FPL Gameweek 4 tips: captain picks and best buys

FPL Gameweek 4 preview: best buys, captain picks and differentials

Fantasy Premier League reaches its Gameweek 4 deadline at 13:30 on Saturday 12 September, and the round is a full one: all 20 clubs play, nobody blanks and nobody doubles. Chelsea at home to Hull City is the softest fixture in the league, Arsenal have the best clean-sheet chance, and the Manchester derby on Sunday is the one big game the model still refuses to call. The obvious names are already on most teams. What decides your Gameweek 4 is the six or seven squad spots after them.

What has moved since the last Gameweek 4 update

The projections were refreshed on 10 September and three things changed enough to act on. Alexander Isak (9.1m, LIV) climbed to 6.0 xP and is now the third-best forward in the round, but his horizon fell apart at the same time — more on that below. Pedro Neto (6.5m, CHE) and Nico Elvedi (4.5m, LEE) both rose into the top of their positions and are almost entirely unowned. And Manchester City's defence is no longer a straight sell: Josko Gvardiol (5.6m, MCI) now projects 5.0 for Gameweek 5, which turns the recommendation from "move him on" into "ride out one bad week". The fixture list itself has not changed.

Gameweek 4 fixtures: where the points are

Chelsea at home to Hull City tops the round on both win probability and expected goals. Arsenal, away at Sunderland, actually edge Chelsea on clean-sheet chance — which matters if you are choosing between a Chelsea and an Arsenal defender rather than an attacker.

Three things follow. Chelsea are the round's target, and the three-players-per-club limit is the only thing stopping managers loading up further. Sunderland and Hull City are the two attacks to avoid entirely — 0.69 and 0.70 expected goals. And the Manchester derby is a genuine coin-flip at 45% City against 31% United, with clean-sheet chances of 23% and 17%: a bad round to own a City or United defender, a fine round to own their attackers.

For planning, note what comes next. Manchester City host Sunderland in Gameweek 5 (76% win probability, 2.47 expected goals, 51% clean sheet) and Ipswich Town in Gameweek 7 (78%, 2.77). City assets are a buy for the rounds after this one. You can walk the full run yourself on the FPL fixture schedule and check the underlying market probabilities.

Best goalkeepers for Gameweek 4

David Raya (6.0m, ARS) is the top-projected keeper in the game for Gameweek 4 at 4.3 xP, and Arsenal's 50% clean-sheet chance at Sunderland is the highest in the round. He also leads all keepers on the six-round weighted view at 15.3, well clear of second place, with Arsenal hosting Leeds United in Gameweek 6 and Everton in Gameweek 8 — 53% and 54% clean-sheet chances. If you are paying up for a keeper anywhere this season, this is the one.

The value route is Emiliano Martinez (5.0m, CHE) at 4.1 xP and 0.82 points per million, the best value of any goalkeeper in the round. He rides the same Hull City fixture as the rest of the Chelsea block, so if you already own two Chelsea outfielders he uses your third slot — and Chelsea's schedule normalises from Gameweek 5. Treat him as a one-or-two-round keeper.

Bart Verbruggen (4.5m, BHA) at 3.6 xP is the cheapest reliable starter in the position; Brighton & Hove Albion travel to Coventry City with a 33% clean-sheet chance. At 22.2% ownership he is already the template's budget answer, so he protects rank rather than gaining it.

Best defenders for Gameweek 4

Defence is where Gameweek 4 is won, because the two best fixtures in the round are also the two best clean-sheet fixtures, and almost none of the cheap routes into them are owned.

Maxence Lacroix (6.0m, CHE) is the top-projected defender in the game for Gameweek 4 at 5.8 xP, and unlike most of the Chelsea back line he does not fall off a cliff afterwards: 5.8 → 3.3 → 3.6 over the next three rounds, with a six-round weighted score of 14.8 that puts him second among all defenders. He is a buy you can keep for a month.

