Track Record
How the model actually performs on Greek Super League matches • Predictions through: 2026-11-07 09:22:13
Matches evaluated
69
finished matches with a stored prediction
Pick accuracy
58.0%
most likely outcome was correct
Log loss
0.916
uniform guess = 1.099 — lower is better
Brier score
0.537
0 = perfect, lower is better
Probability calibration
When the model says 60%, the event should happen about 60% of the time. The dashed line is perfection; the solid line is the model across all three outcomes.
Overall reliability curve, all three outcomes pooled. Sparse buckets (few matches) naturally wobble more — the current sample is small this early in the season.
Per-outcome diagnostics
| Outcome | Precision | Recall | F1 | Matches |
|---|---|---|---|---|
| Home win | 64% | 70% | 67% | 33 |
| Draw | 50% | 7% | 12% | 15 |
| Away win | 52% | 76% | 62% | 21 |
Draws are the hardest outcome in football — precision/recall stay low for every honest model.
Latest results
| Date | Match | Pick | Result | |
|---|---|---|---|---|
| 2026-11-08 | Olympiacos FC – PAOK FC | Home win (64.1%) | 1–1 | ✗ |
| 2026-11-08 | OFI Crete FC – Atromitos FC | Home win (49.4%) | 0–0 | ✗ |
| 2026-11-08 | Iraklis FC – Panathinaikos FC | Away win (54.0%) | 1–2 | ✓ |
| 2026-11-08 | Asteras Tripolis FC – Kalamata FC | Away win (34.6%) | 1–1 | ✗ |
| 2026-11-08 | AEK Athens FC – Levadiakos FC | Home win (68.5%) | 2–1 | ✓ |
| 2026-11-07 | Volos NFC – Kifisia FC | Away win (40.2%) | 1–3 | ✓ |
| 2026-11-07 | Aris Thessaloniki FC – Panetolikos FC | Home win (65.0%) | 3–0 | ✓ |
| 2026-11-02 | Kalamata FC – Iraklis FC | Draw (40.8%) | 1–1 | ✓ |
| 2026-11-02 | Atromitos FC – Volos NFC | Home win (48.7%) | 3–0 | ✓ |
| 2026-11-01 | PAOK FC – Asteras Tripolis FC | Home win (63.6%) | 2–1 | ✓ |
| 2026-11-01 | Panetolikos FC – OFI Crete FC | Away win (46.9%) | 0–1 | ✓ |
| 2026-11-01 | Panathinaikos FC – AEK Athens FC | Home win (40.0%) | 2–0 | ✓ |
What we don't publish: betting simulations, ROI figures or “value picks”. The numbers above are about the quality of the probabilities, not about wagering — see the methodology for how they are produced.