Riot Removed Mel's Reflect but Shipped No Numbers With It: The Real Story of Patch 26.22 Is the Data That Never Arrived

I opened the Patch 26.22 notes looking for the one thing every balance update should lead with — a number. Win rate, pick rate, ban rate, anything. There wasn't one. Riot stated plainly that Mel's projectile reflect was her signature identity, that they tried to preserve it and could not, and that W will now destroy projectiles and grant a shield plus movement speed to Mel and nearby allies. The argument was complete. The receipt was missing.

The Forsaken tape taught me that silence is still a data point. In 2026, running a 180-member CS:GO Discord out of Bangalore, I thought the story was the cheating. The real story was that nobody had kept the VOD with timestamps. Whatever wasn't logged vanished from the argument. The same thing is happening here: an identity-level decision, shipped with zero public data.

Context: the reflect was never just a skill

Mel was designed as a counter-pick. Her W returned enemy projectiles, meaning an opponent's own burst came back at them. Riot cited two examples — Seraphine's ultimate reflecting into a charm, Jhin's fourth shot returning with crit. Those examples are not random. They land precisely on the two classes that feel most helpless against a reflect: poke-crit marksmen and projectile-dependent mages.

Read Riot's language carefully. They did not say Mel was too strong. They said the reflect was too frustrating to play against, and that they had tried to preserve it. That sentence is itself an admission: an internal design-consent threshold was crossed. The company also noted multiple continuous adjustment rounds had already run on Mel. That detail is the loudest thing in the notes. When a champion survives several tuning passes, the next step is not a number change — it is a mechanic removal.

W was not the only change. Passive lost early damage and duration but gained health and armor scaling. Q shifted damage from the first explosion to subsequent hits, with a slower projectile. E's root is now shorter at close range and longer at max range. The ultimate took a nerf. This is a kit rebuild presented as an adjustment.

Core: this is a role shift, not a nerf

In the Mel VODs I've watched this season, one pattern repeats. Mel did not win because her damage was highest. She won because opponents could not stop themselves — they ran their combo and it came back at them. That was psychological pressure, not numerical pressure. Her draft job was to hold the opponent's most expensive habit hostage.

That pressure is gone. W now destroys projectiles and grants shield and movement speed. Notice the direction: reflect was reactive, aggressive, punitive. Shield and movement speed are proactive, enabling, protective. Mel is being moved from burst counter-pick to sustain/enable mid. That is an identity change, not a tuning pass.

Raw damage is going down across passive, Q and ultimate; durability and utility are going up. The net effect: her carry ceiling drops, her enabling floor rises. The solo-queue specialist who won games alone loses. The coordinated mid who runs a team gains.

Who benefits, who pays, and who was never named

No team, pro player or tournament appears anywhere in the notes. That is deliberate — champion-level balance notes are written this way because naming targets is never good business. But the unnamed damage is still damage.

The clearest beneficiaries are the champions who lost priority to the reflect threat: Seraphine, Jhin, and projectile-reliant poke marksmen of that class. I'd add a caveat. That benefit is not permanent. When a hard counter disappears, the champions it suppressed get a one-to-two-patch pick-rate bump and settle back, because real metas are never built by one champion leaving.

The biggest loser has no name in the article: the Mel one-trick. My rule after any change is to ask who is paying the bill. Here it is the player who invested two seasons into one pick, learned its limits, and built a rank on it. Riot's decision favours the collective experience over individual investment. That is defensible. Selling it as 'her identity is preserved' is not.

Skill-indexing: the design intent hidden in E and Q

The least-discussed part of this patch says the most. E's root is shorter at close range and longer at max range. Calling that a nerf is lazy — it is a spacing reward. Players who measure distance gain; players who panic-engage lose. That is a skill-tier filter splitting the Mel player base in two.

Q repeats the logic. Damage moved off the first explosion, projectile slowed. Less guaranteed burst, more value on tracking and aiming. That is a skill-expression tax. Riot is not selling a nerf here; it is raising the skill floor.

Passive is the quietest move and the most structural. Less early damage and duration, more health and armor scaling — that bends build direction. Players who ran a full damage path will drift toward health and armor. Structural changes like this frequently overshoot on the first pass.

The unlisted risk: rework churn

Multi-ability reworks follow one rule — version one is never final. Touching passive, Q, E, W and R at once almost guarantees a hotfix once live data lands. My advice is blunt: do not form a firm opinion on Mel inside the first week of Patch 26.22. Wait for one or two patches of live data.

The second risk is timing. The patch is slated roughly three weeks after announcement. If that window lands on either side of a major event, a tournament-patch versus practice-patch gap opens. To be clear: no tournament is named in the source, so this is a flag, not a claim. Patch numbering convention suggests a late-season window, which would reduce disruption — that is inference, not fact.

The third risk is the largest and the least discussed: no win rate, pick rate or ban rate was published. That does not mean the data didn't exist. Riot clearly had it, or they would not have removed a mechanic after several tuning rounds. The data existed and stayed internal. Arguments about internal data are arguments about feelings, not evidence.

I've watched this pattern play out repeatedly on the mobile side. When something is removed as 'toxic' without a number attached, the outcome is always the same: the people who lost something shout, the people who got relief applaud, and the developer stands in the middle. A thumbnail is a promise; the click is the receipt. Riot shipped the thumbnail.

How I could be wrong

Suppose the reflect was never Mel's real power. Suppose her actual strength was wave control through Q and E, and the reflect was a dramatic outer layer that looked good in clips. If that's true, removing it barely moves her win rate, and my entire identity-shift argument collapses.

Second, the shield and movement-speed aura may land harder than Riot modelled. If Mel becomes an enabler every coordinated team must pick, a patch that began as a nerf ends as a new problem. That is plausible precisely because so many variables changed at once — and multi-variable patches get corrected in the following patch.

Third, the frustration argument may simply be correct and I may be over-suspicious. If millions of players had bad games into Mel for years, the decision is right even without published numbers. My case weakens the moment someone asks whether a publisher owes full data disclosure on every balance call. It doesn't. But when the publisher itself says it removed a champion's signature identity, one number in support of that call is the minimum.

I'll also concede the structural limit: I come from the mobile-first scene, where ban cycles and delistings are daily work. My receipt-audit on PC League is not as deep as it is on BGMI. My VOD sample is limited. What I have is an estimate, not a measurement — and my strongest claim is retractable once live data lands.

Takeaway: a bet with a timestamp

Every public hot take is a bet with a timestamp attached. My bet: within two weeks of Patch 26.22 going live, Mel's win rate drops, and a hotfix follows that either raises W's shield scaling or returns some passive damage. If neither happens and she settles into the new role, I'll accept that Riot did what it said and my suspicion was wrong.

Second bet: within six months Riot removes another 'anti-fun' mechanic and presents it without public data. If that pattern repeats, the question stops being champion balance and becomes design governance — how often, and how quietly, a publisher can overrule the investment its players made. Third-party discipline means checking your gut against the dashboard. This patch never published the dashboard.