Ezri Konsa (4.4m, ARS) is the single best points-per-million pick in the entire game this week at 1.14 — ahead of every midfielder, forward and keeper. Arsenal's 50% clean-sheet chance, at the cheapest price in a top-two defence, is close to a free upgrade. Be honest about the horizon though: 5.0 → 3.0 → 3.5 → 2.4, so he is a strong two-round pick rather than a season-long hold.

Gabriel (8.0m, ARS) at 5.4 xP is the premium version of the same bet and leads all defenders on the six-round weighted view at 15.3. The model gives him nearly twice Konsa's goal threat (0.11 expected goals to 0.06), and his Gameweek 6 and Gameweek 8 numbers (4.3 and 4.2) are the best of any defender in the game. At 24.7% owned he is template, not a differential — but he is the correct template.

Reece James (5.4m, CHE) carries the highest assist projection of any defender in the round at 0.18 expected assists on top of Chelsea's clean sheet, which makes him the Chelsea defender to own if you want ceiling. His projection drops to 2.7 in Gameweek 5 and keeps sliding, so buy him as a short-term play and plan the exit.

Budget and differential defenders for Gameweek 4

Nico Elvedi (4.5m, LEE) is the hidden gem of the round and he is new to this list. Essentially nobody owns him, he projects 4.6 xP at home to Newcastle United, and and unlike the cheap Chelsea and Arsenal defenders he does not fall away afterwards — 4.6 → 4.0 → 2.5 → 3.0 → 3.5, weighted 13.0, which is more than Konsa, Colwill, Acheampong or Jorrel Hato manage. He also carries a higher defensive-contribution projection (1.2) than any of them, which is where cheap defenders now earn their extra points.

Chris Richards (5.0m, CRY) and Jaydee Canvot (4.9m, CRY) remain the best-balanced cheap picks. Both project 4.7 xP against Ipswich Town, both are owned by under 1%, and both hold their value where the cheap Chelsea options do not: Richards runs 4.7 → 3.8 → 3.7 for a weighted 13.5, Canvot 4.7 → 3.8 → 3.5 for 13.1. Crystal Palace host Nottingham Forest in Gameweek 6 and Newcastle United in Gameweek 8.

If you want the Chelsea clean sheet as cheaply as possible, Josh Acheampong (4.5m, CHE) at 4.9 xP and 0.4% ownership is now better value than Levi Colwill (4.9m, CHE), though both collapse from Gameweek 5 onwards — Acheampong to 2.3, Colwill to 3.5 and then 1.4 by Gameweek 9. Buy either knowing the exit is already scheduled. Jan Paul van Hecke (4.9m, TOT) at 4.2 xP is the more durable cheap defender at 13.6 weighted, with Tottenham Hotspur hosting Coventry City in Gameweek 7 and Crystal Palace in Gameweek 9.

Best midfielders for Gameweek 4

Cole Palmer (9.6m, CHE) is the highest-projected player in Fantasy Premier League for Gameweek 4 at 8.6 xP — a full 1.5 clear of the next name in any position. The components explain it: 0.67 expected goals, 0.40 expected assists (the most of any midfielder in the game), Chelsea's 50% clean-sheet chance behind him, and a penalty share of 0.99 against a side the model gives 0.70 expected goals. At 22.6% ownership he is still not template. He is the pick of the round.

Bruno Fernandes (12.0m, MUN) has the highest six-round weighted score of any player in the game at 23.9 — above Haaland, above Palmer. His 5.7 xP for Gameweek 4 understates him: the run reads 5.7 → 7.2 → 7.3 → 6.4 → 7.6 → 6.1, he is on penalties at a 0.99 share, and he projects 0.31 expected assists in the derby. If you have been waiting for the entry point, this is it — he is good now and better from Gameweek 5.

