Premier League · 2026-27 · after 100 matches
Premier League power rankings 2026-27
Arsenal top the Premier League power rankings: if every club played every other club home and away now, the model expects them to outscore opponents by 0.93 goals per match. Manchester City are second (+0.92), and Ipswich are bottom (−0.91).
Premier League team ratings, ranked
Rating is expected goal difference per match against the rest of the league, home and away. Goals and points per match are on the same basis. The rank range covers the middle 80% of 2,000 plausible sets of ratings.
| # | Club | Rating | Goals for | Goals against | Pts / match | Rank range | Table | Last 5 |
|---|---|---|---|---|---|---|---|---|
| 1 | Arsenal | +0.93 | 1.78 | 0.85 | 1.99 | 1–2 | 1st · 27 pts | W W W W W |
| 2 | Manchester City | +0.92 | 1.98 | 1.06 | 1.95 | 1–2 | 3rd · 21 pts | W W L L L |
| 3 | Liverpool | +0.55 | 1.78 | 1.23 | 1.73 | 3–4 | 4th · 18 pts | L W W L W |
| 4 | Brighton | +0.36 | 1.62 | 1.26 | 1.61 | 3–7 | 2nd · 22 pts | W W L W W |
| 5 | Chelsea | +0.20 | 1.59 | 1.39 | 1.50 | 4–10 | 5th · 16 pts | W W L W L |
| 6 | Brentford | +0.11 | 1.50 | 1.39 | 1.44 | 5–12 | 8th · 14 pts | D L D W L |
| 7 | Aston Villa | +0.09 | 1.50 | 1.41 | 1.43 | 5–13 | 10th · 13 pts | D W W D D |
| 8 | Manchester United | +0.05 | 1.51 | 1.45 | 1.41 | 5–14 | 18th · 8 pts | D D L L D |
| 9 | Newcastle | +0.05 | 1.51 | 1.46 | 1.41 | 5–14 | 12th · 12 pts | L L W L D |
| 10 | Bournemouth | +0.05 | 1.45 | 1.41 | 1.41 | 5–14 | 7th · 15 pts | L W W W W |
| 11 | Nottingham Forest | −0.12 | 1.26 | 1.37 | 1.29 | 8–16 | 16th · 9 pts | L L D L W |
| 12 | Everton | −0.12 | 1.16 | 1.28 | 1.28 | 8–16 | 6th · 16 pts | D L L W W |
| 13 | Leeds United | −0.13 | 1.23 | 1.36 | 1.28 | 8–17 | 9th · 14 pts | L D D L W |
| 14 | Tottenham | −0.14 | 1.37 | 1.50 | 1.29 | 8–17 | 11th · 12 pts | D W W W L |
| 15 | Fulham | −0.16 | 1.22 | 1.38 | 1.26 | 9–17 | 15th · 10 pts | L W W D D |
| 16 | Hull City | −0.27 | 1.09 | 1.35 | 1.18 | 8–19 | 13th · 11 pts | D L D D L |
| 17 | Sunderland | −0.34 | 1.08 | 1.42 | 1.13 | 12–18 | 19th · 8 pts | L L D L W |
| 18 | Crystal Palace | −0.39 | 1.18 | 1.57 | 1.12 | 14–19 | 20th · 7 pts | W L L L L |
| 19 | Coventry City | −0.75 | 1.09 | 1.84 | 0.92 | 16–20 | 17th · 9 pts | W L L W L |
| 20 | Ipswich | −0.91 | 1.16 | 2.07 | 0.85 | 19–20 | 14th · 11 pts | W L D D L |
Ratings come from completed league results only; injuries, suspensions and transfers are not included here, though they can move the individual match predictions. Last 5 is the club's five most recent league results this season, oldest first.
Rating versus the league table
Higher in the table than their rating
- Everton: 6th in the table, 12th on rating (6 places).
- Ipswich: 14th in the table, 20th on rating (6 places).
- Leeds United: 9th in the table, 13th on rating (4 places).
Rated better than their table position
- Manchester United: 18th in the table, 8th on rating (10 places).
- Nottingham Forest: 16th in the table, 11th on rating (5 places).
A gap does not predict that a club will fall or rise by that many places; the predicted final table combines ratings with points already won and the remaining fixtures.
How the power rankings are calculated
These rankings use the published Premier League season model (dc-v4-cc0f617b20a0), the same Dixon–Coles ratings behind the match predictions and season projection. Each club has an attack and a defence rating, fitted to league results with older matches counting less. For every pair of clubs we take the expected goals at home and away, and average across all 38 games a club would play. Points per match come from the same calibrated scoreline probabilities: 3 × win probability + draw probability.
The ratings are estimates. The season projection publishes 2,000 draws from the fit's own uncertainty; we rank the clubs in each draw, and the rank range is the 10th to 90th percentile. Clubs with overlapping ranges are hard to separate. Rankings are recalculated whenever the season projection is updated, usually after each round of matches.
The rankings use only match results, with no pundit polls or manual adjustments. Read the full methodology or check the model's prediction record.
Questions about the Premier League power rankings
Who is the best team in the Premier League right now?
On the model's current ratings, Arsenal: an expected goal difference of +0.93 per match and 1.99 expected points per match against the rest of the league. They ranked first in 52% of 2,000 rating draws; Manchester City are next.
How are these power rankings calculated?
They use the same Dixon–Coles team ratings that produce the match predictions and the predicted table, fitted to Premier League results with recent matches weighted more heavily. Each club is ranked by expected goal difference per match in a full home-and-away round against the other 19 clubs. No pundit opinion or manual adjustment is added.
Why is a club ranked differently from the league table?
The table counts points already won; the rating estimates how strong a club is now from goals scored and conceded, adjusted for opponents and venue. Everton (6th in the table, 12th on rating), Ipswich (14th in the table, 20th on rating), Leeds United (9th in the table, 13th on rating) are higher in the table than their rating. Manchester United (18th in the table, 8th on rating), Nottingham Forest (16th in the table, 11th on rating) are rated better than their table position.