Saturday, 21/11/2026 at 18:30
Full timeNewcastle United 2:2 Arsenal
Arsenal were the model's likeliest winner before kick-off at 52%. It finished 2–2. The biggest difference was Newcastle United's attack: 1.71 xG against 0.92 projected (+0.79).
Goals
- 33' Eberechi Eze Arsenal · 0.47 xG chance 0–1
- 49' Harvey Barnes Newcastle United · 0.26 xG chance 1–1
- 62' Malick Thiaw Newcastle United · 0.27 xG chance 2–1
- 78' Noni Madueke Arsenal · 0.08 xG chance 2–2
The model against the result
- Newcastle United
- 19%
- Draw
- 29% What happened
- Arsenal
- 52%
Data coverage Both teams have the full 20-match model window available.
Projected vs produced
Newcastle United produced 0.79 xG more than projected
Projected before kick-off Produced on the night
The shots behind it
- Shots
- 12
- 13
- On target
- 3
- 4
- Inside the box
- 5
- 10
- xG per shot
- 0.14
- 0.11
- Open play xG
- 0.68
- 1.40
- Set-piece xG
- 1.05
- 0.06
- Penalties
- —
- —
Newcastle United's shots add up to 1.73 xG and Arsenal's to 1.46; the published figures are 1.71 and 1.49. Leaving out penalties and own goals, Newcastle United's shots come to 1.73 against a published non-penalty xG of 1.71: more than rounding can explain, so the two figures were not read at the same point of the provider's revisions. Leaving out penalties and own goals, Arsenal's shots come to 1.46 against a published non-penalty xG of 1.49: more than rounding can explain, so the two figures were not read at the same point of the provider's revisions. The provider computes a match's total xG and its non-penalty xG separately, so the difference between the two is not what a penalty was worth.
Shot map
Both teams shown attacking the same goal for comparison.
All shots
25 shots · 3.19 xG
Hover or tap a shot, or step through them with the arrows.
- Newcastle United
- Arsenal
- Filled = goal
- Dot area = xG (the smallest chances drawn at a minimum size)
Figures read 22 November, 08:32 · provisional, next reading 23 November 2026
Recent form
1.28 xG created 1.54 xGA
1.55 xG created 0.81 xGA
- L 1.63 created, 1.31 conceded against Brighton & Hove Albion
- W 2.33 created, 0.62 conceded against Leeds United
- W 1.14 created, 0.90 conceded against Nottingham Forest
- W 1.96 created, 0.63 conceded against Everton
- W 0.96 created, 1.05 conceded against Liverpool
- W 1.26 created, 0.36 conceded against Hull City
xG created xG conceded (xGA) W / D / L: the match result
Head to head
Pre-match model & market record
6 signals flagged1 won5 lost
- Won
- Lost
- Lost
- Lost
- Lost
- Lost
View pre-match record Hide pre-match record
| Selection | Price | Model-implied edge | Result |
|---|---|---|---|
| Draw | 4.10 | +19.5% | Won |
| Arsenal not to score | 6.50 | +16.2% | Lost |
| Arsenal at most 1 goal | 2.20 | +12.6% | Lost |
| Under 2.5 goals | 2.10 | +11.4% | Lost |
| Newcastle United to win | 5.75 | +7.6% | Lost |
| Under 1.5 goals | 4.33 | +5.4% | Lost |
Pre-match model vs market
| Outcome | Model | Market | Disagreement | Model price |
|---|---|---|---|---|
| Newcastle United | 18.7% | 16.5% | +2.2 pp | 5.34 |
| Draw | 29.2% | 23.1% | +6.0 pp | 3.43 |
| Arsenal | 52.1% | 60.4% | −8.2 pp | 1.92 |
Other markets
| Outcome | Model | Market | Disagreement | Model price |
|---|---|---|---|---|
| Newcastle United to score | 63.1% | 58.9% | +4.2 pp | 1.59 |
| Newcastle United not to score | 36.9% | 41.1% | −4.2 pp | 2.71 |
| Arsenal to score | 82.1% | 85.4% | −3.3 pp | 1.22 |
| Arsenal not to score | 17.9% | 14.6% | +3.3 pp | 5.60 |
| Both teams to score | 53.5% | 52.0% | +1.5 pp | 1.87 |
| Both teams not to score | 46.5% | 48.0% | −1.5 pp | 2.15 |
| Newcastle United 2+ goals | 23.1% | 25.0% | −1.9 pp | 4.34 |
| Newcastle United at most 1 goal | 76.9% | 75.0% | +1.9 pp | 1.30 |
| Arsenal 2+ goals | 48.8% | 57.7% | −8.9 pp | 2.05 |
| Arsenal at most 1 goal | 51.2% | 42.3% | +8.9 pp | 1.95 |
| Over 1.5 goals | 75.7% | 78.0% | −2.3 pp | 1.32 |
| Under 1.5 goals | 24.3% | 22.0% | +2.3 pp | 4.11 |
| Over 2.5 goals | 47.0% | 54.9% | −7.9 pp | 2.13 |
| Under 2.5 goals | 53.0% | 45.1% | +7.9 pp | 1.89 |
Prices and probabilities are the ones stored before kick-off; nothing here is recalculated. The model-implied edge is the model's probability times the price, minus one: what a stake of one would return on average if the model is right, not observed return, so a flagged outcome that lost is the ordinary case rather than a contradiction.
Model details
Show Hide
- Fit window
- the last 20 matches, recency weighted (a match 107 days old counts half)
- Newcastle United sample
- 20 matches
- Arsenal sample
- 20 matches
- Computed
- 2 hours before kick-off
- Prices taken
- 3 hours before kick-off
Limitations
The model reads expected goals and nothing else. It does not know the lineups, who is injured or suspended, what either side has to play for, the weather, or how congested the schedule around this match is.