Saturday 19 Sept, 15:00 · engine xGB Core v1 · doesn't know lineups or injuries · reassessed after this week's earlier results (rev 4)

Brighton & Hove Albion

59Index58Attack58Defense
DLWLL
49%ARS WIN
01
Likeliest ARS win · 10%

Arsenal

99Index76Attack99Defense
WWWWW
BRIGHTON 23%DRAW 28%ARSENAL 49%
Over 2.5 goals
48%
Both teams score
52%
Expected goals
1.0 · 1.6
Why
Head to head · 18 league meetings since 2017/18
BHA 55 drawn8 ARS
2025/26BHA01ARS2026-03-04
2025/26ARS21BHA2025-12-27
2024/25BHA11ARS2025-01-04
2024/25ARS11BHA2024-08-31
2023/24BHA03ARS2024-04-06
Brighton — projected scorers
Maxim De Cuyper12%22% ast78'
Pascal Groß8%15% ast90'
Jack Hinshelwood14%3% ast46'
Diego Gómez8%7% ast71'
Georginio Rutter3%12% ast25'
Charalampos Kostoulas7%6% ast46'
Arsenal — projected scorers
Kai Havertz27%13% ast63'
Bukayo Saka15%20% ast67'
Riccardo Calafiori7%12% ast69'
Viktor Gyokeres15%3% ast22'
Gabriel11%1% ast80'
Martin Odegaard2%9% ast52'

Projections assume recent minutes patterns — the model doesn’t know lineups.

What the grid implies
Goals
Over 1.5 goals75%
Over 2.5 goals48%
Over 3.5 goals26%
Both teams score52%
Double chance
BHA or draw51%
Draw or ARS77%
BHA or ARSno draw72%
Draw no bet
BHAstake returned on a draw32%
ARS68%
Clean sheets
BHA clean sheet21%
ARS clean sheet36%

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
BHAARS012345
0
1
2
3
4
5
Darker green = likelier. Rows are BHA goals, columns ARS. Scores of six or more each way carry the remaining probability.
Price movement — 4 pre-kickoff assessments, every one kept
5 Sept, 20:11BHA 23.1% (−1.4 since first)draw 27.9%ARS 49.0%
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

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