Coventry v Sunderland — Prediction, Odds & Stats
Saturday 31 October 2026 · ENG-Premier League · 2627
Premier League
xG Prediction generated
AI prediction
ai-blend v1.0 dailyRaw model output; no calibration record for this fixture's bands yet.
Top 5 scorelines
- 1-1 12.5%
- 0-1 10.6%
- 1-0 8.9%
- 1-2 8.8%
- 0-0 7.6%
Goals
| Over 2.5 | 47.7% | Under 2.5 | 52.3% |
| BTTS yes | 52.2% | BTTS no | 47.8% |
| Coventry 1+ | 69.3% | Coventry 2+ | 33.1% |
| Sunderland 1+ | 75.3% | Sunderland 2+ | 40.8% |
Expected goals 1.18 – 1.40. Every market above is read off one scoreline distribution.
Footing
Positive factors
- Sunderland carry the stronger rating, 1536 Elo against 1493.
- Sunderland have created 1.59 xG a match over their last five while Coventry have allowed 1.68.
- Away from home Sunderland average 1.56 xG a match.
Risk factors
- The component models disagree more than usual (agreement 0.88).
- Coventry have only 7 matches on record.
Why the model thinks this
The price
Sunderland have an estimated 42.0% win probability, against 31.7% for Coventry and 26.3% for the draw. Expected goals from the ensemble are 1.18 for Coventry and 1.40 for Sunderland.
Why Coventry
Coventry have created 1.11 xG a match over their last five while Sunderland have allowed 1.77. At home Coventry average 1.30 xG a match. Their last eleven was closer to full strength: Sunderland were missing 0.80 expected goal involvements a match against 0.36.
Why the draw is live
The draw sits at 26.3%. The sides are closely matched on expected goals, 1.18 to 1.40. Draws run at 25% in this competition.
Why Sunderland can win
Sunderland carry the stronger rating, 1536 Elo against 1493. Sunderland have created 1.59 xG a match over their last five while Coventry have allowed 1.68. Away from home Sunderland average 1.56 xG a match.
Key players
Coventry's leading contributors by expected goal involvements a match: Jack Rudoni (0.31), Taiwo Awoniyi (0.26), Sidiki Chérif (0.25). Sunderland's leading contributors by expected goal involvements a match: Enzo Le Fée (0.39), Brian Brobbey (0.32), Tymur Tutierov (0.25).
Availability
1 of Coventry's top three contributors did not start their last match. 1 of Sunderland's top three contributors did not start their last match. Sunderland had a player sent off last time out, worth about 0.07 expected goal involvements a match. Reported unavailable for Sunderland: Kevin Danso (suspended).
The matchup, as far as the data reaches
Coventry: 32% possession, 11.4 shots a match, 19.2 faced, 0.12 xG a match from set pieces. Sunderland: 47% possession, 11.6 shots a match, 16.2 faced, 0.28 xG a match from set pieces.
Inputs last refreshed: matches 2026-10-22 14:54 · team_stats 2026-10-20 06:04 · availability 2026-10-22 14:58 · prediction 2026-09-18 10:37 ·
Sunderland are the model's most likely outcome at 42.0%. The likeliest scoreline is 1-1 at 12.5%. Confidence 78/100, data quality 80/100. 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 chances say
Coventry are conceding more than the chances against them merit — 2.2 goals above expected, which tends to correct.
Sunderland are finishing sharply below their chances — 4.9 goals fewer than expected across 10 matches, despite creating the openings.
Finishing swings far more than chance creation, so a gap this size usually narrows. Over ten matches it is suggestive, not settled — a genuinely elite finisher beats expected goals for years.
What the model said
| Model | Home / Draw / Away | Expected goals | Likeliest score | O2.5 | BTTS |
|---|---|---|---|---|---|
| ai-blend | 1.18–1.40 | 0% | 0% | 0% | |
| dc-goals | 0.82–1.11 | 0-0 16% | 31% | 39% | |
| dc-xg | 1.42–1.11 | 1-1 13% | 46% | 51% |
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.