Notes · 9 November 2026
Gameweek 10 scored: Barry got 16, Haaland got 2 again
The best week the list has had, by a distance. The twenty averaged more than six points each. Barry, eighth on the list at 8% owned, got 16 and shared the top score in the game. Haaland was top of the list and got 2, for the second week running.
The locked file was written at 06:38 UTC on Friday, 11.9 hours before the deadline. The backstop got there first even with an evening deadline. It only runs once a day, so the morning run on deadline day is always its last chance, and it takes it. A bigger margin than the two hours I aim for, but the file is locked and the margin is written in it.
All ten games are played and the bonus is in. FPL has not marked the week as final yet, and some of its own numbers are clearly stuck from the middle of the weekend (more on that below). The player scores are what they are, so nothing in the table should move.
The list beat the baseline by the most so far
The twenty players averaged 6.15 points. The 177 players the model expects to play an hour or more averaged 3.55. That is 2.60 points a player above the baseline. Last week was +1.41 and I called that the best week. This is almost double it.
Against the 204 players who actually played 60 minutes, who averaged 3.85, it is still 2.30 ahead. And before anyone says it was all Barry: take his 16 out and the other nineteen averaged 5.63. Still the best week by itself.
Last week I said I would keep an eye on the gap between what the twenty were projected and what they got, and that if it was still above a point after this week the model needed fixing. The twenty were projected 4.78 and got 6.15. That is the second week in a row the list has scored more than the model said it would. So that question is answered, at least for now. The early weeks were overconfident, the last two were not.
The captain: Haaland, 2
Top of the locked list was Haaland at 5.59, just ahead of Gabriel at 5.36. City lost 3-1 at Forest and he got 2. He was the most captained player in the game again, so it is a draw again, 4 against 4.
The rest of the captain list: Gabriel 8, White 0, Calafiori 6, Hill 4. Gabriel would have doubled to 16, which a lot of people did. The model had them a quarter of a point apart, and it went with the top number, like it always does.
That makes six scored weeks where the top captain pick has never beaten the most captained player. Four draws, two losses. Before doubling, the top pick has 29 points and the crowd's captain has 40. All of the gap is still from the two weeks it picked someone other than Haaland.
Every pick, scored
| # | Player | Team | xP | Actual |
|---|---|---|---|---|
| 1 | Haaland | MCI | 5.59 | 2 |
| 2 | Gabriel | ARS | 5.36 | 8 |
| 3 | White | ARS | 5.17 | 0 |
| 4 | Calafiori | ARS | 5.12 | 6 |
| 5 | Hill | BOU | 4.96 | 4 |
| 6 | Tarkowski | EVE | 4.93 | 2 |
| 7 | Evanilson | BOU | 4.85 | 9 |
| 8 | Barry | EVE | 4.81 | 16 |
| 9 | Tavernier | BOU | 4.80 | 3 |
| 10 | Muharemović | LEE | 4.71 | 10 |
| 11 | Isak | LIV | 4.68 | 8 |
| 12 | Vuskovic | BHA | 4.67 | 12 |
| 13 | Mbeumo | MUN | 4.63 | 2 |
| 14 | Raya | ARS | 4.63 | 6 |
| 15 | De Cuyper | BHA | 4.54 | 6 |
| 16 | Rogers | CHE | 4.47 | 2 |
| 17 | Dewsbury-Hall | EVE | 4.47 | 9 |
| 18 | Silva | BOU | 4.46 | 3 |
| 19 | Ødegaard | ARS | 4.44 | 11 |
| 20 | Jacquet | LIV | 4.39 | 4 |
| Mean | 4.78 | 6.15 |
Eleven of the twenty got six points or more. Five got two or fewer. Last week it was nine and eight. White's zero stays in the table. He did not play and had no flag on Friday morning, and a player who never plays counts as zero here, otherwise the model would never miss.
Fifteen players in the whole game got 10 or more this week. Four of them were on the list: Barry 16, Vuskovic 12, Ødegaard 11 and Muharemović 10.
What went right
Everton at home to Coventry. Three Everton picks projected 14.2 between them. Everton won 3-1 and they got 27. Barry scored twice, set one up and took all three bonus points for 16, level with Alysson for the top score in the game. Dewsbury-Hall scored as well and got 9. Last week I wrote that Barry had been scoring below his xG and the model kept backing the xG. Two weeks later he has 25 points from them.
