WWCFxWorld Cup Fantasy Advanced Model
Matchday12345678
Backtest · Matchday 1

Prediction Accuracy

After a round finishes, we score the predictions against actual fantasy points (FIFA feed) — error, ranking, scouting-bonus calibration, and how the recommended XI did. The hints feed back into tuning the model.

Model calibration · auto-tunedfrom rounds 1, 2, 3, 4, 5, 6, 7, 8

attacking-output multiplier the tuner learned per position (1.0 = no change)

GK×1.30
was under-predicting
DEF×1.30
was under-predicting
MID×1.14
was under-predicting
FWD×1.22
was under-predicting
Players scored
753
Mean abs error
2.25
pts/player
Rank correlation
0.51
predicted vs actual
Bias
+0.07
over-predict
Scouting-bonus calibration

predicted chance to clear threshold vs what actually happened

BucketPredictedObservedn
0-20%12%10%392
20-40%28%25%260
40-60%46%42%36
60-80%65%40%5
Recommended XI · actual result
XI actual points (capt ×2)
59.0
Captain
Ferran Torres (2 pts)
Captain rank in XI
#9 of 11
Best captain (hindsight)
Alexander Isak
Avg player (field)
2.9 pts
Predicted vs actual · by ownership bucket

where the model flatters players — chasing differentials is only right if the low-owned buckets aren't inflated

<1% owned+0.08
pred 2.61 · actual 2.53 · n=566
1-5% owned+0.37
pred 3.93 · actual 3.57 · n=127
5-15% owned-0.17
pred 3.71 · actual 3.87 · n=39
15%+ owned-1.41
pred 5.55 · actual 6.95 · n=21
Biggest prediction misses
MIDElijah Just
pred 3.3 · actual 16
MIDYasin Ayari
pred 4.8 · actual 17
DEFRamin Rezaeian
pred 3.2 · actual 13
FWDDeniz Undav
pred 4.5 · actual 14
GKVozinha
pred 1.5 · actual 11
FWDLionel Messi
pred 9.5 · actual 19
FWDFolarin Balogun
pred 5.7 · actual 15
DEFNathaniel Brown
pred 5 · actual 14
Post-round autopsy · what actually happened

Every squad's real score vs the hindsight-best XIs, plus every big miss classified to the model factor that caused it — so each round makes the next projection better.

Maximum Points
108
capt Kai Havertz (13) · best: Lionel Messi
Wildcard
90
capt Ferran Torres (2) · best: Lionel Messi
Balanced
66
capt Ferran Torres (2) · best: Alexander Isak
Recommended squad
59
capt Ferran Torres (2) · best: Alexander Isak
Scouting Bonus
59
capt Ferran Torres (2) · best: Alexander Isak
Best possible XI
180
formation-legal, whole pool
Best legal squad XI
180
budget + country limits respected
Field-average XI
~34.8
random valid team benchmark
Under-projected · what we missed
  • MIDElijah Just3.316

    Scored 16 largely through tackle/chance volume the per-90 rates didn't credit (model expected 0.5 pts from those routes).

  • MIDYasin Ayari4.817

    Low goal prior (16%) but hauled 17 — likely a finishing spike. Check role support before chasing.

  • DEFRamin Rezaeian3.213

    DEF returned 13 vs 3.2 projected — clean-sheet/save/conceded points are systematically under-modelled for DEFs.

  • FWDDeniz Undav4.514

    Model gave him only 30% to start, but he played and returned 14. The appearance model missed his role.

  • GKVozinha1.511

    GK returned 11 vs 1.5 projected — clean-sheet/save/conceded points are systematically under-modelled for GKs.

  • FWDLionel Messi9.519

    Returned 19 vs 9.5 projected — team/fixture strength for this matchup was likely underrated.

Over-projected · what we chased
  • FWDLautaro Martínez9.61

    Ceiling pick projected 9.6 but returned 1. High-variance blank; the miss cost is amplified if captained.

  • FWDFerran Torres10.32

    Ceiling pick projected 10.3 but returned 2. High-variance blank; the miss cost is amplified if captained.

  • FWDMikel Oyarzabal9.22

    Ceiling pick projected 9.2 but returned 2. High-variance blank; the miss cost is amplified if captained.

  • FWDCody Gakpo8.42

    Ceiling pick projected 8.4 but returned 2. High-variance blank; the miss cost is amplified if captained.

  • FWDDarwin Núñez6.51

    Ceiling pick projected 6.5 but returned 1. High-variance blank; the miss cost is amplified if captained.

  • FWDGonçalo Ramos6.21

    Ceiling pick projected 6.2 but returned 1. High-variance blank; the miss cost is amplified if captained.

Model factors behind this round's misses
  • 7×Ceiling pick blanked — captaincy/exposure sizing, not projection
  • 4×Defensive scoring calibration (clean sheets, saves, conceded)
  • 3×Team strength / fixture odds for this matchup
  • 2×Variance blank — played, no returns; only act if it repeats
  • 1×Per-90 tackle / chance-created rates (volume routes)
  • 1×One-off finishing spike — do NOT over-correct, check role support
  • 1×Start-probability model (appearances / rotation signal)
  • 1×Start-probability model (predicted a start that didn't happen)

backtested 4 Aug, 08:15 UTC