We live in aggregates because we have to. At least, that's how the Premier League awards its champion. You play 38 games, one home and away against everyone, and whoever accrues the best combination of three-point wins and one-point draws gets to lift the trophy. If, somehow, there's still a tie, then we look to another aggregate: the differential between goals scored and conceded across those 38 games.
Unlike in tournament-based competitions, the best team -- on aggregate, by points won -- always wins the league.
Of course, in a single moment, no team is exactly its aggregate. No, each team is performing somewhere along the spectrum of matches -- good, bad, or average -- that add up to make its aggregate.
Take Liverpool's match with Wolves last weekend. At halftime, the league leaders were on pace to win 4-0 -- well ahead of their seasonlong aggregate of plus-1.3 per game. Then, in the second half, they got outshot 10-0 and were outscored 1-0. Two vastly different halves add up to a performance well below their 25-game aggregate.
At this point in the season, the league table and each team's goal differential do a pretty good job of summarizing the average performance level of all 20 teams in England's top flight. But it doesn't really tell you anything about how they got there -- including the top end of what teams have proven capable of this season.
Today, we're going to move beyond the aggregates. To do so, we've ranked all 20 Premier League teams, in order, by how good their best game was, which in turn tells us how high each team's ceiling is.
The methodology
As I've written about before, analysis from Ben Torvaney -- formerly of AC Milan and now working for Liverpool's former head of research, Ian Graham, at the consultancy Ludonautics -- found that the best goal-based metric for assessing team quality was a blend of 70% expected goals and 30% goals.
I like this, particularly, because it's pretty intuitive. Hook any soccer manager, or fan, or player, up to a lie detector, and they'll admit that creating better chances than your opponent is the main objective in a match. But it's also not the only thing.
When the score changes, teams switch up their objectives. And, of course, the ability to convert your chances and prevent your opponent from doing the same is important -- if not as important as the ability to generate the chances in the first place.
Based on Torvaney's metric, here's how the league sorts out through 25 matches:
These ratings are adjusted slightly for the quality of the opposition and for home-field advantage. By the 70-30 metric, home teams average plus-0.37 per game, while away teams, of course, average the inverse.
And so, this is the metric we'll be using to determine each team's best performance in a match this season: 70% expected goals, 30% actual goals, adjusted for the location of the match and the quality of the opponent.
All data comes from FBref and Stats Perform.
20. Ipswich Town (team rating: minus-1.01)
- 0.88 adjusted rating (0.74 performance, plus 0.14 opponent)
- Away vs. Tottenham on Nov. 10, 2024
- Score: 2-1
- xG: 1.6-1.5
The "performance" rating here is 0.37 for the xG and goals, plus another 0.37 because Ipswich were on the road. (Home teams lose 0.37 from their performance rating.) And then it gets a further boost of plus-0.14 because that's Tottenham's rating so far this season.
Consider this Ipswich's ceiling: a 2-1 away win where the chances were roughly equal against the 12th-placed team in the table.
19. Leicester City (team rating: minus-0.95)
- 1.23 adjusted rating (1.09 performance, plus 0.14 opponent)
- Away vs. Tottenham on Jan. 26, 2025
- Score: 2-1
- xG: 1.6-1.0
While this is meant to be an exercise in showcasing each team's ceiling, we're also seeing which team might have the lowest floor in the league, too.
18. Aston Villa (team rating: 0.02)
- 1.28 adjusted rating (0.68 performance, plus 0.60 opponent)
- Home vs. Bournemouth on Oct. 26, 2024
- Score: 1-1
- xG: 1.8-0.3
While Unai Emery is one of the best cup managers of his generation, he has never quite had the same level of success in league play, in part, because he has been unwilling to dial up the variance and play the kinds of more open matches that might allow his teams to dominate more often but also occasionally get upset. See: Villa finishing top eight in the Champions League's league phase and landing 18th on this list.
17. Everton (team rating: minus-0.22)
- 1.42 adjusted rating (2.37 performance, minus 0.95 opponent)
- Home vs. Leicester City on Feb. 1, 2025
- Score: 4-0
- xG: 2.5-0.3
And we have our first dominant win over a bad team. Since Leicester have a minus-0.95 rating for the season, that gets subtracted from Everton's overall rating for this match.
