Gameweek 1 · 2026/27

What the model picked, and what I entered instead

Fixed before the deadline of 21 August 2026, 17:30 UTC, from a snapshot taken at 09:45 the same morning. The snapshot’s hash is in the provenance strip below.


The disagreement, in one number

Model’s squad, objective

186.87

My squad, same objective

177.10

Difference

−9.77

That is the model’s own objective function, evaluated on both squads: for each of the five gameweeks in the horizon, the best legal eleven, plus the best starter’s score again for the captain, plus a small share of the bench, all discounted by 0.85 per gameweek away. Run on the model’s own fifteen it reproduces the value the solver itself recorded, 186.86809762841, to within 5 × 10−13.

My six judgement calls cost 9.77 expected points against the model.

Why not just add up the fifteen players?

Summing all fifteen players’ five-gameweek expected points gives 245.09 against 272.89, a gap of −27.80 — nearly three times the real one. It charges full value for all fifteen when only eleven start, so it penalises a good bench. The optimiser never maximised that quantity.


The model’s fifteen

The strokes on the right are the nine additive components of the gameweek’s expected points, at the same scale for every player on this page.

Model squad — cost £100.0m, bank £0.0m. Roles shown are the model’s own gameweek 1 team sheet.

Player Price Owned P(start) GW1 xP 5‑GW xP 123456789
B.Fernandes MID · MUN · starts, captain £12.0m 51.1% 0.92 4.82 23.00
Mbeumo MID · MUN · starts, vice £8.0m 36.6% 0.85 4.52 21.55
Virgil DEF · LIV · starts £6.5m 18.6% 0.98 3.92 21.86
Szoboszlai MID · LIV · starts £7.0m 41.7% 0.91 3.91 21.26
Anderson MID · MCI · starts £6.5m 8.4% 0.94 4.31 21.22
Rice MID · ARS · starts £7.5m 18.6% 0.90 3.81 19.23
Raya GKP · ARS · bench 1 £6.0m 37.0% 0.95 3.10 18.04
Thiaw DEF · NEW · starts £5.0m 1.7% 0.93 3.18 17.90
Lacroix DEF · CHE · starts £6.0m 9.5% 0.90 3.60 17.59
Thiago FWD · BRE · starts £8.0m 17.4% 0.73 3.77 17.14
Lammens GKP · MUN · starts £5.0m 17.2% 0.97 3.76 16.62
Guéhi DEF · MCI · starts £6.0m 18.0% 0.88 3.33 16.96
Evanilson FWD · BOU · bench 2 £6.0m 2.2% 0.86 2.99 16.41
Truffert DEF · BOU · bench 3 £5.5m 4.6% 1.00 2.48 16.32
Emegha no history FWD · CHE · bench 4 £5.0m 0.1% 0.26 1.55 7.80

The table scrolls sideways; the player column stays put.

Slot order, left to right — the slot is always drawn, so an empty slot means that component is exactly zero for that player:

  1. 1 appearance
  2. 2 goals
  3. 3 assists
  4. 4 clean sheet
  5. 5 def. contribution
  6. 6 saves
  7. 7 goals conceded
  8. 8 cards
  9. 9 shrink

 above the rule, solid — adds points.  below the rule, dotted — takes them away. Both halves are drawn at the same rate, so equal lengths mean equal magnitudes, and the faint horizontal rules are one expected point apart, so a stroke can be read off and not merely compared. The tallest on the page is Haaland’s goals term, 3.20 xP.


My fifteen, as a diff

Nine of the fifteen are the same player, so here is only what moved. Both goalkeepers survived; the argument about them is which one plays, and that is judgement call four.

Model squad against entered squad, grouped by whether we agreed. Figures are 5‑gameweek expected points.

