Best FPL Goalkeepers 2026/27: Where the £1.5m Goes

Goalkeeper is the position managers spend the least time on and the most money defending out of habit. This year the numbers say that habit is expensive. Ranked on projected points across the opening six gameweeks, the best goalkeeper in the game and the sixteenth-best are separated by 5.9 points, which is the narrowest spread of any position on the pitch.

All projections below are from The Algorithm's 7 August 2026 snapshot, and every figure is on the free FPL predicted points table if you want to check the workings yourself. The table updates daily, so it will move before the 21 August deadline.

Why goalkeeper is the flattest position in FPL

David Raya (£6.0m) tops our goalkeeper projections at 23.2 points over GW1-6. Antonín Kinský (£4.5m) is sixteenth at 17.3. That is the whole market, top to bottom, in a 5.9-point band across six gameweeks.

Compare that with what the same money buys anywhere else. Virgil van Dijk (£6.5m) projects 22.4 over the window. Gabriel (£8.0m) projects 30.2. One £1.5m step in defence is worth 7.8 points.

The reason is structural rather than seasonal. Goalkeepers score from clean sheets, saves and the occasional bonus, and every starting keeper in the league gets a broadly similar volume of all three. There is no goalkeeper equivalent of a penalty taker on set pieces. The ceiling is capped by the position itself.

One honest caveat before the picks. August is the one month where team motivation, normally the strongest lever in fixture reasoning, tells you nothing: nobody is chasing Europe, nobody is safe, nobody is on the beach. What is left is defensive quality and the fixture, and goalkeeper returns are the noisiest in FPL week to week. These are six-gameweek expectations, not predictions of GW1.

The £1.5m test

Here is the trade in its simplest form.

Raya to Bart Verbruggen (£4.5m) costs you 3.6 projected points across the opening six and frees £1.5m. Spending that £1.5m on the step from van Dijk to Gabriel gains 7.8. Net, the swap is worth roughly four points before you have made a single transfer.

That is not an argument that Raya is a bad pick. He is the best goalkeeper in the game on these numbers and he is 30.9% owned for a reason. It is an argument that goalkeeper is the wrong place to spend a premium, because the position will not pay you back for it.

The goalkeepers worth owning

David Raya (£6.0m) — Arsenal

Fixtures: Coventry (H), Friday night, in the season's opening match.

Raya scored 162 points in 3,330 minutes last season, the best goalkeeper return in the game, and he carries the lowest minutes risk available at a 7.5% chance of not starting any given gameweek. His 4.2 projected points in GW1 at home to promoted Coventry is the highest single-gameweek figure of any keeper.

However, he is 30.9% owned, which makes him the consensus rather than an edge, and the Arsenal defensive case has just taken a real hit (see the caveat below).

The verdict: correct, expensive, and owned by almost one in three managers.

Gianluigi Donnarumma (£5.5m) — Manchester City

Fixtures: Bournemouth (H) on the Sunday of the opening weekend.

Donnarumma is second on our goalkeeper projections at 21.2 over the window, 2.0 behind Raya for £0.5m less. He managed 135 points in 3,060 minutes last season at 4.0 points per game.

He is 11.1% owned, which is the interesting part: he costs half the gap to Raya for a third of the ownership, and City's home opener against Bournemouth is the softest GW1 assignment any of the top-priced keepers get.

The verdict: the premium keeper for managers who want the Arsenal alternative, at 11.1% owned.

Bart Verbruggen (£4.5m) — Brighton

Fixtures: Aston Villa (H) on the Sunday.

Verbruggen played all 3,420 available minutes last season for 130 points, which is the cleanest minutes-security signal in the position: not one gameweek missed. He projects 19.6 over the opening six, 3.6 behind Raya at £1.5m less, and 4.36 points per million, the best rate of any goalkeeper in the game.

At 16.9% owned he is not a differential, but he is the specific enabler that funds a premium defender.

The verdict: the best value goalkeeper in FPL, and the one this whole article argues for, at 16.9% owned.

