Smart stats. Honest predictions.

Recent form

LIV DDWDW
MCI WWWWW
Last 5 league results · 60% vs 100% of points taken

Head to head

Liverpool vs Manchester City
6
Liverpool wins
6
Draws
8
Manchester City wins
From 20 meetings on record

Expected goals (xG)

1.42 vs 1.65
LiverpoolManchester City
The model's expected goals, adjusted for attack/defence/h2h/injuries. From kickoff, Match day below charts these against the real goals.

Fixture details

VenueAnfield, Liverpool
Referee TBC
StatusNot Started
Referees are typically only confirmed a few days before kickoff.
Prediction
Predicted score
Our Poisson model rates each team's attack and defence mostly on expected goals, the quality of the chances it makes and allows (blended with last season's ratings early on), then weighs head-to-head history and injuries to calculate the most likely outcome.
Liverpool
1 2
Manchester City
Model's expected scoreline · 9.0% probability of this exact score
PROVISIONAL This prediction locks on 10 Oct, 16:30 BST, 24 hours before kickoff, using the latest data.
Outcome certainty
43.0%
Low
ScorelineProbability
Liverpool 1-1 Manchester City 10.9%
Liverpool 1-2 Manchester City 9.0%
Liverpool 2-1 Manchester City 7.7%
Liverpool 0-1 Manchester City 7.7%
Liverpool 1-0 Manchester City 6.6%

In brief

Sixth-placed Liverpool host leaders Manchester City. Liverpool arrive in good form and Manchester City in excellent form, and Manchester City have won their last five. Liverpool are still looking for a first home win and Manchester City have won every away game this season. Manchester City are one of the league's highest-scoring sides. Past meetings between the two have been fairly even. Liverpool have the longer injury list. The model sees little between the sides. The bookmakers broadly agree with the model.

Written automatically from our statistics, so it updates when they do. Not betting advice.

Key factors in prediction

Statistics baseline: 2026 season
Liverpool vs Manchester City
Attack strength
1.07 / 1.31
Defence strength
1.17 / 1.12
Head-to-head
0.95 / 1.05
Injury factor
0.95 / 1.04
No units: every number compares a team with an average Premier League side, which scores 1.00. Attack is a team's scoring rate at home or away (built from goals and expected goals) against the league average, so 1.16 means 16% more than average. Defence is the same for goals conceded, turned round so higher is better. Head-to-head and injuries nudge each side's expected goals: 1.15 adds 15%, 0.85 takes 15% off, 1.00 changes nothing.

Win probability

Liverpool win33%
Draw24%
Manchester City win43%

Odds comparison — 1x2

36
Bookmakers
36
Offering 1x2
6.5%
Overround
Liverpool —
Model33.2%
Market32.3%
Difference +0.9 points
Draw —
Model23.8%
Market25.6%
Difference -1.8 points
Manchester City —
Model43.0%
Market42.1%
Difference +0.9 points
Where we differ from the bookmakers, they have usually been right: in our backtests, backing the outcomes our model rated higher lost money. Treat a difference as something to look into, not a bet.

Odds comparison — total goals

2.5
Goals line
12
Bookmakers
6.9%
Overround
Over 2.5 goalsAverage odds 1.57 —
Model59.2%
Market59.6%
Difference -0.4 points
Under 2.5 goalsAverage odds 2.32 —
Model40.8%
Market40.4%
Difference +0.4 points

Injuries

PlayerTeamReason
C. BradleyLiverpoolKnee Injury
H. EkitikeLiverpoolAchilles Tendon Injury
G. LeoniLiverpoolKnee Injury
F. ChiesaLiverpoolBack Injury
P. FodenManchester CityRed Card

Weather at kickoff

UP TO 3 DAYS BEFORE KICKOFF

The forecast at Anfield fills in once the match is within 3 days.

Match day

Confirmed lineups

ABOUT AN HOUR BEFORE KICKOFF

Both starting XIs and formations.

Expected v real goals

FROM KICKOFF, 16:30 BST

The model's expected goals against the real goals, minute by minute.

Live match stats

FROM KICKOFF, 16:30 BST

Shots, possession, corners and cards as the game goes on.

Result

AT FULL TIME

The final score against our locked prediction.

See how we've actually done
Every prediction is locked before kickoff and checked against the final score, good weeks and bad.
See the results