Gameweek Debrief 9 min read 1,776 words

Gameweek 4 Debrief: winners, blanks, and what the underlying numbers say

Groß led Gameweek 4 with 17 points against an average of 69. Underneath the scoreline, the expected-goals numbers point at a different set of names.

· Generated from Gameweek 4 player history and FPL's published round summary

The headline

Groß was the story of Gameweek 4: 17 points from 1 goal and 2 assists away at COV, in 90 minutes. He went into the round owned by 18.6% of managers.

The round in numbers

These are the figures FPL publishes about the gameweek itself rather than about any one player, and they set the bar every squad is measured against.

MetricValueWhat it means
Average score69What the field scored
Highest score151The best single squad in the game
Managers ranked10,673,913Squads that scored this round
Most captainedHaaland72.3% owned, returned 9
Most selectedJoão Pedro75.3% owned

Chips played

ChipManagersShare of the field
Wildcard760,6827.1%
Free Hit328,7403.1%
Triple Captain323,2263%
Bench Boost170,0701.6%

Wildcard was the round's most-played chip. A chip played in the same week as everybody else converts a good week into an average one relative to the field, which is the argument for holding until your own squad is ready rather than until the template says go.

Who actually scored

PlayerTeamPosPtsMinsGABonusxGIOwned
GroßBHAMID17901231.0118.6%
SchadeBREMID15902030.315.9%
BogleLEEDEF15821030.463%
RayaARSGK1490003040.4%
DavisIPSDEF14901210.716.1%
BelloumiHULMID13592020.233%
DunkBHADEF12901000.061.5%
João PedroCHEFWD12901130.3375.3%
EmersonnIPSFWD12621130.492.6%
De CuyperBHADEF11900120.0723.6%
MykolenkoEVEDEF11900030.073%
GvardiolMCIDEF11900130.0222.9%

The scoreline against the underlying data

Expected goals measure the chances a player got into. Goals measure what he did with them. Over one round the gap between the two is mostly noise, but it is the right place to look for the returns that have not landed yet.

Beat their chances

PlayerTeamGoalsxGDifferenceMins
SchadeBRE20.21+1.7990
TavernierBOU10.2+0.882
KluivertBOU10.21+0.7965
RogersCHE10.21+0.7990
KhalailiCRY10.22+0.7890
João PedroCHE10.29+0.7190

Created more than they took

PlayerTeamGoalsxGDifferenceMins
E.Le FéeSUN00.85-0.8590
MukieleSUN00.7-0.790
NdoyeNFO00.7-0.774
BarryEVE00.67-0.6790
SilvaBOU00.58-0.5890
IsakLIV00.5-0.590

E.Le Fée generated 0.85 expected goals and scored 0. That is the kind of return that tends to arrive late rather than never — provided the chances keep coming.

The defensive audit

7 teams kept a clean sheet in Gameweek 4: ARS, BHA, EVE, FUL, LIV, MCI, TOT. Clean sheets are worth four points to a defender and a goalkeeper, which makes the defensive picture worth as much attention as the attacking one.

Goalkeepers

PlayerTeamPricePtsSavesConcededOwned
RayaARS£6.0m142040.4%
LenoFUL£4.5m9303%
A.BeckerLIV£5.5m9303.9%
VerbruggenBHA£4.5m86022%
DonnarummaMCI£5.5m7307.6%
KinskyTOT£4.5m73018.5%

A goalkeeper behind a leaky defence is not automatically a bad pick: save points accumulate in exactly the weeks a clean sheet does not arrive, which is why the busiest keeper in the division is often better value than the one behind the best defence.

Defenders

PlayerTeamPricePtsBonusxGIOwned
BogleLEE£4.5m1530.463%
DavisIPS£4.0m1410.716.1%
DunkBHA£4.5m1200.061.5%
De CuyperBHA£4.8m1120.0723.6%
MykolenkoEVE£4.5m1130.073%
GvardiolMCI£5.7m1130.0222.9%

Attacking output by team

TeamScoredConcededResult
BHA50Won
LEE41Won
IPS32Won
ARS20Won
BOU22Drew
BRE22Drew
CHE22Drew
CRY23Lost

BHA scored the most goals of the round with 5. Team-level attacking output is the cleanest signal available after a single round: it takes eleven players to create a goal and one to finish it, and the finisher changes far more often than the creating does.

The bonus point picture

Bonus is decided by the Bonus Points System, which rewards goals and assists heavily but also pays for tackles, recoveries, clean sheets and passing accuracy. Over a season it is worth roughly a point a game to the players who consistently rank in it, and it is the most overlooked component of a captaincy decision.

