PAOK v Iraklis — Prediction, Odds & Stats
Saturday 17 October 2026 · GRE-Super League · 2627
AI prediction
ai-blend v1.0 dailyRaw model output; no calibration record for this fixture's bands yet.
Top 5 scorelines
- 2-0 13.6%
- 3-0 11.5%
- 1-0 9.6%
- 2-1 8.6%
- 1-1 8.0%
Goals
| Over 2.5 | 61.0% | Under 2.5 | 39.0% |
| BTTS yes | 44.2% | BTTS no | 55.8% |
| PAOK 1+ | 92.0% | PAOK 2+ | 71.8% |
| Iraklis 1+ | 46.7% | Iraklis 2+ | 13.2% |
Expected goals 2.52 – 0.63. Every market above is read off one scoreline distribution.
Footing
Positive factors
- PAOK carry the stronger rating, 1724 Elo against 1490.
- Form is good: 1.80 points a match from the last five.
Risk factors
- Data quality is 30/100, so this price rests on less than the model would like.
- Iraklis have only 2 matches on record.
Why the model thinks this
The price
PAOK have an estimated 77.3% win probability, against 16.7% for the draw and 6.0% for Iraklis. Expected goals from the ensemble are 2.52 for PAOK and 0.63 for Iraklis.
Why PAOK
PAOK carry the stronger rating, 1724 Elo against 1490. Form is good: 1.80 points a match from the last five.
Why the draw is live
The draw sits at 16.7%. Draws run at 25% in this competition.
Why Iraklis can win
Nothing in the data singles out Iraklis.
Key players
No player-level data for this fixture; the leagues with lineups are the big five.
Availability
No lineup on record for PAOK, so availability is unknown. No lineup on record for Iraklis, so availability is unknown.
The matchup, as far as the data reaches
PAOK: 49% possession, 12.2 shots a match, 17.6 faced, 0.30 xG a match from set pieces. Iraklis: 37% possession, 10.0 shots a match, 17.0 faced, 0.30 xG a match from set pieces.
Inputs last refreshed: matches 2026-09-05 22:20 · team_stats 2026-09-05 09:41 · availability 2026-09-05 21:03 · prediction 2026-09-05 18:39 ·
PAOK are the model's most likely outcome at 77.3%. The likeliest scoreline is 2-0 at 13.6%. Confidence 49/100, data quality 30/100. 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 |
|---|---|---|---|---|---|
| dc-xg | 2.58–0.71 | 2-0 12% | 64% | 48% |
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.