WWCFxWorld Cup Fantasy Advanced Model
Matchday12345678
Backtest · Matchday 6

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
130
Mean abs error
1.96
pts/player
Rank correlation
0.29
predicted vs actual
Bias
+0.45
over-predict
Scouting-bonus calibration

predicted chance to clear threshold vs what actually happened

BucketPredictedObservedn
0-20%10%4%54
20-40%26%22%32
Recommended XI · actual result
XI actual points (capt ×2)
67.0
Captain
Kylian Mbappé (13 pts)
Captain rank in XI
#2 of 11
Best captain (hindsight)
Jude Bellingham
Avg player (field)
2.78 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+0.13
pred 2.29 · actual 2.16 · n=51
1-5% owned+0.94
pred 3.27 · actual 2.32 · n=37
5-15% owned+0.37
pred 3.79 · actual 3.42 · n=24
15%+ owned+0.41
pred 5.08 · actual 4.67 · n=18
Biggest prediction misses
MIDJude Bellingham
pred 6.8 · actual 14
FWDDan Ndoye
pred 2.9 · actual 10
MIDFabián Ruiz
pred 2.4 · actual 9
FWDAndreas Schjelderup
pred 3.2 · actual 9
MIDAlexis Mac Allister
pred 3.4 · actual 9
MIDCharles De Ketelaere
pred 4.7 · actual 10
FWDHarry Kane
pred 8.2 · actual 3
MIDMikel Merino
pred 3.9 · actual 9
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.

Wildcard
76
capt Kylian Mbappé (13) · best: Jude Bellingham
Balanced
74
capt Kylian Mbappé (13) · best: Jude Bellingham
Recommended squad
67
capt Kylian Mbappé (13) · best: Jude Bellingham
Maximum Points
65
capt Kylian Mbappé (13) · best: Jude Bellingham
Scouting Bonus
19
capt Bukayo Saka (4) · best: Désiré Doué
Best possible XI
117
formation-legal, whole pool
Best legal squad XI
117
budget + country limits respected
Field-average XI
~33.4
random valid team benchmark
Under-projected · what we missed
  • MIDJude Bellingham6.814

    Returned 14 vs 6.8 projected — team/fixture strength for this matchup was likely underrated.

  • FWDDan Ndoye2.910

    Low goal prior (12%) but hauled 10 — likely a finishing spike. Check role support before chasing.

  • MIDFabián Ruiz2.49

    Scored 9 largely through tackle/chance volume the per-90 rates didn't credit (model expected 0.3 pts from those routes).

  • FWDAndreas Schjelderup3.29

    Low goal prior (7%) but hauled 9 — likely a finishing spike. Check role support before chasing.

  • MIDAlexis Mac Allister3.49

    Scored 9 largely through tackle/chance volume the per-90 rates didn't credit (model expected 0.5 pts from those routes).

  • MIDCharles De Ketelaere4.710

    Low goal prior (19%) but hauled 10 — likely a finishing spike. Check role support before chasing.

Over-projected · what we chased
  • FWDHarry Kane8.23

    Played but blanked (3 vs 8.2 projected) — mostly variance.

  • MIDBradley Barcola5.91

    Played but blanked (1 vs 5.9 projected) — mostly variance.

  • FWDMikel Oyarzabal62

    Ceiling pick projected 6.0 but returned 2. High-variance blank; the miss cost is amplified if captained.

  • FWDBreel Embolo3.90

    Projected 73 minutes but recorded 0 points — benched, unused or the round hadn't scored him. Start model called this wrong.

  • DEFPedro Porro5.92

    Played but blanked (2 vs 5.9 projected) — mostly variance.

  • MIDÁlex Baena4.61

    Played but blanked (1 vs 4.6 projected) — mostly variance.

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

backtested 4 Aug, 08:15 UTC