Oddsprey

Forgot your password?

Don't have an account yet? Create one

← All match write-ups
Primeira Liga

FC Porto vs Benfica

Kick-off 21:30 · 2026-09-20
Big match Medium confidence

Result

HT1-0 FT3-1 UNDER 1.5 HT won
Games this big draw sharp, heavily traded markets, so a real betting edge is rare here. We usually find better value in quieter fixtures. Still, here is how we see it landing.

Predicted score

1-1
13%
1-0
9%
2-1
8%
Combined, there's a 31% chance one of these three scorelines is the exact final result.

Our call

UNDER 1.5 HT 1.44 72% model chance +3% edge

Match result (1X2)

FC Porto 39% Draw 28% Benfica 33%

Recent form

TeamLast 6WDLGoals
FC Porto WWWWWW 600 15:2
Benfica DWWWWW 510 21:4

Oldest on the left, most recent on the right. League matches only.

Head to head

Last 6 meetings: FC Porto 1, Benfica 3, 2 drawn, averaging 3.33 goals a game.

Scores, most recent first: 2-2, 0-0, 1-4, 1-4, 5-0, 0-1.

Analyst write-up

A top-of-table Clássico between two sides in near-identical WWWWW form, so expect a tight, tactically cautious game rather than a shootout. Porto's home defence is elite, while Benfica carry the sharper attack, leaving the sides close to level with a slim home edge. Lineups are not yet confirmed (they publish about an hour before kickoff), and with no injury data the read may shift on team news. Honest coin-flip between a 1-1 and a narrow home win.

Main risk: Derbies between two in-form, evenly matched sides are high-variance; either the game opens up into a 3-goal-plus affair as the H2H history suggests, or one side's elite defence produces a 1-0 grind. A single early goal could swing the shape decisively.

Unknowns: Injuries and suspensions are unavailable, so squad availability is unknown; lineups shown are last-match XIs, not confirmed teams. A key absentee for either side, especially in attack or central defence, would shift these λ materially.

Predictions are model estimates, not certainties. 18+. Bet responsibly. Odds and prices move, so always confirm before betting.