16. Southampton (team rating: minus-1.32)
- 1.49 adjusted rating (1.12 performance, plus 0.37 opponent)
- Away vs. Newcastle Utd on Aug. 17, 2024
- Score: 1-0
- xG: 1.8-0.3
And we have our first and only ... loss? When this happened on the opening weekend, we all should've known this wouldn't be Southampton's year.
15. West Ham United (team rating: minus-0.42)
- 1.62 adjusted rating (2.63 performance, minus 1.01 opponent)
- Home vs. Ipswich Town on Oct. 5, 2024
- Score: 4-1
- xG: 3.6-0.6
14. Wolves (team rating: minus-0.43)
- 1.66 adjusted rating (1.48 performance, plus 0.18 opponent)
- Away vs. Fulham on Nov. 23, 2024
- Score: 4-1
- xG: 1.3-1.0
13. Fulham (team rating: 0.18)
- 1.67 adjusted rating (1.16 performance, plus 0.51 opponent)
- Away vs. Chelsea on Dec. 26, 2024
- Score: 2-1
- xG: 1.8-1.1
Fulham are the best team in the bottom 10 of these rankings. But when your leading non-penalty goal scorer is Alex Iwobi, who really has developed into a nice midfielder but whose main skill is moving the ball toward the opposition box and creating chances for others, your ceiling can only be so high.
12. Brighton (team rating: minus-0.09)
- 1.68 adjusted rating (1.90 performance, minus 0.22 opponent)
- Away vs. Everton on Aug. 17, 2024
- Score: 3-0
- xG: 1.4-0.5
There's no red-card adjustment in these ratings, and this was the game when Kaoru Mitoma nearly made Ashley Young retroactively retire five years ago. Brighton played more than 20 minutes of their first game of the season up a man. Throw that into the mix, and Brighton's best game was really last week's 3-0 win at home to Chelsea (1.4 xG to 0.5).
Uncoincidentally, Brighton's best games are the ones when Mitoma briefly becomes the best player in the Premier League.
11. Crystal Palace (team rating: minus-0.04)
- 1.88 adjusted rating (2.02 performance, minus 0.14 opponent)
- Away vs. Manchester United on Feb. 2, 2025
- Score: 2-0
- xG: 2.6-1.1
A dominant, multigoal win at Old Trafford ... only lands you in 11th place on this list.
10. Manchester United (team rating: minus-0.14)
ESPN FC Live's Mark Donaldson is joined by Shaka Hislop to discuss what competitions Ruben Amorim's Manchester United should prioritise.
- 2.02 adjusted rating (1.51 performance, plus 0.51 opponent)
- Away vs. Manchester City on Dec. 15, 2024
- Score: 2-1
- xG: 2.1-0.9
A decisive victory on the road against the four-time-defending Premier League champs ... only lands you in 10th place on this list.
9. Arsenal (team rating: 0.80)
- 2.25 adjusted rating (2.67 performance, minus 0.42 opponent)
- Away vs. West Ham on Nov. 30, 2024
- Score: 5-2
- xG: 3.5-1.5
This is the most useful part of this exercise.
Due to a combination of reasons -- tactics, injuries, player-personnel decisions, red cards, Raheem Sterling beginning his professional career at age 16 -- Arsenal just haven't had many matches where they've totally overwhelmed their opponent and eliminated any real chance of getting done in by bad luck.
But didn't they beat Manchester City, uh, 5-1? Yes, but that was in London, City's rating isn't all that special this season, and that was more due to some really impressive finishing than overwhelming chance creation:
Arsenal probably have the highest floor of any team in the world. The challenge going forward, if there even is one, is to find a way to raise that ceiling.