Morgan Rogers (7.6m, CHE) at 6.1 xP is the second-highest midfielder in the round and the Chelsea attacker most managers already own alongside Joao Pedro. Bukayo Saka (9.5m, ARS) at 5.8 xP and only 11.4% ownership is the more differential premium: 0.33 expected goals, 0.22 expected assists and a 50% clean sheet behind him, weighted 16.9.

Florian Wirtz (7.4m, LIV) at 5.6 xP is the best-value premium midfielder in the round — 0.33 expected goals and 0.26 expected assists against Fulham, at 8.3% ownership. Dominik Szoboszlai (7.0m, LIV) at 5.3 xP leads Liverpool's midfield in expected assists (0.32) and holds their penalties at a 0.99 share, which is what carries his weighted score to 17.4.

Budget and differential midfielders for Gameweek 4

Pedro Neto (6.5m, CHE) is the biggest riser in the position and the best differential in the round at his price: 5.1 xP, 1.3% ownership, 0.27 expected goals and 0.26 expected assists in the league's easiest fixture. He is the seventh-highest-projected midfielder this week, ahead of every midfielder priced under 6.5m, and 27 percentage points less owned than Morgan Rogers, who costs more.

Diego Gomez (5.0m, BHA) is the best points-per-million midfielder in the game at 0.94, projecting 4.7 xP with 0.27 expected goals at Coventry City, and a weighted 13.9 makes him a keeper rather than a one-week punt. Mathys Tel (5.8m, TOT) at 4.8 xP is almost unowned at 0.3% and stays strong into Gameweek 5 (4.5), for a weighted 14.6.

Marcus Tavernier (6.0m, BOU) and Daizen Maeda (5.4m, IPS) hold the two best six-round weighted scores of any midfielder under 6.5m, at 15.7 each. Tavernier projects 4.7 xP and runs 4.7 → 4.4 → 3.8 → 4.7; Maeda is the cheaper and the only one of the two whose line is rising, 4.4 → 4.5 → 4.7, at 0.3% ownership. His 69 projected minutes are the caveat. Emiliano Buendia (5.9m, AVL) is the highest-ceiling differential in the bracket at 4.7 xP with a 0.86 penalty share, and Abdul Fatawu (5.5m, IPS) at 4.6 xP and 0.2% ownership takes Ipswich Town's penalties at a 0.88 share.

Best forwards for Gameweek 4

Joao Pedro (7.7m, CHE) is the top forward in the round at 7.1 xP and the most-owned player in the game at 72.5%. The projection is enormous for the price — 0.72 expected goals, 0.92 points per million — but at that ownership he is a hold, not a differential. If you do not own him, he is the most urgent buy on this page.

Erling Haaland (15.5m, MCI) at 6.1 xP still projects more expected goals than any player in the game (0.72, a fraction ahead of Joao Pedro) despite the awkward derby. He remains the highest-projected forward across six rounds at 23.5 weighted, with a run of 6.1 → 7.7 → 5.3 → 7.9 driven by City hosting Sunderland and then Ipswich Town. He is not this week's captain, but the fixture swing straight after the derby is exactly why you do not sell him.

Alexander Isak (9.1m, LIV) at 6.0 xP projects 0.54 expected goals at home to Fulham, the third-highest goal projection of any forward in the round. Be careful with the horizon though, because it has deteriorated sharply — 6.0 → 4.9 → 4.2 → 3.8 → 3.6 → 2.6. He is a buy for this week and next, not an investment. His 74 projected minutes are the other caveat, so re-check the FPL predicted lineups before the deadline.

Charalampos Kostoulas (5.5m, BHA) is the cheap differential of the round: 4.9 xP, 0.5% ownership, 0.39 expected goals and 0.89 points per million, which is better value than any forward priced above 6.0m other than Joao Pedro. Thiago Rodrigues (7.9m, BRE) at 4.7 xP is the mid-price forward with the best horizon (15.8 weighted) and takes Brentford's penalties at a 0.99 share. Dominic Calvert-Lewin (6.0m, LEE) at 4.6 xP is the penalty-taking budget option at a 0.89 share, though his Gameweek 6 projection falls to 2.5 when Leeds United visit Arsenal.

