Premier League · Preview

Premier League predictions: matchweek 8 (23–25 October 2026)

Every Premier League match in matchweek 8 with the model's probabilities, the likeliest score, form, table positions and head-to-head behind each call.

Published 16 Oct 2026, updated 18 Oct 2026 · Written from the site’s own data

Premier League, matchweek 8: ten matches, played across 23–25 October 2026. Our model has published a forecast for every one.

The strongest favourite is Arsenal, 60% to beat Everton at home; the most evenly balanced game is Crystal Palace v Newcastle (Crystal Palace 39%, draw 25%, Newcastle 36%); the fixture expected to produce most goals is Chelsea v Tottenham, at 3.39; the tightest looks Sunderland v Leeds, at 2.44.

Going into the round, Manchester City lead on 21 points, six points clear of Arsenal, with Brighton third on 13. The bottom three are Sunderland (4), Bournemouth (3), Tottenham (3).

The numbers this season

Premier League has averaged 2.91 goals a game over 65 matches this season, with over 2.5 goals in 55% of them and both sides scoring in 54%. Home sides have won 38% of matches, 29% have been drawn and away sides have won 32%. Those rates are the base the model's probabilities are judged against: a call only stands out where it beats what the competition normally produces.

Match by match

Ipswich v Nottingham ForestFri 23 Oct, 20:00

Our model makes Ipswich 34%, the draw 27% and Nottingham Forest 39%. The likeliest score is 1-1 (13%), 2.73 goals are expected, over 2.5 goals 51%, both teams to score 56%. The headline call is both teams to score at 56%. Ipswich are 10th on 9 points from 7; Nottingham Forest are 17th on 5 points from 6. Ipswich have won two of their last five; Nottingham Forest have won one of their last five.

Aston Villa v Manchester CitySat 24 Oct, 12:30

Our model makes Aston Villa 26%, the draw 23% and Manchester City 50%. The likeliest score is 1-1 (10%), 3.27 goals are expected, over 2.5 goals 64%, both teams to score 64%. The headline call is Manchester City to win at 50%. Aston Villa are 12th on 8 points from 7; Manchester City are 1st on 21 points from 7. Aston Villa have won two of their last five; Manchester City have won all of their last five. Aston Villa have won two of the sides' last two meetings.

Arsenal v EvertonSat 24 Oct, 15:00

Our model makes Arsenal 60%, the draw 21% and Everton 19%. The likeliest score is 2-1 (10%), 3.28 goals are expected, over 2.5 goals 64%, both teams to score 60%. The headline call is Arsenal to win at 60%. Arsenal are 2nd on 15 points from 6; Everton are 7th on 10 points from 7. Arsenal have won four of their last five; Everton have won one of their last five. Arsenal have won two of the sides' last two meetings.

Coventry v FulhamSat 24 Oct, 15:00

Our model makes Coventry 40%, the draw 27% and Fulham 34%. The likeliest score is 1-1 (13%), 2.71 goals are expected, over 2.5 goals 51%, both teams to score 56%. The headline call is both teams to score at 56%. Coventry are 15th on 6 points from 6; Fulham are 16th on 5 points from 7. Coventry have won two of their last five; Fulham have won one of their last five.

Chelsea v TottenhamSat 24 Oct, 17:30

Our model makes Chelsea 54%, the draw 22% and Tottenham 24%. The likeliest score is 1-1 (10%), 3.39 goals are expected, over 2.5 goals 66%, both teams to score 65%. The headline call is Chelsea to win at 54%. Chelsea are 4th on 13 points from 7; Tottenham are 20th on 3 points from 6. Chelsea have won two of their last five; Tottenham are without a win in five (three draws, two defeats). Chelsea have won two of the sides' last two meetings.

Crystal Palace v NewcastleSun 25 Oct, 14:00

Our model makes Crystal Palace 39%, the draw 25% and Newcastle 36%. The likeliest score is 1-1 (11%), 3.08 goals are expected, over 2.5 goals 60%, both teams to score 62%. The headline call is both teams to score at 62%. Crystal Palace are 13th on 7 points from 6; Newcastle are 11th on 8 points from 7. Crystal Palace have won two of their last five; Newcastle have won one of their last five. The sides' last two meetings split 1–0–1 between Crystal Palace wins, draws and Newcastle wins.

Hull City v BrentfordSun 25 Oct, 14:00

Our model makes Hull City 25%, the draw 28% and Brentford 48%. The likeliest score is 1-1 (13%), 2.46 goals are expected, over 2.5 goals 44%, both teams to score 49%. The headline call is Brentford to win at 48%. Hull City are 9th on 9 points from 7; Brentford are 6th on 10 points from 7. Hull City are without a win in five (three draws, two defeats); Brentford have won one of their last five.

Liverpool v BrightonSun 25 Oct, 14:00

Our model makes Liverpool 29%, the draw 23% and Brighton 48%. The likeliest score is 1-1 (10%), 3.29 goals are expected, over 2.5 goals 64%, both teams to score 65%. The headline call is Brighton to win at 48%. Liverpool are 5th on 12 points from 7; Brighton are 3rd on 13 points from 6. Liverpool have won three of their last five; Brighton have won three of their last five. Liverpool have won two of the sides' last three meetings.

Manchester United v BournemouthSun 25 Oct, 14:00

Our model makes Manchester United 43%, the draw 25% and Bournemouth 32%. The likeliest score is 1-1 (11%), 3.09 goals are expected, over 2.5 goals 60%, both teams to score 62%. The headline call is both teams to score at 62%. Manchester United are 14th on 6 points from 6; Bournemouth are 19th on 3 points from 6. Manchester United have won one of their last five; Bournemouth are without a win in five (three draws, two defeats). Two of the sides' last two meetings were drawn.

Sunderland v LeedsSun 25 Oct, 16:30

Our model makes Sunderland 28%, the draw 28% and Leeds 44%. The likeliest score is 1-1 (13%), 2.44 goals are expected, over 2.5 goals 44%, both teams to score 50%. The headline call is under 2.5 goals at 56%. Sunderland are 18th on 4 points from 6; Leeds are 8th on 9 points from 6. Sunderland have won one of their last five; Leeds have won one of their last five. One of the sides' last two meetings was drawn.

How to read these

Every probability above comes from a Dixon–Coles adjusted Poisson model fitted on recent results and expected goals, and every market on a match page is derived from the same scoreline distribution, so the likely score, the totals and the match result cannot contradict each other. A 70% call is expected to be wrong three times in ten; nothing here is a certainty.

The model has published seven picks on this round so far, where the best available price stands clear of its own probability and the market's consensus together. Each day's free picks appear on the homepage by 09:00 UK time; the rest are for members, and every pick is listed with its outcome once the day is over.