On-hit items in League of Legends: a balance signal that is not yet an event

Opening

Two weeks ago, on a Vietnamese-language League of Legends forum, a screenshot of a Riot August comment began circulating. The content fit into a few lines: the item Gươm Vô Danh needs a buff; the change should apply to both melee and ranged champions, with ranged users prioritised.

The screenshot carried no source link. No patch number. No date. Just one sentence, placed next to an item name.

Within four days, that sentence produced dozens of aggregation posts, hundreds of comment threads, and a debate far larger than itself: where exactly is League of Legends' on-hit item group broken, and why.

Years ago I wrote a two-page piece in secondary school, purely to prove that a bench player wearing number 14 had a better defensive rating than the team's star number 7. It took three straight losses before the coach agreed to try. We won five games after that. The lesson I kept was not the result. It was that a claim only stands when there is a table behind it. This time, the table does not exist.

Context

To read this debate properly, two things that are usually merged must be separated: a balance signal and a balance event.

A designer saying item X is weak is a signal. Item X gaining stats in patch notes is an event. Between those two points lies the entire gap between expectation and reality, and it is also where most community coverage is stuck.

The on-hit item group in League of Legends triggers effects on each basic attack. It does not exist independently. It lives on attack speed, on champions with attack-reset mechanics, and on an implicit assumption that games last long enough for sustained damage to outpace burst damage.

The most recent precedent the community keeps citing is Patch 25.14. In that patch, Riot buffed a series of items in this group: Móc Diệt, Gươm Vô Danh, and Vũ Điệu Tử Thần. The goal was clear, to revive items considered weakened after repeated system-level changes.

After Patch 25.14, community feedback did not shift. The group was still described with one word: trash. That is the single most important data point in the whole story, and it is routinely skipped. A large-scale buff already happened, and it did not solve the problem. Anyone waiting for another buff to fix everything should first ask why the last one was not enough.

Analysis

When the stage lights go off, the numbers start speaking. Here the lights went off long ago, and the numbers have not been published by anyone.

A decent analysis of the on-hit group needs four layers of data stacked together: item pick rate by game-time bracket, to identify when the group loses its footing; gold efficiency per point of damage, to separate the feeling of weakness from actual weakness; build diversity, measured as the number of viable build paths per champion; and win rate by game-minute bracket for the champions that depend on this item group.

None of those four layers appears in the current debate. The entire argument is built on gameplay feel and on memory of past seasons. That is why I call this a signal, not an event.

The notable detail sits in the phrase prioritising ranged. If that information is accurate, it is a directional clue. Ranged users consume on-hit items differently from melee users: their damage arrives from a safe distance and does not require committing. When a designer says ranged is prioritised, they are most likely thinking of an item-level fix, not a champion-level one. That choice carries consequences, because changing an item changes an entire class of players at once.

The structural problem lies elsewhere. An item shared by melee and ranged users has only one stat line. You cannot raise damage-per-hit for ranged users without raising it for melee users at the same time. This is a stat amplification problem: an item balanced around the median user will inadvertently push champions with special basic-attack mechanics far above or far below that median.

The three names the community cites most are Irelia, Yasuo and Yone. All three have attack-reset mechanics, all three build toward on-hit paths, and all three have a high skill ceiling. In other words, they are edge users: the group that benefits most or suffers most from any change to this item group's stat line. A small change does not spread evenly. It concentrates on those three names first.

Data does not lie; only interpretation betrays. The same 49 percent win rate can be a sign of a weak item group, or a sign of a group being built incorrectly. Telling those two apart requires build-order data, which nobody is publishing right now.

I have done this work in another sport. In 2026 I rebuilt a basketball defensive framework and applied it to football, coding more than thirty matches and calculating France's average successful pressing actions per game. That framework was never designed for football. It still produced a correct conclusion, because the core assumption, that active defending generates measurable value, held in both sports. For on-hit items, the equivalent assumption has not been tested. The debate is circling a metric nobody has defined.

Based on my experience tracking matches, one repeating pattern stands out. The on-hit group is usually not weak in the early game. It is weak in the mid game, when matches are decided by short skirmishes. Sustained damage needs time; modern skirmishes do not grant time. Read that way, the problem is not the stat line on Gươm Vô Danh. It is the tempo of the game, and tempo cannot be fixed by adding a few points of damage.

Contrarian angle

Most of the debate assumes Riot will act, and that the only open question is by how much. That assumption has no basis yet.

What exists is a Reddit comment, screenshotted, translated, and re-shared through at least three intermediary layers. The Vietnamese article carries no original link. Nobody has confirmed whether the comment survives in its original context. A sentence cut from its thread can mean something entirely different, shifting from an exploratory remark into a promise.

Every objection is an equation still missing a variable. Here, the largest missing variable is scope. The community wants the whole item group fixed. Riot, when it acts, tends to start with one item. Those two scopes produce two different outcomes, and only one of them will be counted as success.

The next risk is cyclical. The on-hit group has been adjusted many times, most recently in Patch 25.14. A dense adjustment history like that reflects an unstable design zone, where each fix generates the next one. If only one item receives stats this time, the probability of a revert within two patches is not small. The buff, overperform, revert sequence has happened repeatedly before.

A less discussed risk is content farming. The weak-item topic is cheap, easy to produce, and infinitely repeatable. The same headline can be published every month without new data. When content volume grows faster than evidence volume, debate temperature detaches from the substance of the problem. That is when a balance topic becomes a commenting habit.

The data gate does not open for the impatient. Until official patch notes arrive, any conclusion about how far this item group will be adjusted remains a probabilistic guess.

What to watch

Three things matter over the next two patch cycles. Scope comes first: whether Riot touches one item or the whole group. One item means the issue is treated as a local defect; the whole group means they have accepted a class-level diagnosis. Type of change comes next: stats or mechanics. Stat buffs close a temporary power gap; mechanic changes reach the tempo of the game. And the reaction of the edge champions is the easiest thing to miss: if Irelia, Yasuo and Yone all rise in presence after the change, the amplification problem has not been solved, only pushed into another cycle.

We tend to look for stars where the light is brightest, forgetting that darkness also has a shape. In this story, what deserves tracking is not the item that gets named. It is the items that do not, and the champions who never appear in a single comment.