Guide · 5 min read

Overall rank, gameweek rank and percentiles

Rank is the only number in FPL that measures you against everyone at once. It's also the most misread. Here's what each version means, and why a single point can be worth a hundred thousand places one month and a few hundred the next.

Three different ranks

Overall rank is your position among every team in the game, ordered by total points for the season. Gameweek rank is your position for one gameweek only, ordered by that week's score. Mini-league rank is your position among the people you actually care about. They move independently. You can have your best gameweek rank of the season and still drop in your league because a rival did better, or fall a hundred thousand places overall while winning your league week.

Teams on the same total share the same rank. In this season's first gameweek, two teams on 67 points both sat at exactly 787,025th; there's no tiebreak, just a shared position until the next gameweek separates them.

The size of the field

This season the game passed ten million teams within the first three gameweeks, a little over 10.4 million at the time of writing. That number is what makes rank figures feel so large. A rank of 1,000,000 sounds terrible and is roughly the top 10%. A rank of 100,000 is the top 1%. A rank of 10,000 is the top 0.1%, and 1,000 is a place most managers will never see.

The game's own settings define twenty percentile bands: the top 1%, then every 5% from 5% to 95%. When you translate your rank into a percentile, that's the scale the game itself uses. Ten million teams also means the field is dense; at many total-points values in the middle of the table, tens of thousands of teams share the same score.

Why a point is worth different amounts

Rank moves when you pass people, and how many people you pass with one point depends on how many are packed into the points band you're moving through. Picture the spread of total points as a hill: a few teams far out in front, a few far behind, and a dense middle where huge numbers sit within a few points of each other.

In the middle of that hill, one extra point might carry you past sixty thousand teams, because sixty thousand of them were sitting on exactly the score you just passed. Near the top, where the field is thin, the same point might move you three hundred places. The point is identical; the rank consequence differs by two hundred times. That's why managers in the top ten thousand talk about rank in hundreds and managers around a million talk in hundreds of thousands. They're describing the same game from different densities.

Early season is different

In August the whole field is bunched. After one gameweek, the gap between a good start and an average one is a dozen points, and millions of teams sit inside that gap. In this season's first gameweek the average was 50 and the top score was 131. A manager on 63 finished the week around 1,280,000th; a manager on 67 was around 787,000th. Four points, half a million places: roughly 120,000 places per point at that part of the table.

By May the same spread might be two hundred points wide, and each point covers a far thinner slice of the field. A red arrow in September is not a disaster; a green arrow in May is hard-earned. Judge the same movement differently depending on the month.

Gameweek rank and the average

Your gameweek rank is decided by how far you finished from the week's average score and by how spread out the week was. In the first gameweek of this season the average was 50; in the second it was 81. A 70 in the first week was well above par and produced a good gameweek rank. The same 70 in the second week was below par and cost overall rank, despite being a good-looking score. Every finished gameweek's average and top score are on the gameweek pages, one page per week.

The right question after each week isn't "how many points did I get?" but "how far from the average was I, and which of my decisions made the difference?" A 60 in a 50 week is a better performance than a 75 in an 81 week.

Live rank is an estimate

During a gameweek the game's rank figures lag; the field is still scoring and bonus points are provisional until each match's data is finalised. Any rank you see mid-gameweek, on the game or on this site, is a projection. FPL Rivals labels its live figures as estimates for that reason, and they can move by a good margin when late bonus lands. How bonus points work explains why.

What you can see live is which of your players is doing the moving. My Team shows, for each player in your squad, the rank places his score is currently moving you, given his effective ownership at your rank. A 6 from a player nobody around you owns moves you further than a 6 from one everybody has. Effective ownership is the reason.

Rank trajectory: the shape of a season

A single rank number is a snapshot. The more useful picture is the trajectory: where you started, where you peaked, where you bottomed out, and whether the line is heading the right way. Two managers finishing on 2,100 points with the same rank can have had completely different seasons, one in the top 50,000 in December and fading, one outside two million in October and climbing. The Rivalry Report compares both managers' rank trajectories and season checkpoints as one of its categories, and treats a manager who's rising as stronger than one coasting on an early lead.

For the mini-league version of all this, where the only rank that matters is first to eighth, read Playing the mini-league, not the world.