Path of Exile 2 / Builds
A thousand characters is the entire visible player base for build data here. Everything on this page is about reading that honestly — what a filter costs you, why an incomplete class roster distorts every share, and what the gem system changes about what "a build" even means.
The ladder returns the top 1,000 characters. That number sounds workable until you narrow it to something you care about, at which point it stops being a sample and starts being a group chat.
Divisions are illustrative. Real distribution is never even, and popular classes hold far more than an equal share while others hold almost none.
The same three filters in Path of Exile leave roughly a hundred and eighty characters, which is a pattern. Here they leave an anecdote. Both games use identical methods; only the divisor differs.
Rough thresholds, offered as judgement rather than statistics. The point is not the exact numbers but that the same percentage means very different things at different sample sizes, and a build page never tells you which you are looking at.
Convert before you react. On a thousand-character ladder, multiplying a percentage by ten gives you the headcount, and the answer is frequently smaller than the percentage made it feel. The Path of Exile 2 overview has the full translation table.
This one is specific to a game still shipping content, and it is almost never mentioned. Path of Exile 2 was designed around a larger class roster than the one currently playable, with the remainder arriving in later updates.
A class's popularity share will drop when a new class ships, without anything about that class changing. Players redistribute into the new option, and every existing share shrinks arithmetically. If you see a build's popularity fall after an update, check whether the roster grew before concluding it got worse.
The same effect works in reverse at launch: with fewer classes available, players pile into what exists, which makes early shares look more decisive than they are. Neither direction is a signal about build quality.
In Path of Exile, support gems live in gear sockets, which ties a build's skill setup to specific items. In Path of Exile 2 supports socket into the skill gem itself. That is a quality-of-life change in play and a genuinely different thing when reading a build page.
Path of Exile
The skill setup is bound to a piece of gear. Reading a build means reading the item that carries the links, and swapping that item unravels the setup.
Gear and skill are one decision, and the linking cost belongs in the gear budget.
Path of Exile 2
The skill setup travels with the gem, independent of what you are wearing. Reading a build means reading the gem configuration and the gear as two separate things.
Which also means a character's gear tells you less about its build than the equivalent would in the older game.
Practically, when you open three characters running the same skill and compare them — the habit that does most of the work in reading ladder data — you are comparing two separable things here. Compare the gem setups to find the build, and compare the gear to find the budget. In Path of Exile those two comparisons are partly the same one.
It almost certainly is not eligible, and that is the normal state rather than a fault. With a thousand entries per ladder, the overwhelming majority of characters in this game are outside the window at any moment — considerably more so than in Path of Exile, where the cap is fifteen times deeper.
The eligibility checker works through the causes in order, and the full guide covers the awkward cases, including the rename and skill-attribution problems that look like something is broken when nothing is.
Build data in Path of Exile 2 is usable and fragile. It will show you what high-ranked players are running, which is genuinely informative, and it will do so through a window narrow enough that a single popular video can visibly bend the numbers.
Read it as evidence rather than instruction, convert percentages into people before reacting, and open the characters instead of trusting the aggregate. Do those three things and the thin sample stops being a problem — it just becomes something you have accounted for.
Sign in to your account