 Model picked5‑GW xP I entered5‑GW xP
Goalkeepers
agreedRaya £6.0m18.04Raya £6.0m18.04
agreedLammens £5.0m16.62Lammens £5.0m16.62
Defenders
agreedVirgil £6.5m21.86Virgil £6.5m21.86
agreedLacroix £6.0m17.59Lacroix £6.0m17.59
agreedTruffert £5.5m16.32Truffert £5.5m16.32
swappedThiaw £5.0m17.90Mitchell £4.5m14.52
swappedGuéhi £6.0m16.96Coyle £4.0m6.42
Midfielders
agreedMbeumo £8.0m21.55Mbeumo £8.0m21.55
agreedAnderson £6.5m21.22Anderson £6.5m21.22
agreedSzoboszlai £7.0m21.26Szoboszlai £7.0m21.26
swappedB.Fernandes £12.0m23.00Tzolis £6.5m8.79
swappedRice £7.5m19.23E.Le Fée £6.0m16.83
Forwards
agreedThiago £8.0m17.14Thiago £8.0m17.14
swappedEvanilson £6.0m16.41Haaland £15.5m22.22
swappedEmegha £5.0m7.80Walle Egeli £4.5m4.72

The fifteen actually entered — cost £99.5m, £0.5m in the bank. Roles are mine, not the model’s.

Player Price Owned P(start) GW1 xP 5‑GW xP 123456789
Raya GKP · ARS · starts £6.0m 37.0% 0.95 3.10 18.04
Lammens GKP · MUN · bench, GK £5.0m 17.2% 0.97 3.76 16.62
Virgil DEF · LIV · starts £6.5m 18.6% 0.98 3.92 21.86
Lacroix DEF · CHE · starts £6.0m 9.5% 0.90 3.60 17.59
Mitchell DEF · CRY · starts £4.5m 6.5% 0.90 2.87 14.52
Truffert DEF · BOU · bench 3 £5.5m 4.6% 1.00 2.48 16.32
Coyle no history DEF · HUL · bench 4 £4.0m 0.7% 0.38 1.10 6.42
Mbeumo MID · MUN · starts, captain £8.0m 36.6% 0.85 4.52 21.55
Tzolis no history MID · ARS · starts, vice £6.5m 22.2% 0.43 1.75 8.79
Anderson MID · MCI · starts £6.5m 8.4% 0.94 4.31 21.22
E.Le Fée MID · SUN · starts £6.0m 10.9% 0.89 3.87 16.83
Szoboszlai MID · LIV · starts £7.0m 41.7% 0.91 3.91 21.26
Haaland FWD · MCI · starts £15.5m 69.3% 0.79 4.71 22.22
Thiago FWD · BRE · starts £8.0m 17.4% 0.73 3.77 17.14
Walle Egeli no history FWD · IPS · bench 2 £4.5m 2.2% 0.15 0.99 4.72

The table scrolls sideways; the player column stays put.


Where the gap opens up

The objective is a weighted sum over five gameweeks. Its weight falls by 0.85 each week out, so gameweek 1 counts for nearly twice as much as gameweek 5 — a squad that is wrong now is punished harder than a squad that will be wrong later. Both columns are scored the same way, re-picking the best legal eleven and the best captain each week, so this compares the two squads rather than my team-sheet decisions.

Best legal eleven per gameweek, before and after the decay weight.

GWModel, rawMine, raw WeightWeighted difference
1 49.08 47.25 1.0000 -1.83
2 53.97 49.78 0.8500 -3.56
3 49.81 47.20 0.7225 -1.89
4 50.10 48.66 0.6141 -0.89
5 48.20 45.13 0.5220 -1.60
Total251.16 238.01 -9.77

The six judgement calls

Where I overruled the model, or fed it a number I chose myself. Listed separately so each can be graded on its own when the week is scored.

#The callWhat it costs if I am wrong
1Belgian → Premier League attack factor set to 0.58Tzolis becomes the squad’s best player at 23.64 xP. At a factor of 0.30 he falls to 17.1 and Rice is the better pick.
2Bruno Fernandes fitness factor set to 0.85He drops out, and the £12.0m he frees funds Haaland. At 0.95 he keeps his place and Haaland is unaffordable.
3Diop P(start) set to 0.85Corrects a stale Fulham bench role. He still did not make the squad, so this one costs nothing either way.
4Raya started over Lammens, against the modelThe model prefers Lammens on a fixture priced by the FDR fallback rather than by expected goals.
5Le Fée over AmpaduThe model prefers Ampadu by 1.22 xP. Le Fée’s 16 bonus points last season are invisible to a formula with no bonus term.
6Egeli over Thomas-AsanteBought £0.5m of headroom against price rises. The model prefers Thomas-Asante.

