LiverpoolvsManchester City

Manchester City visit Anfield for Round 6, and the model makes them the more probable winner at 53.3%, against 24.9% for Liverpool and 21.7% for the draw. The table reference points here come from 2025-26, when City finished second on 78 points and Liverpool fifth on 60; this season's data is too thin to be meaningful. Both sides arrive 21 days after their last match, so the layoff is identical.

Liverpool finished 2025-26 fifth on 60 points, scoring 63 and conceding 53 across 38 games — the division's fourth-ranked attack but only its 13th-ranked defence. At home their scoring rate rises to 2.05 per game with 1.15 conceded over the last 20. Their most recent five read four wins and a draw, averaging 1.6 scored and 0.4 conceded.

City's 2025-26 brought second place on 78 points, with 77 scored and 35 conceded and rankings of first for attack and second for defence. The last five are all wins, at 2.8 scored and 0.6 conceded per game. Away from home over the last 20, though, that rate falls to 1.7 scored with 1.3 conceded, and the win rate is 55%.

Four meetings are on record and they split evenly: two wins each, with the goals level at 5-5 and an average of 2.5 per meeting. Each side has won once at home and once away in the sample. City were kept off the scoresheet twice across the four, Liverpool once. Two of the four produced two goals or fewer, but four matches is a small sample to lean on.

Liverpool's home split improves on their overall figures — 2.05 scored and 1.15 conceded per home game against 1.66 and 1.39 overall — though the win rate barely shifts, 45% at home versus 44.7% overall. City's away numbers move the other way: 1.7 scored and 1.3 conceded, against 2.03 and 0.92 overall. That defensive gap is the widest home-away split in this fixture's data.

Both teams have had 21 days since their 2026-09-20 match, so the rest intervals are identical. The schedule-strength notes differ — Liverpool's three recent fixtures averaged an opponent rank of 11.3, City's two averaged 5 — but City's sample is the smaller one. Every table figure quoted in this preview comes from 2025-26.

The Poisson model gives City 53.3%, the draw 21.7% and Liverpool 24.9%, and states high confidence. The projected scoreline is 1-1 on expected values of 1.3 and 1.99; 1-1 is also the single most likely exact score at 9.7%, just ahead of 1-2 at 9.6%. Roughly 3.29 goals are projected in total, with an 84% chance of two or more and a 62.8% chance both teams score at least once. The margin distribution tilts toward a multi-goal away win: away by two or more 31.2%, away by one 22.2%, draw 21.7%, home by one 14.5%, home by two or more 10.4%.

The data points to a City win: a higher 2025-26 finish, the division's top-ranked attack, five straight wins, and a model making them a 53.3% favourite. The counterweights are real — Anfield lifts Liverpool to 2.05 goals per home game, 1-1 is the most probable exact scoreline, and the draw or a home win still carries 46.6% combined. A low-scoring home result is what would prove this read wrong.