Serie A · engine xGB Current v1 · graded — the claim below was frozen before kickoff, never edited · reassessed after earlier results (rev 3)

Roma vs Atalanta

Saturday 5 Sept, 19:45
Full time
2–1
chances 3.54 – 0.77 xG
✓ called — we picked ROM; the ROM win we priced at 49%.
The claim — frozen pre-kickoff after 3 assessments
50 · 28 · 22ROM · draw · ATA
1–0
Likeliest ROM win · 11%
ROMA 50%DRAW 28%ATALANTA 22%

Expected goals: ROM 1.63 · ATA 1.05

Goals
Over 1.5 goals74%
Over 2.5 goals49%
Over 3.5 goals29%
Both teams score54%
Double chance
ROM or draw78%
Draw or ATA51%
ROM or ATAno draw71%
Draw no bet
ROMstake returned on a draw69%
ATA31%
Clean sheets
ROM clean sheet35%
ATA clean sheet20%

Model probabilities, not odds and not advice. Our own leaderboard shows no model here beats the closing market once the bookmaker’s margin is paid.

Every scoreline, priced — tap a cell
ROM↓ ATA→012345
0
1
2
3
4
5
2–1 · 9.6%
A ROM win. 3 goals — over 2.5. Expect it roughly 10 in 100 plays. This is what happened.
What this is built from
The engine weighs
  • Chance quality created and conceded, both sides, all season
  • Opponent strength and venue
  • Recent play, weighted toward now
It cannot see
  • Confirmed lineups
  • Injuries and suspensions
  • Cup rotation plans

Thin evidence: the engine has seen little recent top-flight data on Roma and Atalanta — this assessment leans on conservative defaults more than usual.

Engine internals are proprietary. The complete performance record — including every miss — is public on the record.

Head to head · 24 meetings since 2014/15
ROM 48 drawn12 ATA
2025/26Roma1–1Atalanta2026-04-18
2025/26Atalanta1–0Roma2026-01-03
2024/25Atalanta2–1Roma2025-05-12
2024/25Roma0–2Atalanta2024-12-02
2023/24Atalanta2–1Roma2024-05-12

Match-level assessment. No scorer projections here: we hold every appearance in this league, and a player model has been measured on it, but its probabilities came out over-confident for the players you would actually read. Recorded before kickoff, revised only until kickoff, graded after full time. Probabilities, never advice.