FPL GW1 squad 2026/27: our current £100.0m draft

Here is the recommendation before the explanation.

If the Gameweek 1 deadline were today, SmartPlay's balanced starting squad would be a £100.0m 3-5-2, with Erling Haaland captain and Gabriel vice-captain. The starting XI projects at 51.9 points for GW1 after the captain bonus.

That is a dated model decision, not a claim that these are permanently the “best” 15 players. The deadline is Friday 21 August at 6:30pm UK time. Prices, injuries and expected minutes can all change the answer before then, so this article records the 9 August snapshot and the live squad builder remains the final check.

First, this is not a GW1 Wildcard chip

Before GW1, FPL already gives every manager unlimited transfers. You do not need to play a Wildcard to enter this team.

We use SmartPlay's Balanced Wildcard setting because it solves the right kind of problem: build a legal 15-player squad from a blank sheet and value the short run ahead, not only opening weekend. It blends GW1 xPts with the three-gameweek outlook, expected-minutes durability, captain routes, value and autosub cover.

The Free Hit setting answers a different question. It pushes harder toward the best possible one-week XI because the squad disappears after the deadline. That is useful for an actual Free Hit later in the season, but it is too short-sighted for a team intended to survive beyond GW1.

The current 15

Balanced GW1 starting squad

3-5-2 · £100.0m

Goalkeeper

• Raya

  ARS · COV (H)

  £6.0m · 3.7 xPts

Defenders

• Gabriel (VC)

  ARS · COV (H)

  £8.0m · 4.8 xPts

• Mosquera

  ARS · COV (H)

  £5.5m · 3.7 xPts

• Virgil

  LIV · NEW (A)

  £6.5m · 3.5 xPts

Midfielders

• Mbeumo

  MUN · HUL (A)

  £8.0m · 4.6 xPts

• E.Le Fée

  SUN · IPS (A)

  £6.0m · 4.1 xPts

• Doku

  MCI · BOU (H)

  £7.5m · 4.0 xPts

• Szoboszlai

  LIV · NEW (A)

  £7.0m · 3.8 xPts

• Groß

  BHA · AVL (H)

  £5.5m · 3.4 xPts

Forwards

• Haaland (C)

  MCI · BOU (H)

  £15.5m · 6.2 xPts

• Calvert-Lewin

  LEE · NFO (A)

  £6.0m · 3.9 xPts

Bench order

• Forster (GK)

  BOU · MCI (A)

  £4.0m · 0.7 xPts

• Justin (B1)

  LEE · NFO (A)

  £4.5m · 3.0 xPts

• Rodon (B2)

  LEE · NFO (A)

  £4.5m · 3.0 xPts

• McBurnie (B3)

  HUL · MUN (H)

  £5.5m · 2.5 xPts

The raw arithmetic matters. The XI contributes 45.7 xPts before captaincy. Haaland's 6.2 is counted a second time, producing the displayed 51.9. The four substitutes are not quietly added to the headline number.

Why the model lands here

One premium, strength everywhere else

Haaland is the only true premium in this structure. His 6.2 GW1 xPts and 16.6 across the next three gameweeks give the squad its clearest captain route. Avoiding a second £12.0m-plus player then funds three premium or mid-priced defenders and five midfield starters; there is no £4.5m midfielder being asked to carry the XI.

That does not mean Bruno Fernandes is a bad pick. He is at 6.1 xPts for GW1, second only to Haaland in the current player pool. SmartPlay's captain-first alternative includes both and reaches 53.0 GW1 xPts, compared with 51.9 here, but gives up enough elsewhere to rank lower on the balanced multi-week objective. The choice is explicit: more opening-week power with Bruno, or the stronger all-round starting structure shown above.

Arsenal are the big bet

Raya, Gabriel and Mosquera occupy all three Arsenal slots. Together they project at 12.3 xPts in GW1 and 30.3 across the next three gameweeks. Gabriel also becomes vice-captain because his 4.8 is the highest unconditional projection in this XI after Haaland.

The upside is obvious; so is the risk. Three defensive slots from one club concentrate the result around the same clean sheet. This is not diversification, and it should not be described as one. If Arsenal's expected lineup or defensive outlook moves before the deadline, this is the first part of the draft to re-solve.

The midfield is built from small edges

Mbeumo leads the group at 4.6 GW1 xPts and 11.9 over three gameweeks. E.Le Fée, Doku and Szoboszlai sit in a tight band from 3.8 to 4.1 for the opener, while Groß closes the midfield at 3.4. None is presented as a secret certainty. Their value is collective: five playable starters, several price points and enough three-week strength to avoid building the whole team around one fixture.

The bench is cheap, not invisible

Justin is first substitute, Rodon second and McBurnie third. The two Leeds defenders each carry 3.0 GW1 xPts, so the first autosub is not a dead slot. McBurnie's 2.5 is weaker but still reflects a route to minutes.

Forster is different. At 0.7 xPts away to Manchester City, he is a budget reserve rather than a planned rotation partner. The bench also doubles up on Leeds defenders, so its outcomes are correlated. Those are real compromises created by spending £100.0m; calling the bench “strong” without stating them would be misleading.

Captaincy: Haaland leads on the number

Haaland6.2

Gabriel4.8

Mbeumo4.6

E.Le Fée4.1

Doku4.0

The armband follows the largest current projection. Gabriel is vice because he is next among players in this particular XI, not because defenders are automatically safer.

There is no need to turn ownership into the argument. Haaland captains because his current projection is highest and his expected minutes support it. Ownership can matter to a manager's risk posture, but it does not create expected points by itself.

If late news cuts Haaland's expected minutes, the recommendation must be rerun rather than defended with an old chart. The vice-captain is insurance for a non-appearance, not a substitute for checking news.

What is uncertain this early

Dastan v13 is the production model behind these values, but GW1 is a cold start: there are no 2026/27 league matches in the feature history yet. SmartPlay therefore applies a separate, auditable early-season layer through GW5:

• sourced team-news facts can adjust expected-minutes bands;
• players without an applied team-news adjustment are blended toward a position-and-price prior fitted on the previous season;
• the unadjusted model values remain stored alongside the displayed values;
• both early-season overlays switch off from GW6, when current-season evidence can take over.

The 9 August run also uses the most recent valid cached team-strength data because Understat's 2026/27 league feed was not available to the pipeline. The pipeline passed its publication checks, but the warning is visible on the builder because a fallback should be disclosed, not hidden.

The model gives us a disciplined starting point. Team news still gets the final word.

What could change this draft

The most sensitive points are the triple Arsenal defence, the exact-budget constraint and any late change to a starter's minutes. A single £0.1m rise makes this exact draft impossible. A single lineup downgrade can make a previously inferior player the correct replacement.

Deadline: Friday 21 August, 6:30pm UK time. Use this article to understand the structure and its compromises. Use the live Balanced Wildcard build for the names and numbers you actually submit.