Patch 26.22: Riot Removed Mel's Reflect — What Changed Isn't a Number, It's a Champion's Identity

In 2026, the international athletics federation moved the centre of gravity of the men's javelin four centimetres forward. Two years earlier, in Berlin, East Germany's Uwe Hohn had thrown 104.80 metres. That mark still stands, and probably always will, because the implement that produced it was quietly retired. The javelin had begun landing beyond the field, so the federation shortened the javelin — not through physics, but through a rule.

I think about that every time a publisher deletes something that was, by its own admission, a champion's signature. League of Legends patch 26.22 is that kind of document. Riot Games has removed projectile reflection from Mel's W — the exact mechanic her identity was built on. A magic projectile coming at her used to bounce back into the hands of the person who threw it.

Riot's notes cite two examples. Seraphine's ultimate, reflected back into a charm on her own team. Jhin's fourth shot, its critical damage returned to sender. Two clips, two seconds of video. Those two frames built the champion, and those same two frames supplied the justification for dismantling her.

The Bhubaneswar notebook still opens to the race nobody clocked. At the 2026 Asian Athletics Championships I filed 41 pieces in 96 hours and missed the name that would stand on an Olympic podium four years later. Patch notes are the same kind of document. They tell you what changed; they do not account for what had to be changed.

Context: a champion built out of her own defence

Riot designed Mel as a mage who wins by contact. The logic was simple: throw a projectile at her and you carry the cost of your own decision. That is a specific slot in the mid lane, which is normally occupied by champions who deal damage at range — Xerath, Vel'Koz, Lux, Zoe, and in recent seasons enchanter-mages like Seraphine. All of them are built on a single assumption: that a projectile will travel.

Mel interrogated that assumption. She was a counter-pick, not a blind first pick. Teams reached for her when the opposing draft leaned on poke. At 2 a.m., the World Cup desk taught me to read the transfer window, and what that desk really taught me is that a pick's value is set by the conditions of the draft, not by the champion's raw numbers. A champion who only works in one situation is priced by that situation.

Patch 26.22 is not confined to the W. Passive, Q, E and R have all been touched in the same pass. Riot states that it tried to preserve the reflect and failed before pivoting. That sentence matters, because it marks this as a considered design decision rather than a reactive hotfix.

One caution on the numbering. Riot moved to a season-based scheme some years ago, in which the first number denotes the year. Under that scheme, 26.22 should mean the 22nd patch of the 2026 season. That is an inference, not a verified fact — the source does not state it outright. The gap between announcement and release is given as roughly three weeks.

Core: from reflection to cover

The new W has three parts. First, the projectile is no longer returned; it is destroyed. Second, Mel gains a shield. Third, and most interesting, nearby allies gain movement speed.

That third part is the real story. Reflection was a solo duelling tool — you stood alone and sent the incoming damage home. A shield and a movement-speed aura are collective tools. You do not win alone; you make your team capable of winning. The difference between those two things is not statistical. It is a difference of role. Mel is sliding from burst counter-pick toward sustain-enable mid.

Net power direction is downward — passive, Q and the ultimate all lose damage, and the reflect is gone entirely — while utility and durability move upward. Calling this a number tweak misreads it. This is a change to how the champion does her job.

Passive: not a number, a build direction

Early damage and duration are down; health and armour scaling are up. On the surface that reads as the familiar 'weaker early, stronger late' trade. In practice it is a pointer at itemisation. When a champion's scaling tilts toward health and armour, players' instincts tilt toward the items that multiply those stats.

This is the same manoeuvre you see in track and field when a federation fixes spike length or sole thickness. The athlete's technique does not change, but the entire preparation apparatus rearranges itself around that one number. Riot has signalled clearly that it wants Mel as a durable mid rather than a fragile burst mage. The open question is whether the tankier path was fully stress-tested. My suspicion is that it was not. First-pass changes of this shape routinely overshoot or undershoot, and a hotfix follows.

Q: the first explosion gets cheaper, the later hits get dearer

Damage has been shifted off the first explosion and into subsequent hits, and the projectile has been slowed. This is what I'd call a skill-expression tax. The first explosion was the part that landed whether or not you aimed well — a floor of guaranteed damage. That floor is now gone, and the damage has moved to the part that demands tracking, with a slower projectile giving the opponent more time to move.

I look for the quiet footwork beneath the loud scoreline, and this is where it shows. When a publisher shifts damage from 'guaranteed' to 'earned', it is making a statement about its own confidence in that champion's player base. That is not an accusation; it is a pattern. Riot consistently works on mechanics where damage is delivered without requiring skill.