Arsenal at home to Hull. Five Arsenal players on the list, the same kind of pile-in I was nervous about last week. It held up. Arsenal won 1-0 and the four who played got 31, against 19.6 projected for those four. Ødegaard scored the only goal and took the full bonus for 11. Gabriel kept the clean sheet with ten defensive actions for 8. Calafiori and Raya got 6 each. White is the fifth one, covered below.
Brighton at home to Brentford. Vuskovic got an assist, a clean sheet and three bonus for 12. De Cuyper got the clean sheet for 6.
Muharemović. Leeds beat Spurs 1-0, and he had 14 defensive actions, a clean sheet and two bonus for 10. Last week he made 12 defensive actions and got 3 because Leeds conceded twice. Same player, same work, very different score.
Isak. Scored in the 3-1 at Palace and got 8, after 2 last week.
What went wrong
Haaland. Covered above. A 3-1 defeat at Forest and 2 points from the top pick, again.
White. Third on the list at 5.17 and did not play a minute. There was no flag when the file locked, so the model had no way of knowing, but it still costs a quarter of a point off the average.
Mbeumo in a 4-4. United and Villa scored eight between them and Mbeumo got 2. Hard to see how you get 2 out of that, but he managed.
Rogers and Tarkowski. Chelsea lost 1-0 at Sunderland and Rogers got 2. Tarkowski got 2 in the same game Barry got 16. Everton conceded once, and that was enough to take his clean sheet.
Bournemouth, exactly as projected. Four Bournemouth picks projected 19.1 between them. They won 2-1 at Ipswich and got 19. Evanilson scored and took the bonus for 9. Hill had 14 defensive actions and still only got 4 after the goal they conceded. Not a miss, not a win, just the model being right on the average.
Against the crowd. The locked file left out Groß at 47% owned and backed Hill at 2.6% instead. Groß scored, kept a clean sheet and got 10. Hill got 4. That ends a two-week run of the against-the-crowd pick winning.
The model's own team
The Deadline FPL team in the official game, picked by the wildcard planner and captaining the top projection, had 63 points live this morning. Belloumi did not play, so Groß should come off the bench for him when FPL runs the automatic subs, and that makes it 73.
FPL itself still shows 28 for the team and 25 as the average manager score. Those are clearly numbers from the middle of the weekend that have not been updated. I am going on the live points per player, and I will correct it here if the final number is different.
The bench after the sub: Trafford 10, Vuskovic 12, Thomas 0. Twenty-two points left on it, and Vuskovic's 12 hurts the most because he was on the list as well. No transfer was made before Gameweek 10, so the team goes into Gameweek 11 with two free transfers.
The record so far
| GW | List avg | Likely starters | Edge | Top captain | Crowd captain |
|---|---|---|---|---|---|
| 4 | 4.20 | 3.84 | +0.36 | Isak 2 | Haaland 9 |
| 5 | Not captured. What happened | ||||
| 6 | 3.65 | 3.35 | +0.30 | Saka 2 | Haaland 6 |
| 7 | 3.40 | 3.66 | −0.26 | Haaland 13 | Haaland 13 |
| 8 | 4.30 | 3.56 | +0.74 | Haaland 8 | Haaland 8 |
| 9 | 4.80 | 3.39 | +1.41 | Haaland 2 | Haaland 2 |
| 10 | 6.15 | 3.55 | +2.60 | Haaland 2 | Haaland 2 |
Across the six scored weeks the list is ahead by about 0.86 points a player, five weeks out of six, and the edge has gone up every week since Gameweek 7. Same caveat as before: "likely starters" is filtered on what the model expects now, not what it expected before the week, so the baseline is a bit soft. Not soft enough to explain 2.60.
The honest summary
Three good weeks in a row, and the last two the list has beaten its own projection. I said I would call it after this week, so here it is: I am not changing how the model projects points. I would be more worried if it was beating the projection by this much every week, because that would mean it is too cautious. One week of 6.15 is a good week, not a new normal.
The part that still does not work is the captain. Six weeks, never better than the crowd. When it picks Haaland it draws, and when it picks someone else it has lost. That is not much of a case for following its captain pick over just picking Haaland.
Last week I worried about the list backing both ends of the same game. It did not happen this week. The five Arsenal players were the other risk, and they came good.
Now an international break. Gameweek 11's deadline is Saturday 21 November at 11:00 UK time. Early numbers have Haaland on top again at home to Fulham, then Isak, Evanilson, Van Hecke and Vuskovic. The numbers lock before the deadline and get scored the same way.