FPL GW4 projection accuracy scorecard
The accepted xpts_v1 release was frozen before the deadline, then compared with official points, minutes and starts after the Gameweek finished.
Snapshot2026-09-12-gw4
Likely players290
Likely-player MAE2.74 pts
xMins MAE23.6 min
No post-result reconstruction
Accuracy by cohort
| Cohort | Players | MAE | RMSE | Bias | Within 2 pts | xMins MAE | Start Brier |
|---|---|---|---|---|---|---|---|
| All captured players | 656 | 1.57 | 2.35 | +0.60 | 66.2% | 15.9 min | 0.105 |
| Projected ≥45 minutes | 290 | 2.74 | 3.31 | +0.79 | 33.8% | 23.6 min | 0.188 |
| Predicted top 25 | 25 | 3.11 | 3.44 | -0.24 | 16.0% | 13.0 min | 0.059 |
Highest pre-deadline xPts
Predicted top 10 outcomes
| Player | Team | Predicted | Actual | Error | xMins | Minutes |
|---|---|---|---|---|---|---|
| #1 Haaland | MCI | 6.88 | 9 | -2.12 | 87.6 | 90 |
| #2 B.Fernandes | MUN | 6.54 | 2 | +4.54 | 86.5 | 90 |
| #3 Palmer | CHE | 5.77 | 5 | +0.77 | 82.6 | 90 |
| #4 João Pedro | CHE | 5.65 | 12 | -6.35 | 77.4 | 90 |
| #5 Saka | ARS | 5.61 | 8 | -2.39 | 70.5 | 90 |
| #6 Gabriel | ARS | 5.48 | 9 | -3.52 | 89.6 | 90 |
| #7 Szoboszlai | LIV | 5.45 | 3 | +2.45 | 88.0 | 90 |
| #8 Calafiori | ARS | 5.44 | 6 | -0.56 | 75.9 | 90 |
| #9 Cherki | MCI | 5.35 | 1 | +4.35 | 82.8 | 45 |
| #10 De Cuyper | BHA | 5.31 | 11 | -5.69 | 75.5 | 90 |
How this scorecard is calculated
Point error is predicted next-Gameweek xPts minus settled official points. MAE averages absolute error; RMSE gives larger misses more weight; positive bias means overprediction. Minutes MAE compares xMins with played minutes. Start Brier score measures the probability forecast against the official start flag, where lower is better.