Captain picks for Gameweek 4

The captain scores double in Fantasy Premier League, and this week the decision is unusually clear.

Cole Palmer is the captain. He has the highest projection in the game, the best fixture in the round and the penalties, and a 17.2-point captained projection that is 3.0 clear of the next option — a wider gap than most weeks offer.

Joao Pedro is the safe alternative and where most managers will land, sharing the same fixture with 0.72 expected goals. Erling Haaland is the option for anyone who wants to keep the armband still through the derby; on the model that costs roughly 5.0 points of expectation against Palmer, defensible only if you think a 45%/31% coin-flip is undervalued.

Captaincy is a plan, not a single pick. Haaland is the model's best captain for Gameweek 5 (7.7) and Gameweek 7 (7.9), and Bruno Fernandes tops Gameweek 6 (7.3) and Gameweek 8 (7.6). Owning at least one of those two means you have an armband answer for the next month without spending another transfer on it.

Who to sell before Gameweek 4

The sell list is shorter than it was, because Manchester City's Gameweek 5 fixture rescues part of their defence. What is left is the genuinely stuck.

• Rayan Cherki (7.8m, MCI) — 30.3% owned, 2.9 xP. The third-most-owned midfielder in the game, with a six-round weighted score of 10.9 that trails mid-price midfielders costing 2m less. His best round in the next six is 3.5, so this is not a fixture problem that fixes itself. The clearest sell in the game.
• Marc Guehi (6.0m, MCI) — 18.6% owned, 2.7 xP, weighted 10.0, and no round above 3.8 in the next six. The expensive way to own the wrong City defender.
• Nico O'Reilly (6.5m, MCI) — 14.8% owned, 2.0 xP, and the model has him flagged unavailable with a return date of 13 September, which is the day of the derby itself. A real doubt on top of a bad fixture.
• Martin Dubravka (4.0m, TOT) — 17.8% owned and projected for no meaningful minutes at all. He is a bench keeper. If he is starting for you this round you are giving points away for free.
• Luke Shaw (4.4m, MUN) — 13.8% owned, 2.4 xP behind a 17% clean-sheet chance in the derby, weighted 8.7.

Josko Gvardiol (5.6m, MCI) comes off the sell list this week. He is only 3.4 xP for the derby, but he jumps to 5.0 in Gameweek 5 at home to Sunderland — the highest single-round defender projection anywhere from Gameweek 5 onwards — and a weighted 13.5 keeps him inside the top ten defenders. Take the one bad week rather than pay twice to move.

Two more to plan around rather than panic-sell. Riccardo Calafiori (5.7m, ARS) is the most-owned defender in the game at 47.9% and his Gameweek 4 is genuinely fine at 4.7 xP — but the run reads 4.7 → 2.8 → 3.2 → 2.4 → 2.3 → 1.8 for a weighted 11.5, against Konsa's 12.2 at 1.3m less and Gabriel's 15.3 for the ceiling. Same team, same clean sheets, a better answer at both ends of the price range. Maxim De Cuyper (4.8m, BHA) is the same story at 20.0% ownership: 4.2 now, 1.5 by Gameweek 9. Compare the whole pool side by side on the FPL player picks page rather than working from a shortlist.

Best Free Hit team for Gameweek 4

Gameweek 4 is not an obvious Free Hit week — nobody blanks and nobody doubles — but if you are chipping anyway, this is the highest-projecting legal squad the model builds for the round, respecting the three-players-per-club limit. It comes out as a 3-5-2 with a starting XI projected at 63.8 points before the captain's double.

