WWCFxWorld Cup Fantasy Advanced Model
Matchday12345678
Backtest · Matchday 8

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
66
Mean abs error
2.67
pts/player
Rank correlation
0.37
predicted vs actual
Bias
-0.28
under-predict
Scouting-bonus calibration

predicted chance to clear threshold vs what actually happened

BucketPredictedObservedn
0-20%10%13%23
20-40%25%13%8
Recommended XI · actual result
XI actual points (capt ×2)
93.0
Captain
Kylian Mbappé (17 pts)
Captain rank in XI
#2 of 11
Best captain (hindsight)
Bukayo Saka
Avg player (field)
3.27 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+0.59
pred 1.59 · actual 1 · n=11
1-5% owned+0.45
pred 2.6 · actual 2.15 · n=20
5-15% owned-0.79
pred 3.15 · actual 3.94 · n=18
15%+ owned-1.18
pred 4.17 · actual 5.35 · n=17
Model tuning hints
  • MID: under-predicting by 1.04 — adjust MID rates.
  • FWD: under-predicting by 1.09 — adjust FWD rates.
Biggest prediction misses
MIDBukayo Saka
pred 4.4 · actual 21
MIDDeclan Rice
pred 2.8 · actual 13
FWDKylian Mbappé
pred 8.4 · actual 17
FWDFerran Torres
pred 2.1 · actual 9
DEFMaxence Lacroix
pred 2.6 · actual -3
MIDEberechi Eze
pred 1.5 · actual 7
MIDNico Williams
pred 1.6 · actual 7
MIDMichael Olise
pred 4.6 · actual 10
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.

Recommended squad
93
capt Kylian Mbappé (17) · best: Bukayo Saka
Maximum Points
91
capt Kylian Mbappé (17) · best: Kylian Mbappé
Balanced
82
capt Kylian Mbappé (17) · best: Kylian Mbappé
Wildcard
74
capt Kylian Mbappé (17) · best: Kylian Mbappé
Scouting Bonus
55
capt Bukayo Saka (21) · best: Bukayo Saka
Best possible XI
135
formation-legal, whole pool
Best legal squad XI
135
budget + country limits respected
Field-average XI
~39.2
random valid team benchmark
Under-projected · what we missed
  • MIDBukayo Saka4.421

    Low goal prior (4%) but hauled 21 — likely a finishing spike. Check role support before chasing.

  • MIDDeclan Rice2.813

    Scored 13 largely through tackle/chance volume the per-90 rates didn't credit (model expected 0.6 pts from those routes).

  • FWDKylian Mbappé8.417

    Returned 17 vs 8.4 projected — team/fixture strength for this matchup was likely underrated.

  • FWDFerran Torres2.19

    Model gave him only 53% to start, but he played and returned 9. The appearance model missed his role.

  • MIDEberechi Eze1.57

    Model gave him only 34% to start, but he played and returned 7. The appearance model missed his role.

  • MIDNico Williams1.67

    Model gave him only 34% to start, but he played and returned 7. The appearance model missed his role.

Over-projected · what we chased
  • DEFTheo Hernández3.2-2

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

  • GKMike Maignan3.2-2

    Projected 74 minutes but recorded 0 points — benched, unused or the round hadn't scored him. Start model called this wrong.

  • MIDDésiré Doué4.71

    Played but blanked (1 vs 4.7 projected) — mostly variance.

  • FWDLionel Messi5.42

    Played but blanked (2 vs 5.4 projected) — mostly variance.

  • MIDEnzo Fernández3.10

    Projected 74 minutes but recorded 0 points — benched, unused or the round hadn't scored him. Start model called this wrong.

  • FWDMikel Oyarzabal4.92

    Played but blanked (2 vs 4.9 projected) — mostly variance.

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

backtested 4 Aug, 08:15 UTC