Caoimhín Kelleher (£5.0m) — Brentford

Fixtures: Spurs (H) in the Saturday evening game.

Kelleher returned 143 points from 3,330 minutes last season at 3.9 points per game, and projects 20.6 over the opening six. That is within 2.6 points of Raya for £1.0m less, at 4.12 points per million.

The number that should get your attention is ownership. He is 5.6% owned. Of the game's top five goalkeepers on our projections, he is the only one under 10%.

The verdict: the differential in the position, at 5.6% owned.

Bernd Leno (£4.5m) — Fulham

Fixtures: Chelsea (H) in the Monday night game.

Leno also played all 3,420 minutes last season, for 122 points. He projects 19.2 over the window at 4.27 points per million, second only to Verbruggen on that measure.

The GW1 fixture is the worst of these five, which is exactly why he is cheap and why his 3.2% ownership exists. Over six gameweeks that opener washes out.

The verdict: the punt version of the Verbruggen argument, at 3.2% owned.

The full goalkeeper table

Projected points across GW1-6, as of 7 August 2026.

Read the right-hand column and the argument makes itself. Four of the top five on points per million cost £5.0m or less. The most expensive goalkeeper in the game is eighth on that measure.

The most-owned keeper who does not survive the numbers

Martin Dúbravka (£4.0m) is the second-most-owned goalkeeper in FPL at 23.5%, and he does not appear in our top sixteen over the opening six at all. He scored 96 points in 3,150 minutes last season at 2.7 points per game, the lowest of any regular starter in the position.

The pull is obvious: he is the cheapest starting keeper in the game and he frees £0.5m for the rest of the squad. The problem is that Kinský is £4.5m and 19.1% owned at the same club, which means a large slice of the FPL population has bought into a goalkeeper situation where two keepers are competing and our projections rate the more expensive one sixteenth.

The verdict: £4.5m, not £4.0m, is the floor for a goalkeeper worth starting.

What would change this call

William Saliba is out for what Arsenal have described as an extended period with a back injury sustained on World Cup duty, without surgery, and he is expected to miss the Coventry opener. Gabriel is expected to be available.

That matters twice over here, because both halves of the headline trade lean on Arsenal: Raya sits top of the goalkeeper list, and Gabriel is the defender the £1.5m buys. Our projections are refreshed daily and the 7 August snapshot already prices the news we had, but a defence losing one half of its first-choice pairing is the single most likely reason these numbers move before the deadline.

If Arsenal's clean sheet base slips, the goalkeeper argument gets stronger rather than weaker: it becomes an even worse idea to pay £6.0m for the position. What weakens is the specific Gabriel destination for the money, and that is worth watching in the ten days before GW1.

FAQ

Who is the best FPL goalkeeper for 2026/27?

On raw projected points, Raya at £6.0m, with 23.2 across the opening six. On points per million, Verbruggen at £4.5m, with 4.36. Which of those is the better pick depends entirely on what you do with the £1.5m.

Should I pay £6.0m for a goalkeeper in FPL?

Our numbers say no. The gap from the most expensive keeper to a £4.5m starter is 3.6 points over six gameweeks, and the same money is worth 7.8 points as a defensive upgrade.

Is a £4.0m goalkeeper worth it in 2026/27?

Not on these projections. The £4.0m bracket is mostly backup keepers, and the one heavily-owned exception does not make our top sixteen over the opening six.

Which goalkeeper is the best differential?

Kelleher at £5.0m and 5.6% owned, who projects 20.6 and sits fifth in the position. Leno at £4.5m and 3.2% owned is the cheaper version of the same idea.

Where this fits in your squad

The goalkeeper decision is small in isolation and large in what it funds. If you want to see how a full six-gameweek plan allocates the budget across all fifteen slots, the demo is a real plan with no signup, and how The Algorithm works explains what it is optimising when it refuses to spend on a keeper.

For the same exercise across the whole player pool, see our read on where the best value in the game sits, and for the other end of the budget question, whether Haaland is worth £15.5m.