Panionios v Zakynthos — Prediction, Odds & Stats
Sunday 27 September 2026 · GRE-Super League 2 · 2627
AI prediction
ai-blend v1.0 dailyRaw model output; no calibration record for this fixture's bands yet.
Top 5 scorelines
- 1-1 12.7%
- 2-0 12.6%
- 1-0 11.8%
- 0-0 9.9%
- 2-1 9.5%
Goals
| Over 2.5 | 47.1% | Under 2.5 | 52.9% |
| BTTS yes | 46.4% | BTTS no | 53.6% |
| Panionios 1+ | 83.5% | Panionios 2+ | 53.7% |
| Zakynthos 1+ | 53.1% | Zakynthos 2+ | 17.6% |
Expected goals 1.80 – 0.76. Every market above is read off one scoreline distribution.
Footing
Positive factors
- Form is good: 3.00 points a match from the last five.
Risk factors
- Data quality is 20/100, so this price rests on less than the model would like.
- Panionios have only 1 matches on record.
- Zakynthos have only 1 matches on record.
Why the model thinks this
The price
Panionios have an estimated 60.1% win probability, against 26.7% for the draw and 13.1% for Zakynthos. Expected goals from the ensemble are 1.80 for Panionios and 0.76 for Zakynthos.
Why Panionios
Form is good: 3.00 points a match from the last five.
Why the draw is live
The draw sits at 26.7%.
Why Zakynthos can win
Nothing in the data singles out Zakynthos.
Key players
No player-level data for this fixture; the leagues with lineups are the big five.
Availability
No lineup on record for Panionios, so availability is unknown. No lineup on record for Zakynthos, so availability is unknown.
The matchup, as far as the data reaches
Panionios: 50% possession, 12.0 shots a match, 12.0 faced, 0.30 xG a match from set pieces. Zakynthos: 50% possession, 12.0 shots a match, 12.0 faced, 0.30 xG a match from set pieces.
Inputs last refreshed: matches 2026-09-08 10:52 · team_stats 2026-09-08 06:03 · availability 2026-09-08 11:54 · prediction 2026-09-08 11:44 ·
Panionios are the model's most likely outcome at 60.1%. The likeliest scoreline is 1-1 at 12.7%. Confidence 21/100, data quality 20/100. These are probabilities, not certainties: the less likely outcome happens as often as its number says.
Written from the numbers the model saw, not by a language model; nothing here is a statistic the snapshot does not hold. How predictions work.
Was this pick made in advance?
-oos is out of sample: it was fitted only on matches
played before this one, so its pick is a genuine forecast rather than hindsight.
Anything without that suffix was fitted on all available data.What the model said
| Model | Home / Draw / Away | Expected goals | Likeliest score | O2.5 | BTTS |
|---|---|---|---|---|---|
| ai-blend | 1.80–0.76 | 0% | 0% | 0% | |
| dc-xg | 1.91–0.65 | 2-0 14% | 47% | 43% |
A model whose name ends -oos was produced by the walk-forward backtest:
it was fitted only on matches played before this one, so the pick is genuinely out of sample.