Outside the snapshot

George Hirst

Our first choice at £5.0m, and reported to be mid-medical at Stoke City on a £10m permanent deal. FPL still lists him as a clean, available Ipswich forward. Nothing in the snapshot, the archive or the Wikipedia collection would have carried it.

Outside the snapshot

Coventry’s striker

Coventry were reported to be close to a £17m signing, which would push both of their existing forwards down the order. Unconfirmed at the deadline and worth re-checking; it did not change this squad, but it is the kind of thing that silently invalidates a price-sorted shortlist.


Where we disagree with the field

Every figure above this point is the model measured against itself. These are the two outside yardsticks.

Two outside opinions ship inside the same snapshot. Ownership is the crowd: the share of FPL managers who have already picked the player. ep_next is FPL’s own expected-points projection for the coming gameweek, produced by someone else’s method entirely. Neither is truth. The crowd is frequently wrong and ep_next is a blunt instrument. Both were fixed before the deadline. A model that agrees with them everywhere is adding nothing; a model that disagrees with them everywhere is probably broken.

For each position: the eight players we rate highest, and where those same players sit in the field’s ordering of the entire position pool. A negative delta means we rank a player above where the field does. The stroke beneath each figure runs left of its tick for a negative delta and right for a positive one, at one scale shared by all four blocks — so the longest left-hand stroke on the page is the model’s biggest bet against the crowd. It belongs to Thiaw, whom we rank 2 of 185 defenders and the field ranks 56.5.

The components for every row below are in the full-precision table at the foot of the page.

Read the right-hand column with care

ep_next carries only 26 distinct values across all 561 candidates, so its ties are enormous — up to 72 midfielders share one value. Ties here take the average rank, which is the standard treatment for tied data and much fairer than awarding every member of a 72-strong tie the best rank in the block. But it does mean a large negative ep_next delta often says more about how coarse ep_next is than about how strong our conviction is. Ownership has 102 distinct values and is the more informative of the two columns.

Goalkeepers

Eight of 60 goalkeepers, numbered in our order. Ranks are over the whole 60; as many as 22 share one ep_next value.

Player Price GW1 xP 5‑GW xP ep_next Owned By ownership
rank · delta
By ep_next
rank · delta
1Raya squad GKP · ARS £6.0m 3.10 18.04 4.0 37.0% 10.0 10.0
2Lammens squad GKP · MUN £5.0m 3.76 16.62 2.6 17.2% 5-3.0 12-10.0
3Verbruggen GKP · BHA £4.5m 3.29 15.71 1.9 21.0% 30.0 31.5-28.5
4Pickford GKP · EVE £5.5m 2.92 15.25 3.3 8.0% 7-3.0 3+1.0
5Leno GKP · FUL £4.5m 2.86 14.98 1.9 3.1% 16-11.0 31.5-26.5
6Donnarumma GKP · MCI £5.5m 2.67 14.35 3.3 8.5% 60.0 3+3.0
7Martinez GKP · AVL £5.0m 2.27 13.94 2.6 4.5% 10.5-3.5 12-5.0
8Sels GKP · NFO £5.0m 2.90 13.65 2.6 1.6% 22.5-14.5 12-4.0

The table scrolls sideways; the player column stays put.

Defenders

Eight of 185 defenders, numbered in our order. Ranks are over the whole 185; as many as 62 share one ep_next value.