The bench is deliberate fodder, as it should be on a Free Hit: the cheapest available second keeper plus two 4.0m defenders and a 4.5m forward, none of whom are expected to contribute. Note what the optimiser does with the money it frees — it takes Nico Elvedi, a defender essentially nobody owns, rather than a third Arsenal or Chelsea name, because the three-players-per-club limit is already binding on both. The FPL optimal teams solver re-solves this live against the current projections and is the authoritative version if you want to add your own constraints.

Gameweek 4 summary

Own Cole Palmer and captain him. Get an Arsenal defender in — Konsa for the value, Gabriel for the ceiling and the horizon — because Arsenal have the round's best clean-sheet chance and the best defensive run over the next five rounds. Take Nico Elvedi or the Crystal Palace pair as your cheap enablers rather than the Chelsea ones, because they keep their projections and the Chelsea cheapies do not. Sell Rayan Cherki, not Josko Gvardiol. And if you are making the transfer that pays for the next month rather than the next Saturday, Bruno Fernandes has the highest six-round weighted projection of anyone in the game.

For the previous round's picks and how the fixture run has shifted since, read the FPL Gameweek 3 preview, and for a manager's-eye view of the same round there is the Gameweek 3 FPL Physio team reveal. More rounds are collected in the FPL preview hub. If you are new to the scoring and the chips, start with the FPL strategy guide or browse all our fantasy football guides.

FAQ

Who should I captain in FPL Gameweek 4?

Cole Palmer. He projects 8.6 xP for Gameweek 4, the highest of any player in the game and 1.5 clear of second place, which makes a captained projection of 17.2. Chelsea host Hull City with a 79% win probability and 2.72 expected goals, and Palmer takes their penalties at a 0.99 share. Joao Pedro (7.1 xP) is the safer, far more heavily owned alternative in the same fixture, and Erling Haaland (6.1) is the option if you want to leave the armband where it is through the derby.

Which budget defenders are best for Gameweek 4?

Ezri Konsa at 4.4m is the standout — 5.0 xP and 1.14 points per million, the best value of any player in the game this round, behind Arsenal's 50% clean-sheet chance at Sunderland. Below 5.1m, Nico Elvedi (4.5m, Leeds United) at 4.6 xP is the near-unowned pick with the best horizon, and Chris Richards (5.0m) and Jaydee Canvot (4.9m) both project 4.7 xP against Ipswich Town at under 1% ownership.

Should I sell Rayan Cherki before Gameweek 4?

Yes. He projects 2.9 xP at 30.3% ownership, and his six-round weighted score of 10.9 is behind mid-price midfielders costing 2m less. His best single round over the next six is 3.5, so this is not a one-week fixture problem that fixes itself. Diego Gomez at 5.0m, Marcus Tavernier at 6.0m and Pedro Neto at 6.5m all project better this week and better over the horizon.

Is Erling Haaland worth 15.5m in Gameweek 4?

Yes as a hold, no as a captain. He projects 6.1 xP at Manchester United, only third in the round, because the derby is a 45%/31% coin-flip. But he still leads the entire game on expected goals at 0.72, and he is the highest-projected forward across six rounds at 23.5 weighted, with Manchester City hosting Sunderland in Gameweek 5 (7.7) and Ipswich Town in Gameweek 7 (7.9). Keep him and move the armband for one week.

Best differentials for FPL Gameweek 4?

Pedro Neto (6.5m, Chelsea, 1.3% owned, 5.1 xP) is the highest-projected low-ownership player in the round, followed by Josh Acheampong (4.5m, Chelsea, 0.4%, 4.9) and Charalampos Kostoulas (5.5m, Brighton & Hove Albion, 0.5%, 4.9). For a differential you can keep rather than flip, Nico Elvedi (4.5m, Leeds United) at 4.6 xP is owned by almost nobody and holds a weighted 13.0, and Mathys Tel (5.8m, Tottenham Hotspur, 0.3%) projects 4.8 now and 4.5 in Gameweek 5.

This preview was written by TheFantasyTool AI.