Fantasy Premier League
Who defends for points
About half a defender's season comes from a clean sheet his club keeps and an award he clears on his own, and no board on this site has ever printed either. Every defender and midfielder the game would let you field, ranked on the defensive half of the round ahead, with the rate behind the award, the clean-sheet probability behind the sheet and what he does at the other end printed beside them.
Gabriel Magalhães
DEF · ARS · 8.0m · owned by 23%
1.75
Def pts
The award rests on 8.1 defensive contributions per 90, shrunk towards what players in his position actually do, against the 10 actions the game pays for.
Gameweek 6 · 3-round window · feed of 01/10 11:39
Gabriel Magalhães is the best defensive return on the board: 1.75 projected defensive points in gameweek 6 as a defender, of which 0.44 is the contribution award, on a 41% chance of a clean sheet.
The award rests on 8.1 defensive contributions per 90, shrunk towards what players in his position actually do, against the 10 actions the game pays for. Of the 163 your filters leave: 163 defenders and 0 midfielders, ranked against each other.
Def pts
1.75
The clean sheet and the contribution award together, in the round ahead, which is what this board is ranked on.
DC/90
0.44
What the contribution award is worth to him: how often the model has him clearing 10 in a match, priced.
CS
41%
How often the model has his side keeping the sheet clean in the round ahead.
The top five on this board
The board as your controls leave it — the same five that open the table below.
-
Ranked 1 Gabriel Magalhães1.75 Def pts
- DC/90
- 8.1
- CS
- 41%
- BPS/90
- 18.0
2.3- LEE (H)
- NFO (A)
- EVE (H)
-
Ranked 2 John Egan1.69 Def pts
- DC/90
- 11.4
- CS
- 25%
- BPS/90
- 14.6
3.0- EVE (H)
- FUL (A)
- BRE (H)
-
Ranked 3 Bobby Thomas1.56 Def pts
- DC/90
- 10.2
- CS
- 28%
- BPS/90
- 16.1
2.7- NEW (H)
- TOT (A)
- FUL (H)
-
Ranked 4 Chris Richards1.50 Def pts
- DC/90
- 10.3
- CS
- 25%
- BPS/90
- 9.6
3.0- NFO (H)
- BHA (A)
- NEW (H)
-
Ranked 5 Murillo1.38 Def pts
- DC/90
- 10.1
- CS
- 24%
- BPS/90
- 15.2
3.0- CRY (A)
- ARS (H)
- IPS (A)
- DC/90
- 8.1
- DC pts
- 0.44
- xGI
- 0.14
- BPS/90
- 18.0
- Start
- 92%
- DC/90
- 10.2
- DC pts
- 0.86
- xGI
- 0.16
- BPS/90
- 16.1
- Start
- 92%
- DC/90
- 10.3
- DC pts
- 0.95
- xGI
- 0.03
- BPS/90
- 9.6
- Start
- 92%
- DC/90
- 10.6
- DC pts
- 1.03
- xGI
- 0.23
- BPS/90
- 6.0
- Start
- 92%
- DC/90
- 4.6
- DC pts
- 0.01
- xGI
- 0.27
- BPS/90
- 16.4
- Start
- 92%
- DC/90
- 10.3
- DC pts
- 0.95
- xGI
- 0.03
- BPS/90
- 19.4
- Start
- 92%
- DC/90
- 10.9
- DC pts
- 1.11
- xGI
- 0.11
- BPS/90
- 20.2
- Start
- 92%
- DC/90
- 9.8
- DC pts
- 0.83
- xGI
- 0.05
- BPS/90
- 27.6
- Start
- 92%
- DC/90
- 6.4
- DC pts
- 0.16
- xGI
- 0.13
- BPS/90
- 10.0
- Start
- 92%
- DC/90
- 9.6
- DC pts
- 0.79
- xGI
- 0.05
- BPS/90
- 14.6
- Start
- 92%
- DC/90
- 9.3
- DC pts
- 0.71
- xGI
- 0.07
- BPS/90
- 19.2
- Start
- 92%
- DC/90
- 10.8
- DC pts
- 0.65
- xGI
- 0.09
- BPS/90
- 6.9
- Start
- 75%
- DC/90
- 12.2
- DC pts
- 0.61
- xGI
- 0.02
- BPS/90
- 10.1
- Start
- 75%
- DC/90
- 11.1
- DC pts
- 0.71
- xGI
- 0.26
- BPS/90
- 11.0
- Start
- 75%
- DC/90
- 12.5
- DC pts
- 0.48
- xGI
- 0.04
- BPS/90
- 4.8
- Start
- 58%
- DC/90
- 8.8
- DC pts
- 0.59
- xGI
- 0.08
- BPS/90
- 10.6
- Start
- 92%
- DC/90
- 7.4
- DC pts
- 0.04
- xGI
- 0.08
- BPS/90
- 11.5
- Start
- 58%
- DC/90
- 5.5
- DC pts
- 0.02
- xGI
- 0.09
- BPS/90
- 7.3
- Start
- 75%
- DC/90
- 9.9
- DC pts
- 0.40
- xGI
- 0.31
- BPS/90
- 12.8
- Start
- 75%
- DC/90
- 7.8
- DC pts
- 0.37
- xGI
- 0.04
- BPS/90
- 21.8
- Start
- 92%
- DC/90
- 8.6
- DC pts
- 0.55
- xGI
- 0.12
- BPS/90
- 18.2
- Start
- 92%
- DC/90
- 8.3
- DC pts
- 0.23
- xGI
- 0.04
- BPS/90
- 9.5
- Start
- 75%
- DC/90
- 8.7
- DC pts
- 0.51
- xGI
- 0.10
- BPS/90
- 20.2
- Start
- 92%
- DC/90
- 7.8
- DC pts
- 0.27
- xGI
- 0.03
- BPS/90
- 24.3
- Start
- 92%
- DC/90
- 5.8
- DC pts
- 0.06
- xGI
- 0.23
- BPS/90
- 14.6
- Start
- 92%
- DC/90
- 6.1
- DC pts
- 0.12
- xGI
- 0.20
- BPS/90
- 17.0
- Start
- 92%
- DC/90