Player Price GW1 xP 5‑GW xP ep_next Owned By ownership
rank · delta
By ep_next
rank · delta
1Virgil squad DEF · LIV £6.5m 3.92 21.86 3.1 18.6% 6-5.0 2.5-1.5
2Thiaw squad DEF · NEW £5.0m 3.18 17.90 2.2 1.7% 56.5-54.5 47.5-45.5
3Lacroix squad DEF · CHE £6.0m 3.60 17.59 2.8 9.5% 15.5-12.5 5.5-2.5
4Gabriel DEF · ARS £8.0m 3.14 17.52 4.0 29.7% 2+2.0 1+3.0
5Guéhi squad DEF · MCI £6.0m 3.33 16.96 2.8 18.0% 8-3.0 5.5-0.5
6Tarkowski DEF · EVE £6.0m 3.23 16.67 2.8 9.0% 18-12.0 5.5+0.5
7Truffert squad DEF · BOU £5.5m 2.48 16.32 2.5 4.6% 28-21.0 19-12.0
8Van Hecke DEF · TOT £5.0m 2.68 15.66 2.2 9.5% 15.5-7.5 47.5-39.5

Midfielders

Eight of 251 midfielders, numbered in our order. Ranks are over the whole 251; as many as 72 share one ep_next value.

Player Price GW1 xP 5‑GW xP ep_next Owned By ownership
rank · delta
By ep_next
rank · delta
1B.Fernandes squad MID · MUN £12.0m 4.82 23.00 4.0 51.1% 10.0 10.0
2Mbeumo squad MID · MUN £8.0m 4.52 21.55 2.8 36.6% 3-1.0 6-4.0
3Szoboszlai squad MID · LIV £7.0m 3.91 21.26 2.5 41.7% 2+1.0 14-11.0
4Anderson squad MID · MCI £6.5m 4.31 21.22 2.3 8.4% 17-13.0 25-21.0
5Palmer MID · CHE £9.5m 4.08 20.11 3.2 10.4% 14.5-9.5 2.5+2.5
6Semenyo MID · MCI £8.5m 4.07 19.45 2.9 26.3% 4+2.0 4+2.0
7Rice squad MID · ARS £7.5m 3.81 19.23 2.6 18.6% 70.0 9.5-2.5
8Gakpo MID · LIV £7.0m 3.37 18.98 2.5 3.6% 28-20.0 14-6.0

Forwards

Eight of 65 forwards, numbered in our order. Ranks are over the whole 65; as many as 15 share one ep_next value.

Player Price GW1 xP 5‑GW xP ep_next Owned By ownership
rank · delta
By ep_next
rank · delta
1Haaland FWD · MCI £15.5m 4.71 22.22 4.0 69.3% 10.0 10.0
2Thiago squad FWD · BRE £8.0m 3.77 17.14 2.5 17.4% 4-2.0 3.5-1.5
3Watkins FWD · AVL £8.0m 3.22 16.84 2.5 10.9% 7-4.0 3.5-0.5
4Evanilson squad FWD · BOU £6.0m 2.99 16.41 2.0 2.2% 15.5-11.5 17.5-13.5
5Calvert-Lewin FWD · LEE £6.0m 2.99 15.93 2.0 30.4% 3+2.0 17.5-12.5
6João Pedro FWD · CHE £7.5m 3.01 15.15 2.4 63.4% 2+4.0 60.0
7Gyökeres FWD · ARS £7.5m 3.00 14.59 2.4 9.6% 8-1.0 6+1.0
8Mateta FWD · CRY £6.5m 2.68 14.41 2.1 5.8% 10-2.0 9.5-1.5

Provenance

The hash below is of the data snapshot the solve consumed. The same snapshot and the same formula version return the same fifteen names.

Snapshot hash
b7f5cf2940cb5b25e30df1fec0216b7debb45f59cc9b68316c0656a7a1eebef4
Snapshot taken
2026-08-21T09:45:01Z
Fixtures hash
18745575078b4bbf1c3a1c85862945e8d5ff15825130c68b4d11124e2c7bbc54
Deadline
2026-08-21T17:30:00Z
Formula version
v1.0-phase0, rules 2026-27, fidelity A
Solver
highs — proven optimal, 13,986 columns × 8,491 rows over 561 candidates, solved in 1,867 ms
Objective
186.86809762841
What optimal means
Proven optimal means optimal given the formula — which is the thing actually on trial. The solver found the best squad the scoring function can describe; whether the scoring function describes football is a separate question, and the one the track record is for.
Team strength
chain[xgls>fdr] — 82 clubs priced by expected goals, 18 by the difficulty-rating fallback
Coverage
455 of 561 selectable players matched to history (81.1%); 89.4% at clubs the archive knows
Overrides applied
departed 38, doubtful 21, out-indefinite 44, out-with-date 7, suspended 3

Every figure on this page, at full precision

Every player named anywhere on this page — both squads and all thirty-two rows of the disagreement report, 38 distinct people — unrounded, with the nine components for each.

