FPL Transfers GW36: After the City Double, Setting Up GW38

Data current as of: 2 May 2026 Window covered: GW36 to GW38 Assumptions: Salah ruled out for the season (return 24 May per FPL API). Manchester City and Crystal Palace have a confirmed double in GW36. GW37 and GW38 are single gameweeks across the board.

GW36 is a transfer week with two distinct goals overlapping. First, you want to be in on the City and Crystal Palace doubles if you somehow aren’t already. Second, you want to start setting up GW38, which is the canonical end-of-season chip-play week. Most managers go into GW36 with one free transfer and a hard choice: spend it now for the DGW, or hold it for GW37 to set up GW38 properly.

Below is the case for which moves to make this week and which to defer.

Top transfer target 1: Erling Haaland (FWD, £14.6m) if you don’t own him

Fixture run: GW36 (2 fixtures, double) | GW37 | GW38 Projected GW36: 7.60 pts combined (3.64 + 3.96) Form context: 24 goals, 25.88 season xG, 6.8 home pts/game

If you somehow reached GW36 without Haaland, this is your transfer. The combined GW36 projection of 7.60, the GW37 follow-up at 3.37, and the GW38 close at 4.07 produce a 15+ point expected run from a player who can also catch the Triple Captain chip if you’ve been holding it.

The price is also still rising (net transfers in 13,420 high), which means the longer you wait, the more it costs.

Why now: The Triple Captain math collapses if you don’t own the underlying asset. Buy first, captain second.

Top transfer target 2: Rayan Cherki (MID, £6.5m) if you don’t own him

Fixture run: GW36 (2 fixtures, double) | GW37 | GW38 Projected GW36: 8.14 pts combined (3.79 + 4.14, leading City field player) Price: Net transfers in 13,632 (high, leading riser)

Cherki is the cheaper end of the City DGW play and has a higher combined GW36 projection than Haaland. At £6.5m he fits in any squad shape, his season xA of 10.06 is elite for the price, and he’s the leading price riser this week.

If you own Haaland and have one transfer left, Cherki is the second buy. If you own neither, take Cherki first because the price is moving fastest.

Squad fit: Best as a Palmer downgrade (frees £4m) or as a fifth-midfielder slot upgrade from any sub-£6m placeholder.

Top transfer target 3: Morgan Gibbs-White (MID, £7.6m)

Fixture run: GW36 | GW37 | GW38 Projected GW36-38: 4.46 + 5.29 + 5.51 = 15.3 pts combined Price: Net transfers in 3,830 (high)

Gibbs-White is the GW38 setup buy. His projection climbs through the run-in (4.46, 5.29, 5.51 across GW36-38), his pStart is stable at 75%, and the price is rising. He’s not in the City DGW conversation, but he’s the highest-projected non-MCI midfielder for the GW37-38 window and the right buy if your free transfer is best spent on GW38 prep rather than chasing the GW36 double.

Why now: Buy in GW36 to bank the price rise and have him locked for the GW38 chip play. Buying in GW37 means paying £0.1m more and missing the GW36 ceiling.

Sell candidate 1: Mohamed Salah (MID, £14.0m) if you still hold

If you held Salah through GW35 hoping for a Friday team news swing, the swing didn’t come. The expected return date is 24 May, after the final whistle of GW38. Sell. Buy Haaland or Cherki with the funds.

The only edge case for holding is if you’re using a Free Hit chip in GW36 and Salah is parked on your real squad (then he doesn’t count for that week). Otherwise, this is a forced move.

Sell candidate 2: Cole Palmer (MID, £10.5m)

Form: 1.5 (collapsed) GW36 projected: 2.16 pts (single fixture) Price: Net transfers minus 3,879 (high), price falling risk

Palmer’s GW36 single-fixture projection is below £5m midfielders. The form line is dead, the underlying numbers in the recent stretch don’t support a £10.5m valuation, and the net transfer market has decided. If you’re holding because you bought him as a captain anchor in February, the position needs to come off the books before another £0.1m fall.

Replacement: Cherki at £6.5m frees £4m. Bruno at £10.4m is the like-for-like.

Sell candidate 3: Liverpool attackers (Ekitiké, Szoboszlai, Wirtz)

Ekitiké: £9.0m, net transfers minus 3,760 high, fell £0.1m today Szoboszlai: £7.0m, net transfers minus 3,497 high, fell £0.1m today Wirtz: £8.3m, GW36 projected 3.33 pts

Liverpool’s attack post-Salah has not produced the redistribution that owners hoped for. Two of these three lost £0.1m today. The model rates none of them above 3.5 pts in GW36. If you’re holding any as a Salah hedge, the hedge is failing.

Why now: Same logic as last week. Each day held is a likely £0.1m loss.

The hold call: Bryan Mbeumo (MID, £8.4m)

GW36-38 projected: 2.60 + 2.25 + 2.22

Same hold call as last week. Mbeumo’s projections aren’t exciting, but he’s not in price-fall danger and his minutes at Manchester United are stable. A free transfer paid to upgrade Mbeumo is a free transfer wasted when Cherki or Gibbs-White is a more impactful move elsewhere. Hold.

Fixture swing alert: Crystal Palace in GW36 only

Crystal Palace also have a double in GW36. Their assets (Mateta, Sarr, the £4-5m defenders) are real DGW pieces but only for that one week. If you’re picking between a CRY DGW asset and Gibbs-White, Gibbs-White is the better hold for the full GW36-38 window because his projections climb. CRY assets have one good week then drop back to single-fixture territory.

Transfer strategy by squad shape

One free transfer, no Haaland or Cherki: Buy Cherki. Cheapest entry to the City DGW, leading price riser, highest combined projection.
One free transfer, own one of the City pair: Buy the other. You want both for the DGW.
One free transfer, own both Haaland and Cherki: Buy Gibbs-White. Set up GW38.
Two free transfers: Salah/Palmer out, Haaland/Cherki in. Standard rebuild.
Hit considered: A 4-point hit to fund Cherki + Gibbs-White is defensible if both replace dead weight. Don’t take a 4 to upgrade an already-functional position.

The captaincy side of the same calendar is in our GW36 captain picks and GW37 captain picks. For the chip-stack logic across the run-in, see the Triple Captain final gameweeks guide and the Bench Boost chip guide. The final 4 gameweeks rank recovery plan ties it all together. The GW38 single-gameweek preview is the post to read for the closing chip play.

Get transfer suggestions tailored to your squad and budget.

Prices, ownership, and projections from FPL Pilot production data as of 2 May 2026. City DGW36 confirmed via projections (two fixtures returned per City asset). GW37 and GW38 are single gameweeks across the board. Net-transfer numbers update hourly, sub before the deadline if the trend has held.