the reflect returns — Riot's record has been, to a degree, one of restoring baselines. Second, the rework honeymoon: an initial pick-rate spike on a new kit is natural, and reading it as "balance succeeded" is the oldest error in the book. Third, and most interesting — watch whether this change becomes the template for the next round of anti-fun mechanic pruning. I rebuilt the season from patch notes and timestamps, not from memory, and the record says Riot rarely walks back a direction once chosen.