PlayerCode GW1 xP5‑GW xP P(start)P(60)P(play) appearancegoalsassistsclean sheetdef. contributionsavesgoals concededcardsshrink
B.Fernandes141746 4.818213 22.996021 0.9216 0.9000 0.9339 +1.8338+1.5937+1.4118+0.3552+0.1357+0.0000+0.0000-0.1358-0.3763
Mbeumo446008 4.522344 21.548135 0.8469 0.8469 0.9217 +1.7686+2.1836+0.5610+0.3343+0.0000+0.0000+0.0000-0.0850-0.2402
Virgil97032 3.917366 21.863943 0.9846 0.9846 0.9846 +1.9691+0.7048+0.1387+1.0827+0.7622+0.0000-0.6357-0.1045+0.0000
Szoboszlai424876 3.908611 21.255942 0.9069 0.8868 0.9507 +1.8374+0.7583+0.7270+0.2438+0.5037+0.0000+0.0000-0.1615+0.0000
Anderson215379 4.305016 21.219433 0.9443 0.9393 0.9964 +1.9356+0.5459+0.6112+0.2420+1.3594+0.0000+0.0000-0.2489-0.1403
Rice204480 3.809453 19.234768 0.8993 0.8889 0.9164 +1.8052+0.4733+0.6314+0.2874+0.7160+0.0000+0.0000-0.1038+0.0000
Raya154561 3.101838 18.038888 0.9528 0.9528 0.9528 +1.9056+0.0000+0.0089+1.2322+0.0000+0.5346-0.5379-0.0415+0.0000
Thiaw465642 3.178674 17.898815 0.9253 0.9016 0.9753 +1.8769+0.7154+0.0768+0.7380+0.5897+0.0000-0.7152-0.1029+0.0000
Lacroix437499 3.602392 17.589268 0.9011 0.8750 0.9011 +1.7762+0.4568+0.0805+0.9041+1.0819+0.0000-0.5922-0.1049+0.0000
Thiago502500 3.765919 17.139404 0.7334 0.7334 0.8169 +1.5503+2.0996+0.1755+0.0000+0.0628+0.0000+0.0000-0.1223+0.0000
Lammens465247 3.762660 16.619504 0.9711 0.9711 0.9711 +1.9422+0.0000+0.0063+1.5333+0.0000+0.7563-0.4513-0.0243+0.0000
Guéhi209036 3.325341 16.955643 0.8771 0.8771 0.8771 +1.7542+0.7331+0.2382+0.9038+0.4345+0.0000-0.5948-0.1435+0.0000
Evanilson444102 2.987232 16.410730 0.8608 0.7955 0.9400 +1.7355+1.1744+0.1417+0.0000+0.0000+0.0000+0.0000-0.0643+0.0000
Truffert494521 2.478710 16.321769 1.0000 1.0000 1.0000 +2.0000+0.3141+0.2536+0.5015+0.5931+0.0000-1.0382-0.1453+0.0000
Emegha503037 1.553475 7.802433 0.2580 0.2562 0.6074 +0.8636+0.6512+0.0955+0.0000+0.0000+0.0000+0.0000-0.0569+0.0000
Mitchell244723 2.873791 14.516805 0.8996 0.8423 0.9847 +1.8271+0.2856+0.2148+0.9392+0.2612+0.0000-0.5380-0.1161+0.0000
Coyle193278 1.096357 6.418043 0.3815 0.3624 0.5400 +0.9023+0.1233+0.0512+0.2441+0.1766+0.0000-0.3227-0.0785+0.0000
Tzolis439509 1.753798 8.787404 0.4277 0.3681 0.6135 +0.9816+0.4466+0.2502+0.1190+0.0289+0.0000+0.0000-0.0725+0.0000
E.Le Fée484420 3.868532 16.830921 0.8938 0.8938 0.9514 +1.8452+0.6275+0.6613+0.3489+0.4804+0.0000+0.0000-0.0947+0.0000
