How to use this

Every week a model reads every Premier League player — form, opponent, set-piece duty, how likely he is to play the full ninety — and works out how many points each should score. From those numbers it builds the strongest legal team, names a captain, and names the transfer it would make.

You don't have to take its word for anything. Every prediction is frozen before the deadline and scored against what really happened, in public. Nothing is edited afterwards. The Model Scoreboard is the receipt.

Under the pitch are the week's four decisions. Tap any of them for the reasoning.

Captain — doubles your points, so it's the biggest call of the week. One name, the player most likely to deliver.

Bring him in — the single best transfer, ranked not by ceiling but by how unlikely he is to waste your move. Two more sit beside him under Consider, and that isn't hedging: replayed across a full season, the top name returned in 54% of weeks, but at least one of the three returned in 94%.

For scale, the most reliable player in the league returned in 68% of his own games, and captaining Haaland every single week would have hit 60%. Anyone promising a nine-in-ten single pick is selling something.

The eleven is shown in formation, bench below, with a copper C on the captain. Tap anyone for the full breakdown.

xP is expected points — roughly what a player would average if the weekend were replayed a thousand times. On the pitch it's a whole number; open a player and you'll see a range instead, like 4–9. That range is the honest version. A single player's week is wrong by about three points either way no matter how good the model is, so showing you “6.3” would be inviting you to split hairs the engine can't split.

Return chance, on the transfer cards, is the probability he scores five or more — the question that actually matters when you only get one free transfer.

Early in a season everyone asks the same thing: how much should one good week change my mind? We measured it by replaying past seasons, and the answer splits cleanly in two.

Whether a player starts is settled almost immediately. After a single gameweek, this season's minutes predict next week's minutes far better than a whole previous season does. One match tells you who's in the team.

How good he is takes far longer. Judging a player on this season's points alone is worse than ignoring this season entirely until around Gameweek four. One hattrick or one blank is mostly noise. So the model moves fast on minutes and slowly on scoring rate — which is roughly what a sensible manager does by instinct.

The chips under the header rewind the season. Tap a past week to see the exact team named before that deadline, what everyone actually scored, and whether the eleven beat the average manager.

A chip marked SCORING is a week that has been played but not yet confirmed. FPL only finalises bonus points a few hours after the last whistle, and scoring a week before then would lock in numbers that are still moving — so that week shows the locked eleven with dashes where the results will go. It scores itself as soon as the official numbers land. Nothing is missing; it's waiting.

The Model Scoreboard reduces the season to a few lines: points per week against the average manager's, how many weeks it came out ahead, and — the harder test — how it compares with simply picking on points-per-game. That last one is the bar worth watching. Beating the average manager is table stakes; beating the obvious shortcut is the actual claim. Some weeks it wins, some it doesn't, and the losses stay on the page.

1. Open the dashboard a day or two before the deadline.
2. Check the transfer picks against your own shortlist.
3. Weigh the captain call against your gut, then choose.
4. Make your move — or don't, and bank the transfer.
5. After the gameweek, rewind and see how it went.

Worth knowing so you can second-guess it in the right places.

Rotation risks are overrated. The engine is better at judging how a player performs than whether he starts, so a squad player with good underlying numbers will be rated as though he plays every week. If you suspect someone is being rested, trust that over the number.

Anything off the pitch is invisible. New signings, new managers, pre-season fitness, a press conference on Friday. None of it reaches the model.

Players with no Premier League minutes aren't rated at all. Promoted-side players and returners from abroad get no number rather than a guessed one, so they simply won't appear. Their absence is not a verdict.

A high clean-sheet percentage is a ranking, not a promise. The very top of that scale still runs optimistic.

xP — expected points, the model's best estimate for the week; shown as a range because a single week is never that precise.
Return — a week of five points or more.
Return chance — the probability of exactly that.
Haul — a big single week, nine points or more.
Clean sheet — no goals conceded; four points for a defender or keeper.
Nailed — a guaranteed starter, not a rotation risk.
Capture — on the scoreboard, how close the named eleven came to the best possible eleven in hindsight.