Fantasy Premier League
Fantasy points, from an expected-goal model
Every Fantasy Premier League player, projected for the round ahead from the expected goals this site fits for each fixture. Nine ways of reading that projection, one question apiece.
- Deadline
- in 6d 13h Saturday, 10/10/2026
- Gameweek
- 6 the round every board on this page is projected for
Gameweek 6, projected
What the model expects of the round ahead: the three players it projects highest in it, and the cheap ones whose stored rounds say they are about to matter. The round's own page carries every player projected for it and the fifteen the budget buys.
Hidden gems
Points over the next 5 rounds rather than over this one alone: an enabler is bought to be held, which is what that board ranks on.
Who we project highest
Projected for gameweek 6 alone. The boards below rank the same players over the 5 rounds in front of a squad.
Gameweek 5, marked
How the projections for the last completed round held up, and who actually took the points.
Our squad, projected · actual
52.0 46
Erling Haaland took the captain's armband, doubled on both sides.
The best fifteen the projections frozen before the deadline could buy, scored the way the game scores a week — the eleven it would have started, with substitutions where a player never came on.
Model error over the board
2.45 a player, on average
60 players projected before the deadline, against what the round actually paid.
Who actually scored
Every player who appeared, not only those the model projected highest.
Nine questions, nine pages
Every tool in this section is a presentation of the same projected board — 9 of them, one to a question. The figures behind each are the ones explained below.
- Best value Who scores most for what he costs? Every selectable player ranked on projected points, with what each is worth per million of today's price.
- Captain Who should wear the captain's armband? The captain's armband ranked on a distribution over the round rather than on the projection alone — a floor, a ceiling and a chance of a haul.
- Differentials Who is good and barely owned? The board gated on ownership, with each player's edge split into what the model projects and what the field has not noticed.
- Hidden gems Which cheap players are on the rise? Cheap players qualified on the rounds already stored — rising minutes, rising underlying figures — rather than on a projection of them.
- Regression Whose run is about to end? Five rounds of returns read against the chances behind them, and the candidates split both ways: returns above the chances, and returns below them.
- Minutes Who is actually going to start? Every player banded on the model's own start probability, with the rounds the estimate was blended from beside it.
- Defence Which defender is worth the most for defending? Defenders and midfielders ranked on the defensive half of the projection — clean sheets, contributions and the award behind them.
- Fixtures Whose next few weeks are the softest? Every club's next three, five or eight matches banded on this site's own goal margin, and each run totalled in goals.
- Compare two Which of these two should I buy? Two players side by side across fourteen figures, with the better option in expected points named at the foot of it.
What is measured here, and what is not
-
Measured The projected points
Replayed against gameweek 2 after it was played: an average miss of 1.617 points a player, against 1.842 for “predict what he has averaged so far”, over 456 players. Recorded Sunday, 6 September 2026 and refreshed by hand, so the date is part of the claim.