Haaland223094 4.710027 22.223946 0.7879 0.7647 0.8074 +1.5721+3.2048+0.3297+0.0000+0.0000+0.0000+0.0000-0.0702-0.3265
Walle Egeli557012 0.988937 4.717338 0.1542 0.1148 0.5939 +0.7087+0.2718+0.0471+0.0000+0.0000+0.0000+0.0000-0.0387+0.0000
Verbruggen489639 3.294279 15.707503 0.9960 0.9960 0.9960 +1.9920+0.0000+0.0090+1.1553+0.0000+0.8307-0.6165-0.0763+0.0000
Pickford111234 2.920630 15.246599 1.0000 1.0000 1.0000 +2.0000+0.0000+0.0164+0.9364+0.0000+0.8503-0.7260-0.1564+0.0000
Leno80201 2.861787 14.975320 1.0000 1.0000 1.0000 +2.0000+0.0000+0.0062+0.8689+0.0000+0.8274-0.7634-0.0773+0.0000
Donnarumma204936 2.672288 14.346811 0.9360 0.9360 0.9360 +1.8721+0.0000+0.0324+0.9645+0.0000+0.6491-0.6348-0.2110+0.0000
Martinez98980 2.265094 13.944594 0.8741 0.8275 0.8741 +1.7016+0.0000+0.0074+0.5683+0.0000+0.7988-0.7290-0.0820+0.0000
Sels85633 2.904911 13.654368 0.8558 0.8350 0.8724 +1.7073+0.0000+0.0128+0.9405+0.0000+0.8470-0.5291-0.0736+0.0000
Gabriel226597 3.135530 17.515395 0.7481 0.7307 0.8101 +1.5408+0.4974+0.1399+0.9449+0.5422+0.0000-0.4125-0.1172+0.0000
Tarkowski17761 3.226295 16.665570 0.8976 0.8854 0.8976 +1.7831+0.3265+0.1589+0.8291+0.9849+0.0000-0.6428-0.2134+0.0000
Van Hecke469142 2.676142 15.661362 0.9311 0.9311 0.9340 +1.8651+0.4289+0.1197+0.6169+0.6793+0.0000-0.8370-0.1966+0.0000
Palmer244851 4.076730 20.109050 0.8212 0.7886 0.8504 +1.6390+1.8961+0.4014+0.2037+0.1460+0.0000+0.0000-0.1743-0.0353
Semenyo437730 4.070984 19.449306 0.9722 0.8820 0.9858 +1.8678+1.6521+0.3939+0.2272+0.1143+0.0000+0.0000-0.1517-0.0326
Gakpo243298 3.370016 18.977314 0.8422 0.8019 0.9484 +1.7502+1.1052+0.3509+0.2204+0.0527+0.0000+0.0000-0.1095+0.0000
Watkins178301 3.218899 16.836686 0.8522 0.8154 0.9791 +1.7945+1.4489+0.1129+0.0000+0.0000+0.0000+0.0000-0.1374+0.0000
Calvert-Lewin177815 2.991832 15.932748 0.7249 0.7180 0.8539 +1.5719+1.4536+0.0765+0.0000+0.0000+0.0000+0.0000-0.1101+0.0000
João Pedro475168 3.013375 15.145255 0.7042 0.6858 0.7953 +1.4812+1.4931+0.1765+0.0000+0.0000+0.0000+0.0000-0.1374+0.0000
Gyökeres224117 2.998638 14.592987 0.6047 0.5212 0.9724 +1.4936+1.4226+0.1950+0.0000+0.0000+0.0000+0.0000-0.1125+0.0000
Mateta231747 2.676367 14.410755 0.5480 0.4401 0.8474 +1.2875+1.3052+0.1152+0.0000+0.0000+0.0000+0.0000-0.0315+0.0000

Scrolls sideways.