FPL Clean Sheets: How Many Points and Who Keeps Them

A clean sheet pays a defender four points and a midfielder one.

That ratio is the whole argument. Clean sheets are a defender and goalkeeper decision, and almost nobody else's, which is why pricing a midfielder up for his side's defence is one of the easier ways to waste money in this game.

The second thing worth knowing is that a clean sheet is a minority event even at the very top. Across the next eight gameweeks no club in the league averages better than a 42.9% chance of one, and in Gameweek 6 exactly one side clears 50%.

Here is what the points are worth, what the catches are, and where the chances actually sit.

How many points is a clean sheet in FPL?

Two conditions sit behind that table, and both catch people out.

The first is the 60-minute rule. A clean sheet is only awarded to a player who has been on the pitch for 60 minutes or more, excluding stoppage time. A defender who comes on in the 70th minute of a goalless game gets his appearance point and nothing else.

The second is kinder than most managers expect. If a player is substituted off before a goal is conceded, he keeps his clean sheet points, provided he had already played his 60 minutes. Your centre back taken off at 65 in a 1-0 win still banks the four even if the equaliser arrives at 88.

The goals conceded penalty is the other side of the same bet: minus one for every two goals a goalkeeper or defender concedes. Buying a defence is a two-way position in a way that buying an attack is not.

Which clubs are most likely to keep a clean sheet in Gameweek 6?

These are our clean-sheet chances for the round, as of 1 October.

Arsenal at home to Leeds is the only fixture in the round where a clean sheet is better than a coin toss. The step down from Arsenal to Everton is 15.2 points of probability, the widest gap between any two clubs in the round's order, and after that the field compresses hard: seventh to fourteenth sit inside eight points of each other.

That compression is the practical lesson. Below the top two or three fixtures, you are not really choosing between defences at all. You are choosing between prices.

Why a clean sheet is a calendar, not a badge

Here is the number that changed how I read this.

Arsenal have the best defensive outlook in the game over the next eight gameweeks, averaging 42.9%. Hull have the worst at 18.0%. That is a gap of 24.9 points between the best defence in the league and the worst.

Arsenal's own chance swings more than that. Across those same eight fixtures it runs from 28.0% away at Liverpool up to 61.7% at home to Hull, as things stand: a swing of 33.7 points inside one squad.

So the single best defence in the Premier League varies more from week to week than it differs from the single worst defence on average. The badge on the shirt is doing less work than the fixture next to it.

The honest version of that is also worth saying plainly. Even Arsenal are more likely to concede than not in five of their next eight. If you are buying a defender for clean sheets alone, you are buying something that fails more often than it lands, which is why attacking returns and defensive contribution do so much of the lifting in a modern defender's score. We covered that split in our ranked guide to the best FPL defenders.

The defence worth buying is the one about to turn

Man City read 20.1% away at Liverpool in Gameweek 6, level with Liverpool themselves and well down the round. Over the seven gameweeks after that, their average rises to roughly 37%, second only to Arsenal.

That is a buy through the trough, not after it. By the time the run is obvious the price has usually moved.

Gvardiol (£5.7m), Man City. 26.8% owned, 37 points from 434 minutes. He reads 3.2 in Gameweek 6 and 30.0 across the eight. The Gameweek 6 number is the worst of his next five and that is the point of owning him now.

Guéhi (£6.0m), Man City. 18.2% owned, 30 points from a full 450 minutes. 3.1 in Gameweek 6, 31.8 over the eight, and the steadier minutes read of the two.

Against that, the two names the fixture list flatters this week:

Branthwaite (£5.5m), Everton. 2.4% owned, 450 of 450 minutes, 4.4 in Gameweek 6 behind Everton's 35.9% at Hull. He is the differential of the round and I would still not buy him for the run: Everton's eight-gameweek average is 22.4%, twelfth in the league, and Gameweek 8 takes them to Arsenal at 10.5%. A one-week play, named as one.

Bogle (£4.6m), Leeds. 7.2% owned, 42 points and a form score of 11.3, the highest of any defender in the game's top 25 scorers. He reads 2.1 in Gameweek 6, because Leeds go to Arsenal and Leeds are the 11.8% in that table. Form is not a fixture.

All projections are expected points as of 1 October, and you can check every one of them in the free predicted points table, updated daily.

Frequently asked questions

Do midfielders get clean sheet points in FPL? Yes, one point, against four for a defender or goalkeeper. It is a rounding error in a midfielder's score, not a reason to pick him.

Does a substitute get clean sheet points? Only if he plays 60 minutes or more, excluding stoppage time. A player introduced after the 60th minute cannot earn them, however the game finishes.

What happens if my defender is subbed off and his team concedes later? He keeps the clean sheet points, as long as he had played 60 minutes and the goal went in after he left the pitch.

Is it worth paying up for a premium defender? Only where he also attacks. Clean sheets alone are a minority return at every price, which is why the gap between defences matters less than the gap between fixtures. Every fixture is two fixtures goes through that in full.

The bottom line

Buy the calendar, not the crest. Arsenal are the only side in Gameweek 6 whose clean sheet is better than a coin toss, Man City are the side whose run turns the week after, and a defender bought purely for sheets is a bet you lose more often than you win.

The Algorithm plans all of that eight gameweeks out against your actual squad. You can get a free score on your own team at Rate My Team, no signup needed.