Ask ten casual managers how they pick a transfer and you get ten stories — a hat-trick last weekend, a hunch about a fixture, a podcast tip. Ask the managers who finish in the top 10k repeatedly and you hear the same four rules, almost word for word. The system is famously simple. What separates people is whether they actually follow it under pressure.
Rule 1 — Decide with expected points, not with stories
Every decision — transfers, captain, bench order — gets checked against a projection of expected points. Not because models are magic: the public research on this is humbling. A study of six real projection models found the best of them explains barely 15% of week-to-week scoring variance, and even a simulated perfect model — one that knows the true probabilities — tops out around 17%. Football is mostly noise, one week at a time.
So why bother? Because the model's job isn't to be right about one gameweek. It's to stop you making clearly bad decisions — the £8.0m midfielder with two home bankers versus the one walking into four away games; the striker whose underlying numbers collapsed three weeks before his price did. Between two moves the model rates within a point of each other, pick whoever you like. The discipline is refusing moves the model rates as clearly worse.
The trap this rule protects you from has a number on it: in that same research, a player's goals in his last five matches was the single worst predictor tested — on actual goals it scored below zero, worse than guessing the same number for everyone. "He's in form, three in two" is the most expensive sentence in FPL.
Rule 2 — Treat free transfers like money
A saved transfer is an option on information you don't have yet. Press conferences land on Thursday and Friday. Injuries happen in training on Tuesday. The manager who has already spent his transfer on Monday can only watch; the one holding it gets to act on everything that happens before the deadline.
The practical version of the rule: make a move only when it clearly raises your expected points; when the best available move is marginal, roll the transfer instead — next week's version of you, holding two, has strictly more options. And do your business as late as you can. Waiting costs you a possible £0.1m price rise; moving early costs you every piece of news between now and the deadline. That trade is not close.
Rule 3 — Minutes first, everything else second
The best projection in the world is multiplied by zero when your player is on the bench. Before any other quality, the analytics crowd asks one question: does he start every week? A 6.0-rated certain starter beats an 8.0-rated coin flip, because the coin flip also burns your bench, your transfer, and your patience.
This is why expected minutes — not form, not talent — is the first column in any serious model, and why rotation-risk players are quietly absent from the top 10k's teams even when they look brilliant on paper. Every August someone buys the exciting new signing before checking whether his manager actually picks him; by October that transfer has usually been spent twice.
Rule 4 — Never transfer for a price change
Team value feels like winning — the number goes up, the group chat is impressed. But you score points with players, not with team value, and £0.1m buys almost nothing: most upgrades are separated by whole millions, not tenths. Meanwhile a transfer made on Tuesday to "catch the rise" is a transfer made with three fewer days of information, which is exactly what Rule 2 says not to do.
The managers who chase price rises are doing real work, daily, for a reward that almost never changes what team they can field. The ones who ignore prices completely and spend that attention on minutes and fixtures get the actual points. Watch demand trends to read the crowd's mind, not to join the stampede.
The playbook, in one line each
- Project everything; use the model to veto bad moves, not to split hairs.
- Roll a transfer unless spending it clearly wins points.
- No certain minutes, no pick — whatever the upside.
- Prices are weather. Points are the game.
None of this needs a spreadsheet anymore. FPL God's projections lean on the sharpest signal publicly available — the betting market, with the margin stripped out — our player pages lead with minutes and availability before anything glamorous, and the demand charts exist so you can see the stampede without joining it. The four rules are simple. The equipment is the easy part now too.