- 5.7
- DC pts
- 0.00
- xGI
- 0.13
- BPS/90
- 22.3
- Start
- 56%
- DC/90
- 6.2
- DC pts
- 0.00
- xGI
- 0.01
- BPS/90
- 27.9
- Start
- 42%
- DC/90
- 9.4
- DC pts
- 0.27
- xGI
- 0.09
- BPS/90
- 6.2
- Start
- 69%
- DC/90
- 7.0
- DC pts
- 0.07
- xGI
- 0.01
- BPS/90
- 4.5
- Start
- 75%
- DC/90
- 6.9
- DC pts
- 0.02
- xGI
- 0.08
- BPS/90
- 1.8
- Start
- 56%
- DC/90
- 7.6
- DC pts
- 0.14
- xGI
- 0.02
- BPS/90
- 3.0
- Start
- 75%
- DC/90
- 6.6
- DC pts
- 0.16
- xGI
- 0.20
- BPS/90
- 28.0
- Start
- 92%
- DC/90
- 7.1
- DC pts
- 0.08
- xGI
- 0.04
- BPS/90
- 14.7
- Start
- 75%
- DC/90
- 4.9
- DC pts
- 0.04
- xGI
- 0.12
- BPS/90
- 5.6
- Start
- 92%
- DC/90
- 7.3
- DC pts
- 0.12
- xGI
- 0.08
- BPS/90
- 9.3
- Start
- 75%
- DC/90
- 12.4
- DC pts
- 0.27
- xGI
- 0.12
- BPS/90
- 24.9
- Start
- 75%
- DC/90
- 5.1
- DC pts
- 0.02
- xGI
- 0.21
- BPS/90
- 9.5
- Start
- 92%
- DC/90
- 6.7
- DC pts
- 0.15
- xGI
- 0.11
- BPS/90
- 13.6
- Start
- 92%
- DC/90
- 5.9
- DC pts
- 0.00
- xGI
- 0.03
- BPS/90
- -2.6
- Start
- 25%
- DC/90
- 7.8
- DC pts
- 0.01
- xGI
- 0.01
- BPS/90
- 27.0
- Start
- 50%
What these columns are, and which of them is measured
Nothing on this page is derived here. Every column is a component the round-ahead projection has computed since the model was written — the clean sheet off the modelled scoreline, the concession off the same distribution, and the contribution award off the rate below — and the total beside them is those components added up rather than a second opinion about them.
The per-90 rate is shrunk towards what players in that position actually do, in proportion to how little football is behind it. A defender with one busy half has a raw per-90 above anybody's and would head a table sorted on it; that failure is why the shrinkage exists, and it is why the number here is smaller than the one the game publishes for him.
The award does not stack and it is not a rate: it is a threshold cleared once, 10 actions for a defender and 12 for a midfielder, which the game counts differently for the two — a midfielder's recoveries are in his total and a defender's are not. What the column holds is how likely he is to clear it, priced at what the game pays.
A clean sheet here is the probability that the opponent scores exactly zero, off the same fitted scoreline the rest of this site prices matches with. It is not one minus an average, and it is the reason this model exists rather than reading the game's own one-to-five fixture rating.
A dash in the rate column means the game publishes no defensive contribution total for that player at all. It is not a nought: nought would say he has produced none, where the dash says the figure is not in the feed.
The bonus column is the one measured figure on the board — his stored bonus points system total over the minutes he played for it, highlighted above 25 per 90, which is roughly where the three-per-match award starts arriving. This site projects bonus nowhere, so it is a reading of his season rather than a claim about the round.
Goalkeepers and forwards are not here. A goalkeeper cannot earn the contribution award at all and a forward is paid nothing for a clean sheet, so neither has a defensive half this board could rank him on.
Every ranking here is in expected points or in the weighted score for the rounds ahead, which are the units this model is measured in. The value column is those figures restated over a price — it is a way of scanning the board, and nothing on this site backtests it.
The board is every player the game would let you field, the flagged included — whether a doubtful player is worth picking is a judgement the projection makes with his published chance of playing, not one made by hiding him.
What is measured here, and what is not
-
Measured The projected points
Replayed against gameweek 2 after it was played: an average miss of 1.617 points a player, against 1.842 for “predict what he has averaged so far”, over 456 players. Recorded Sunday, 6 September 2026 and refreshed by hand, so the date is part of the claim.