8. Newcastle United (team rating: 0.37)
- 2.40 adjusted rating (3.35 performance, minus 0.95 opponent)
- Home vs. Leicester City on Dec. 14, 2024
- Score: 4-0
- xG: 3.8-0.2
7. Bournemouth (team rating: 0.60)
- 2.50 adjusted rating (2.93 performance, minus 0.43 opponent)
- Away vs. Wolves on Nov. 30, 2024
- Score: 4-2
- xG: 3.3-0.5
Andoni Iraola's side is the third-rated team in the league by our adjusted goal metric, and while it has had some impressive performances, it's more because it has limited the bad matches. Along with Arsenal and Liverpool, Bournemouth are one of only three teams to not produce a single match with a minus-1.0 rating or worse.
6. Chelsea (team rating: 0.51)
- 2.77 adjusted rating (4.09 performance, minus 1.32 opponent)
- Away vs. Southampton on Dec. 4, 2024
- Score: 5-1
- xG: 5.2-1.6
5. Manchester City (team rating: 0.51)
- 2.91 adjusted rating (3.92 performance, minus 1.01 opponent)
- Away vs. Ipswich Town on Jan.19, 2025
- Score: 6-0
- xG: 3.0-0.5
Craig Burley slams Manchester City's attitude in a limp second leg performance as Real Madrid advance at the Bernabeu.
Both City's and Chelsea's best matches might speak to the relative flaws of both teams this season. Neither one has any problem annihilating the worst teams in the league, but once you go higher up the table -- to teams with much more talent and high-level coaching, games you can't just win by virtue of having a better roster -- they have both encountered a bunch of problems.
4. Nottingham Forest (team rating: 0.23)
- 3.32 adjusted rating (3.41 performance, minus 0.09 opponent)
- Home vs. Brighton on Feb. 1, 2025
- Score: 7-0
- xG: 3.3-0.9
I'm not sure there's ever been a single game that has altered my perception of a team more, this deep in a season, than Forest's 7-0 beatdown of Brighton. Up until this point, they'd managed with a bunch of one-goal wins and a lot of good fortune in scoring the first goals in their games. They seemed like the classic not-as-good-as-their-record team, and then it seemed like it was all starting to fall apart with the 5-0 loss against the classic much-better-than-their-record team in Bournemouth.
And then, well, it all whiplashed back in the other direction the following weekend with the 7-0 win against Brighton. I still wouldn't bet on Forest to finish in the top five, but if you're in third place 25 games into the season and you've shown that you can beat a not-bad team 7-0? You've got a real shot of qualifying for the Champions League.
3. Brentford (team rating: minus-0.00)
- 3.42 adjusted rating (4.74 performance, minus 1.32 opponent)
- Away vs. Southampton on Jan. 4, 2025
- Score: 5-0
- xG: 4.4-0.3
Per FBref, Brentford have the second-lowest wage bill in the league. Thomas Frank is magic.
2. Tottenham (team rating: 0.14)
- 3.51 adjusted rating (3.65 performance, minus 0.14 opponent)
- Away vs. Manchester United on Sept. 29, 2024
- Score: 3-0
- xG: 4.4-1.0
All of the discussions over Ange Postecoglu's job security were really missing the forest for the trees. This season, Tottenham produced the second-best single-game performance of any team in the league. And, as you'll soon see, they were also unwilling participants on the other side of the best game of the season.
I'm not sure I can remember another Premier League team capable of putting up some of the best and worst games you'll ever see -- in the same season. I hope Postecoglu manages Tottenham forever.
1. Liverpool (team rating: 1.26)
- 4.42 adjusted rating (4.28 performance, plus 0.14 opponent)
- Away vs. Tottenham on Dec. 22, 2024
- Score: 6-3
- xG: 5.6-1.3
Of the 10 highest-rated games so far this season, four of them were played by Liverpool. They've produced games with a rating of 1.5 or better nine times, while no one else has more than six. And they've produced games with a rating of 1.0 or better 14 times, while no one else has more than nine.
The 6-3 win over Spurs was also, by far, the biggest xG margin of the season, and, incredibly, it does not include a penalty kick. While Arne Slot has certainly tweaked the way Liverpool play, his team hasn't lost the thing that made Jurgen Klopp's teams so special: the ability to just blow their opponents away.