PlayerTeamPosBonusBPSPtsOwned
GroßBHAMID3561718.6%
SchadeBREMID349155.9%
EmersonnIPSFWD345122.6%
BogleLEEDEF345153%
João PedroCHEFWD3441275.3%
Gibbs-WhiteNFOMID342811.8%
RayaARSGK3381440.4%
MykolenkoEVEDEF338113%
JacquetLIVDEF33693.1%
GvardiolMCIDEF3361122.9%

Groß took the maximum with 3 from a BPS score of 56. Defenders and holding midfielders who accumulate BPS through defensive actions are the ones worth identifying early, because that route to bonus is far more repeatable than the one that runs through scoring goals.

The premium watch

A premium at £10m or more is not competing with the other forwards. He is competing with the two mid-price players you could field instead, and with the 72.3% of the field who already own the most popular of them.

PlayerTeamPricePtsMinsGAxGIOwned
HaalandMCI£15.5m990100.8872.3%
B.FernandesMUN£12.0m290000.3641.7%
PalmerCHE£9.7m590010.3826.8%
IsakLIV£9.1m290000.522.1%
SakaARS£9.5m890101.4612.1%

Haaland led the bracket with 9. B.Fernandes returned 2, and at £12.0m that is the single largest hole a squad can have in a week. Not owning the most-selected premium is an active bet against 72.3% of the field — a legitimate one, but it should be deliberate.

Enablers and budget gems

The cheapest player who actually starts is worth more than a better player who does not. An enabler's job is to cost £4.5m and play ninety minutes; everything past that is profit that funds the rest of the squad.

PlayerTeamPosPricePtsMinsOwned
BogleLEEDEF£4.5m15823%
DavisIPSDEF£4.0m14906.1%
DunkBHADEF£4.5m12901.5%
EmersonnIPSFWD£5.5m12622.6%
De CuyperBHADEF£4.8m119023.6%
MykolenkoEVEDEF£4.5m11903%
KostoulasBHAFWD£5.5m10830.7%
LenoFULGK£4.5m9903%

Bogle at £4.5m returned 15 points from 82 minutes, on 3% ownership. Enablers at that price are how squads afford two premiums without hollowing out the bench.

The differentials who paid

Points from a player most of the field does not own are worth more than the same points from one they do. These returned at least six while owned by under 8% of managers, which is where rank is actually won.

PlayerTeamPosPricePtsOwnedMins
SchadeBREMID£6.0m155.9%90
BogleLEEDEF£4.5m153%82
DavisIPSDEF£4.0m146.1%90
BelloumiHULMID£5.0m133%59
DunkBHADEF£4.5m121.5%90
EmersonnIPSFWD£5.5m122.6%62
MykolenkoEVEDEF£4.5m113%90
KostoulasBHAFWD£5.5m100.7%83

Schade at 5.9% ownership returned 15. The managers who owned him gained ground on almost the entire field in a single afternoon — that is the whole case for carrying one or two genuine differentials rather than a squad indistinguishable from the template.

Is Groß worth chasing?

The reflex after a haul is to buy it, and the reflex is usually wrong — you pay a risen price for a performance that has already happened. The question worth asking is not what he just did but what is in front of him:

GW5ARS (H)GW6SUN (A)GW7CRY (H)GW8LIV (A)

An average difficulty of 3.5 across his next four. That is an ordinary run — not a reason to buy and not a reason to avoid, which usually means the transfer is better spent elsewhere.

The early transfer radar

Four calls off the back of this round, with the reasoning attached so you can disagree with the reasoning rather than the conclusion.

BUY: E.Le Fée (SUN, £5.8m)1.44 expected involvements in 90 minutes and still only 4% owned.
BUY: Gibbs-White (NFO, £7.9m)1.07 expected involvements in 90 minutes and still only 11.8% owned.
HOLD: Mbeumo (£7.9m)2 points reads badly, but 0.57 expected involvements says the process was fine. Selling here sells the scoreline, not the player.
SELL: Shaw (£4.4m)Unspecified injury - 75% chance of playing

Looking ahead to Gameweek 5

FixtureHome difficultyAway difficulty
BRE v CHE4.03.0
TOT v AVL3.03.0
BHA v ARS4.03.0
EVE v IPS2.03.0
NEW v HUL2.03.0
NFO v COV2.03.0
BOU v LIV4.03.0
LEE v CRY3.03.0
MCI v SUN2.05.0
FUL v MUN4.03.0

EVE have the kindest assignment of the round on difficulty alone. That is a starting point for captaincy rather than an answer — difficulty ratings know nothing about who is injured.

Every figure in this piece was read from the FPL dataset at the moment it was published, and has not been edited since.