The Premier League Season So Far: Every Team Ranked on the Numbers

Five games is a small sample and this page will not pretend otherwise - at this stage of a Premier League season a single result still moves a per-game number noticeably, and the finishing rates below are the least stable figures on the page. Below is the division ranked on what teams are actually doing rather than what the table has given them for it, with every figure taken from our Premier League stats page, which covers league games only and updates as the season goes. The underlying expected-goals numbers behind much of it live on the Premier League xG table.

Two numbers frame the rest. Manchester City have taken 5.3 points more than their underlying numbers say they should have; AFC Bournemouth have 4.9 fewer. That is the widest the table has been stretched in either direction.

The clearest single case is Manchester United. They lead the Premier League for shots at 22.6 a game and for shots inside the box at 12.6, and they are converting 7.1 per cent of them. They sit 12th.

Attacking threat

Ranked on the volume and the quality of what each side creates.

Manchester United lead the Premier League for shots at 22.6 a game, and take 12.6 of them inside the box. They are scoring 1.6 a game from that, a conversion rate of 7.1 per cent. That is the 16th best conversion rate of 20 teams, so the volume is not turning into goals.

Brighton & Hove Albion carve out 4.4 big chances a game, the most in the Premier League, and convert 40.9 per cent of them. Granit Xhaka of Sunderland leads the division's players with 5 big chances created.

Brighton & Hove Albion generate 2.2 expected goals a game and are scoring 3.2, which puts them 1 a game ahead of what their chances have been worth. Expected goals weight every shot by how often a chance of that type is scored, so it describes the quality of what a side is creating rather than how the finishing has gone.

Tottenham Hotspur win 7.6 corners a game. They also hit 23.2 crosses a game, 3rd of 20 in the division, which is usually the same story told twice: sides that attack down the flanks force more of both.

Brighton & Hove Albion take 12.6 shots inside the box a game, 66.3 per cent of their total. Shot location is the single biggest driver of chance quality, so this table tends to describe who is creating properly rather than shooting on sight.

Defending

Fewest is best in every table below.

Nottingham Forest concede the fewest shots in the Premier League at 10.4 a game. Opponents are converting 9.6 per cent of them, which still leaves 1 goals against a game and a goals-conceded rank of 9th of 20 - a reminder that a good defence and a good goals-against column are not the same thing.

Arsenal give up 0.92 expected goals a game, the best figure in the Premier League. They are conceding 0.8, close to what the chances deserve.

Arsenal concede 3.2 corners a game, fewest in the division and have kept 3 clean sheets in 5. Corners conceded is a decent proxy for how much sustained pressure a side absorbs.

Nottingham Forest concede 1.2 big chances a game, and opponents are putting away 50 per cent of the ones they get. A big chance is one a striker is expected to score, so this is the table that matters most at the defensive end.

Arsenal allow 5.2 shots inside their own box a game, and 49.1 per cent of the shots they face come from there. Keeping opponents outside the area is most of defending.

Possession and play style

How each side plays, rather than how well.

Manchester City play 628.6 passes a game at 89.2 per cent completion, with 63.2 per cent of the ball. Volume and possession usually travel together; completion rate is what separates keeping the ball from moving it forward.

Manchester United hit 26.4 crosses a game, of which 6.8 find a teammate - an accuracy of 25.8 per cent. A high cross count with a low accuracy is often a side with no other way through.

Nottingham Forest commit 14.6 fouls a game, the most in the Premier League, and pick up 2.6 yellow cards a game for it. A high foul count usually means a side that presses hard or breaks play up deliberately. Myles Lewis-Skelly of Arsenal leads the division's players at 2.63 fouls per 90 minutes.

Crystal Palace make 22.4 tackles a game, and 32 tackles and interceptions combined. Anan Khalaili of Crystal Palace makes the most of any player in the division at 4.97 per 90 minutes.

The players

Minimum 270 minutes for the shooting and creating tables, so a player needs roughly three full games to appear.

Erling Haaland of Manchester City leads the Premier League with 5 goals from 20 shots, against an expected goals figure of 4.42 - 0.58 more than those chances were worth.

Pascal Groß of Brighton & Hove Albion leads for assists with 3, from 15 chances created and 4 of those big ones. Assists depend on someone else finishing, which is why chances created is the steadier measure of a creator.

Erling Haaland of Manchester City has taken 20 shots, more than anyone in the Premier League, hitting the target with 10 and scoring 5 - a conversion rate of 25 per cent.

Granit Xhaka of Sunderland has created 5 big chances, the most in the Premier League, which have so far produced 0 assists. The gap between those two numbers is finishing, not creation.

What the table is hiding

Expected points take every chance a side has created and conceded and ask what a league table would look like if those chances had been taken at an average rate. The gap between that and the real table is the part of the season that has not been earned - in either direction.

None of this says the over-performers are about to collapse or that the under-performers are about to win six in a row. Finishing is a real skill and some sides out-run their numbers for a whole season.

What we can say is what we have measured. Across 1,245 fixtures we tested both halves of this idea. Sides whose results were running more than half a goal a game ahead of what their chances deserved went on to win 3.2 percentage points less often than the model rated them - a real, usable correction. Sides running the same distance behind beat their rating by only 0.8 points, which is close enough to nothing to ignore. So the tradeable half of this table is fading the teams it is flattering, not backing the ones it is punishing. That asymmetry is the single most useful thing on this page.

FAQs

Which Premier League team creates the most chances? Manchester United lead for shots at 22.6 a game and shots inside the box at 12.6. Brighton & Hove Albion lead for big chances created at 4.4 a game.

Which Premier League team has the best defence? Arsenal concede the fewest expected goals at 0.92 a game and the fewest shots at 10.6, after 5 games.

Who is the Premier League top scorer this season? Erling Haaland of Manchester City with 5 goals. Pascal Groß of Brighton & Hove Albion leads for assists with 3.

Which Premier League teams are over-performing? Manchester City have 5.3 points more than their chances deserve; AFC Bournemouth have 4.9 fewer. Our own testing says the tradeable side of that gap is fading the over-performers rather than backing the under-performers.