Saturday, 21/11/2026 at 18:00
2d 15h to kick-offParma – Roma
- Parma
- 12%
- Mkt 9% Market 9%
- Price 8.46 Model price 8.46
- Draw
- 22%
- Mkt 16% Market 16%
- Price 4.59 Model price 4.59
- Roma
- 66%
- Mkt 76% Market 76%
- Price 1.51 Model price 1.51
3 signals · See them below →
Each meets the rule the Betting section applies, at this moment: a mechanical filter, not a recommendation. How this is decided →
- Parma 20/20
- Roma 20/20
- Model updated 19 November, 00:30
- Top odds updated 18 November, 18:00
Where the model disagrees with the market
One answer per question, by model-implied edge. Margin removed. Bet365 prices.
Forecast skill is how much the model beat simply knowing how often each outcome happens, when 7 competitions were replayed (measured 17/08/2026). A market without forecast skill above zero never yields a signal, and forecast skill is not the chance a bet wins. How that is measured →
Recent form
xG created xG conceded (xGA) W / D / L: the match result
Goals & scorelines
- Total goals
- 3.11
- Over 2.5 goals
- 62%
- Both teams to score
- 59%
- Roma 2+ goals
- 67%
- Roma clean sheet
- 37%
Most likely scorelines
No exact score is especially likely. The 3 most probable scorelines account for 33% of the model's distribution.
- 1:2
- 11.5%
- 1:1
- 10.9%
- 0:2
- 10.6%
Why the model favours Roma
Attack, defence and recent form all point to Roma.
- Attack
- Favours Roma
- 2nd vs 18th
- Roma attack v Parma defence
- Defence
- Favours Roma
- 4th vs 20th
- Roma defence v Parma attack
- Recent xG form
- Favours Roma
- +0.79 vs -1.28
- xGD per match, last 6
Team comparison
Season to date, per match, with each club's placing in the competition. The last row is the last 6 matches.
| Reading | Parma | Roma |
|---|---|---|
| xG / match | 0.84 · 20th | (better) 2.41 · 2nd |
| xGA / match | 1.95 · 16th | (better) 1.32 · 8th |
| xGD / match | -1.11 · 20th | (better) +1.09 · 2nd |
| xPts / match | 0.66 · 20th | (better) 2.00 · 2nd |
| Last 6 xGD / match | -1.28 | (better) +0.79 |
Head to head
Parma 0 Draws 0 Roma 2
Got a different price?
Type your bookmaker's odds for any market and see the model-implied edge they carry.
How the model-implied edge is calculated
The model-implied edge is model probability × your odds − 1: what a stake of one would return on average if the model is right — not observed return, and not necessarily in this match. The model knows nothing of absences, motivation or the schedule.
Nothing you type is remembered.
What the model doesn't know
It reads expected goals and nothing else — not the lineups, injuries or suspensions, what either side is playing for, the weather or a congested schedule.
Model details- Fit window
- the last 20 matches, recency weighted (a match 107 days old counts half)
- Parma sample
- 20 matches
- Roma sample
- 20 matches
- Home xG sample · Parma
- 6 matches
- Away xG sample · Roma
- 5 matches
- Computed
- Thursday, 19/11/2026 at 00:30
- Odds updated
- 8 hours ago
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.