Fantasy GW 21 – A 5.5 goalkeeper is the round's third-best buy
Last week the point was that the transfer window had closed and that the dead assets were therefore permanent. This week the point is that the window only closed in one direction. Erik Botheim was sold from Malmö to Viking on 2 September. Noah Christoffersson was sold from Örgryte to Thun on 1 September. Swedish clubs cannot buy, but they can still sell to leagues where the window remains open, and the fantasy report has already zeroed both. Botheim, ninth in the xG table with 0.48 per match, is owned by 7.0 per cent and has 0.0 expected fantasy points. Christoffersson, Örgryte's best fantasy asset with 85 points, is owned by 6.6 per cent and carries the same zero.
They join the three already standing there. Robbie Ure is owned by 28.8 per cent, down from 29.9 last week. Elliot Stroud by 15.5, down from 15.7. Rony Jansson by 4.2, down from 4.6. On the three comparable names the market has corrected by 1.7 percentage points in a week, against 3.7 the week before. The correction is slowing, not accelerating, even though the verdict is now final. In total, 62.1 percentage points of ownership, counted across the league's 58,167 teams, sit in players who cannot deliver a point. Ure alone remains in roughly 16,750 squads.
The report's own transfer planner shows how hard it is to keep a season database current. The example of a suggested transfer advises you to sell Bersant Celina and bring in Sagoe Jr, for plus twelve expected points over five rounds. Celina left AIK for Al-Ettifaq on 5 August. It is correct advice for the wrong reason, and anyone who follows it gains more than twelve points.
A 5.5 goalkeeper is the round's third-best buy
Round 21 contains nine matches. The rearranged fixture between AIK and Mjällby is played on Wednesday 16 September and counts, which gives both teams two matches in the same fantasy round. The model responds by lifting Kristoffer Nordfeldt to 8.0 expected points, third in the whole round behind Lind and Bjerkebo on 9.7. Nordfeldt costs 5.5 and is owned by 18.1 per cent. He returned ten points last round with a clean sheet in Malmö, leads Allsvenskan for goals prevented with 5.9, and faces Västerås and Mjällby at home, two teams that create 1.1 and 1.2 expected goals per match respectively.
The double gameweek also lifts players who would otherwise not have been seen. Jacob Bergström of Mjällby has 6.7 expected points at a price of 6.2 and 19.7 per cent ownership, with two away matches against Häcken and AIK. Jeppe Kjær of the same team has 5.8 at 1.6 per cent ownership, the lowest among the round's ten highest-ranked players. Linus Carlstrand, the most bought player before last round, has 6.0 and is owned by 15.9 per cent. Hervé Matthys of AIK sits on 5.1 at 2.9 per cent, and Johan Hove on 5.1 at 8.6.
Two things complicate the picture, and neither is in the report. Mjällby sacked Karl Marius Aksum on Tuesday and go into the double gameweek with Andreas Brännström as their new coach. And AIK announced on Monday that Kevin Filling is out for the rest of the year with a knee injury. Filling remains in the report's list of AIK's top fantasy assets with 6.5 expected points, and he will not score a single one of them. He is owned by 3.7 per cent. It is a small number, but it shows that the model's zeros lag by a week even when the news is public.
Hammarby fill the top list again, and this time the opponent is at home
Last round Hammarby accounted for five of the seven highest-ranked players before the away match at Örgryte. It finished 1–1 and 39 fantasy points, fewer than Örgryte collected in the same match. The model does it again. Victor Stange Lind tops the round with 9.7, Hampus Skoglund is fourth with 7.9, Montader Madjed fifth with 7.4, Frederik Winther sixth with 7.2 and Paulos Abraham ninth with 5.9. Five of the top nine wear the same shirt for the second week running, this time ahead of Brommapojkarna at home, a team Hammarby have beaten ten times out of thirteen.
Ownership has moved a little, and only partly in the model's direction. Lind has gone from 27.4 to 31.1 per cent after eight points as the round's most captained player. Abraham has gone from 39.3 to 41.8. Skoglund has risen from 15.9 to 16.8 and Winther has fallen from 24.1 to 23.8. Madjed stays on 7.3 per cent, the same figure as last week, even though he is now ranked the round's best differential and sits second in the whole league for progressive carrying. Noah Persson at 2.9 per cent has 5.2 expected points.
The people's choice still points to Stockholm but to a different club. Bo Hegland is owned by 52.6 per cent, up from 52.3, and has 4.7 expected points at home to Gais. Kristian Lien is owned by 39.1 and has 4.5. Djurgården have won four in a row with four clean sheets, and Hegland leads the assist table with 0.76 per match. Neither is wrong. But between Hegland's 4.7 and Lind's 9.7 lie five expected points, and that is the difference between the player half the league owns and the player the model actually wants.
Bjerkebo loses owners
Isak Bjerkebo has 9.7 expected points, joint highest in the round, at home to Degerfors who create 0.92 expected goals per match, the lowest in the league. He returned three points last round as the most selected player, and his ownership has fallen from 66.8 to 64.7 per cent after holding steady all August. Two percentage points is not a shift. But the market is selling him in the same week he extended to 2030 and the model ranks him highest.
The player the market bought instead of points was Sam Larsson. Two goals and an assist against Mjällby returned 15 points, the round's highest, and he was owned by 5.1 per cent when it happened. Tobias Heintz returned ten at 8.0 per cent. IFK Göteborg were the round's best real fantasy team with 71 points, and host Halmstad on Saturday, the team with the league's worst defence by expected goals against. Larsson leads the xG table with 0.75 per match. The report does not have him among the round's ten highest-ranked players, and that is either a warning or a buying opportunity.
The wildcard was the week's chip, not the duo
The chips tell you what the league did after the window closed. In round 19, 5,697 Dynamic Duos and 955 wildcards were played. In round 20, 445 Dynamic Duos and 2,342 wildcards. Read the two lines together and you see a market that realised the squad needed rebuilding and did it at once, most likely to get rid of exactly the names carrying zeros above. That is the right reaction. The strange thing is that 62.1 percentage points still have not done it.
Round 20 returned an average of 38 points, down from 50, and a top score of 97. The team of the round contained three AIK players after the 0–0 in Malmö, the goalkeeper and two defenders, and among the round's ten top scorers was a goalkeeper in Etrit Berisha who is owned by 0.8 per cent and returned nine points. That says something about the kind of round it was.
The double gameweek is the rare situation where the model and common sense say the same thing. Nordfeldt at 5.5 with two home matches, Bergström at 6.2 with two away matches under a coach who got the job on Tuesday. It is not a complicated analysis, and that is why it is worth taking seriously. The weeks when the report points at something everyone can see are the weeks when there is no reason not to act on it.
The deadline is Friday at 18.55. Six of the names in this text can no longer score in Allsvenskan Fantasy, and a seventh will not. The market corrected by 1.7 percentage points last week. At that rate, Ure will still have owners in round 30.
Fantasy round 21 fixtures: Häcken–Mjällby (Fri 11/9 19.00), AIK–Västerås (Sat 12/9 15.00), Göteborg–Halmstad (Sat 17.30), Hammarby–Brommapojkarna (Sun 13/9 14.00), Elfsborg–Kalmar (Sun 14.00), Malmö–Örgryte (Sun 16.30), Djurgården–Gais (Mon 14/9 19.00), Sirius–Degerfors (Mon 19.00), AIK–Mjällby (Wed 16/9 19.00).
Source: Fantasy Allsvenskan – Round 21, PlaymakerAI. Ownership and expected fantasy points as of 9 September 2026. Transfer and injury information checked against the clubs' own channels the same day.