E: a two-sided lever, not a straight nerf

E's root is shorter at close range and longer at maximum range. This is the clearest message in the patch: manage your spacing. The player who dove on top of an opponent to root them gets less. The player who held the exact edge of range and landed the skillshot gets more.

This is skill-indexing, not punishment. Which means it will split Mel's player base from within. Players with clean spacing get stronger. Players whose entire habit was close-range lockdown take the kind of hit a 21-year-old athlete takes when a rule changes under their feet.

Who benefits, who pays

The two examples Riot chose — Seraphine's ultimate and Jhin's fourth shot — were not picked at random. They represent the class that had to be most careful around Mel. Seraphine is an enchanter-mage; Jhin is a poke-crit marksman. Ezreal and Caitlyn sit in the same class, their entire damage output routed through projectiles.

Players of those champions have been handed a direct gift. Their projectiles no longer come back to kill them. That is not a small change. The whole logic of a counter-pick is that it neuters the opponent's most reliable tool. Remove the tool and Seraphine's and Jhin's draft value rises on its own — even if not a single one of their numbers moved.

On the other side sit Mel players, and specifically the ones who stake everything on one champion. I follow the athlete first, then the arena, then the money, because that is the order in which the damage usually lands. Here it lands first on the 19-year-old who put two years into Mel and must now relearn a champion who is no longer the champion she was.

Champion-pool erosion works exactly like an ageing curve. The ability is intact; the thing that ability was for has moved.

The missing numbers are the biggest story

This patch note contains not one win rate, pick rate or ban rate. Riot does not say how often Mel was winning, how often she was being banned, or in what share of games her reflect proved decisive.

That absence is not decorative. It directly caps how confident anyone can be about magnitude. Without those figures we can establish direction. We cannot establish size. Direction and magnitude are different things, and the note gives us only the first.

Riot says it was trying to preserve the reflect. That sentence is itself a signal. When a publisher talks about trying to save a mechanic, it already holds data showing the mechanic is a problem. We were not shown that data. We were shown the outcome.

One falsifiable claim, and I'll flag it as inference: this change came from solo-queue metrics, not pro-play pressure. No competitive circuit is referenced anywhere in the note. Mel is not cited in any tournament context. When a champion's ban rate climbs in solo queue and players start banning her because facing her is unpleasant, the publisher's technical options narrow. Riot took this one.

Contrarian: 'identity preserved' is the weakest claim in the document

Riot writes that reflect was Mel's signature characteristic. It then removed it and asserted that the change preserves her identity. That is the least defensible sentence in the patch note, and it is striking how easily it was accepted.

If something is signature, removing it does not preserve identity; it alters it. Shields and movement speed are useful utility, but they are not the identity of the champion people spent two years learning. It is rather like telling a javelin thrower the implement has changed but he is still the same athlete. Technically true. Emotionally hollow.

The second problem is that the relief narrative is entirely emotion-led. For players who found Mel oppressive, this is a justice story. There is no number behind it. We do not know how much less frustrating the game now is. When a change is defended by feeling rather than evidence, it can reverse after launch.

And that raises a larger question than the patch itself. Riot is this game's rule-maker, its stakeholder, and its court of appeal. When it decides unilaterally which mechanic is 'anti-fun', it rises above the consent of the community that invested years in that mechanic. It reminds me of 2 September 2026, when PUBG Mobile was banned under Section 69A. There the rule-maker was a government; here it is a publisher. The scale differs. The grammar is identical: whoever draws the map redraws it, and the player absorbs the cost.

A third angle the note does not touch: if 26.22 lands near a major competitive window, live server and tournament server will diverge. Rework-level changes are typically held out of the active tournament patch, so the professional impact will lag the live impact by a patch cycle. No tournament is named in the source, so anything firmer than that would be speculation. What is certain is that in the week this patch ships, solo queue and the pro scene will be playing two different games.

Takeaway

I would not file this under nerf or buff. I would file it as a document in which a publisher has stated that a player's emotional investment does not outrank its design decisions.

First, watch where Mel's win rate settles. If it collapses, a modified reflect could return — Riot has a record of walking back identity removals, though that is inference, not announced fact. Second, watch Seraphine's and Jhin's pick rates, because that is the real scoreboard: not Mel's numbers, but the numbers of the champions who can now breathe.

The most important watch is on the next few reworks. If Riot starts removing other 'anti-fun' mechanics on the same reasoning, patch 26.22 becomes precedent. Then the question stops being how strong Mel is. It becomes whether identity has any standing in this game at all, and who gets to decide.

The best analysis starts where the press box stops clapping. The patch note has finished applauding itself. I am still waiting for the numbers that sat behind the applause, and that we were never shown.