Premier League Power Rankings: Rating all 20 teams on performances, not results
Ryan O’HanlonJan 23, 2026, 05:00 AM ETCloseRyan O’Hanlon is a staff writer for ESPN.com. He’s also the author of “Net Gains: Inside the Beautiful Game’s Analytics Revolution.”The top four teams in the Premier League have played eight games over the past two matchdays. Care to guess how many combined points they’ve won?It’s six. That’s six out of a potential 24 points — a rip-roaring pace of 0.75 points per game. West Ham United are currently on a 0.77-point-per-game pace. They’re also currently in 18th place and just about every oddsmaker that attempts to predict these things gives them a much-greater-than-50% chance of getting relegated.And it might actually be even worse than the top-line numbers look. Across those eight matches, Arsenal, Manchester City, Aston Villa, and Liverpool have combined for zero wins. Meanwhile, West Ham, Leeds United, Nottingham Forest, and Everton have ripped off four combined wins and 14 combined points over that same stretch.But if we stop picking on the top four for a second and zoom out, a couple more incredible facts become clear. Over the past two weeks, zero Premier League teams have won more than one game. And over the past three weeks? Zero Premier League teams have won more than one games.As I wrote about a couple weeks ago, the Premier League is having something of a contextual crisis because the quality of attacking play is quite low — but more so because the quality of the average team is higher than ever before. Every fanbase seems upset because every team is dropping points to teams they usually don’t drop points to.So, in an effort to get a better handle on the state of the Premier League, we’re breaking out everyone’s favorite 13-letter, two-word phrase: power rankings! Welcome to the first edition of ESPN’s Premier League Power Rankings, our monthly look at the actual strongest and weakest teams.- Hiring a caretaker manager: Big mistake for Man United?- Ranking the top 15 USMNT U-21 prospects of 2026- Predicting Premier League table: How all 20 teams will finishHow the Premier League Power Rankings workFirst, a piece of advice: Do not use this list to try to beat the betting markets. You will lose — eventually.Now, to what the ratings behind this list represent. These aren’t meant to be the most accurate power rankings in the history of power rankings — the goal here is to create something that is simple and intuitive but still has some predictive power. The value comes from comparing the rankings to the table, seeing what looks different, and trying to pry apart the reasons why.As for the metrics that drive these team ratings, there are four inputs. The first is simply the team’s non-penalty expected-goal differential. This is the best single metric we have for predicting future performance. The second is the estimated squad value from Transfermarkt, another metric that has been shown plenty of times to have its own predictive power. Consider this a stand-in for a team’s talent level.That’s already enough for a solid power-rating system right there, but it’s boring to just say “xG and money” even if it’s usually right. With some help from Kevin Cole of Unexpected Points, I looked at a bunch of other metrics and found two that improved the predictions a bit and also made some intuitive sense with how we think soccer works.The first is PPDA, or passes allowed per defensive action in what is essentially the opposition half. This is the most widely-used indicator of a team’s pressing intensity, and my guess for why it helps predictions is that teams that are able to break up play high up the field have access to dominant performances that teams who don’t press aggressively can’t produce. These teams are also less likely to get dominated.The second is pass completion percentage outside of the attacking third — a number I first saw at the site Markstats. This could be viewed as the inverse of PPDA — a team’s ability to withstand an opposition press. These teams are less likely to get blown up and stagnated, and there’s probably also some indicator here of how aggressive opponents play against them. If opposition teams are unwilling to pressure a side in buildup play — thus leading to a high completion percentage — then that tells us something about the team’s quality, too.In the interest of simplicity, our rankings are comprised of just these four numbers. First, we use all of it to rate the teams on a scale of 100, and then convert those ratings into what is essentially a projection of the team’s goal differential in an average match. Here’s how our rankings look:This one is pretty obvious. Based on pure underlying performance through 22 games, there’s Arsenal — and then there’s a group of four other teams. They’re also perfect in the Champions League.This is, quite clearly, the best team in the world right now:What’s of note here is that Man City are significantly closer to Arsenal in the power ratings than they are in the expected-goals table. Why is that? City came into the season with the second-most valuable squad in the league, and they’re completing nearly 90% of their passes outside of the attacking third.When a journalist straight up asks you if another coach is being lined up to replace you, it means … OK, it means a lot of different things that have nothing to do with Arne Slot or Liverpool and more to do with the general state of the world. But still: something like that usually doesn’t happen unless there’s a growing level of unrest around the team you’re coaching.Despite drawing their last four Premier League matches, Liverpool have been one of the better teams in the league over the last 10 games:A lot of the dissatisfaction around the team, I think, is ultimately an aesthetic critique, rather than a qualitative one.Liverpool have created enough chances to win most of their recent matches, but fans are used to watching Jürgen Klopp’s sides fly up and down the field and guarantee matches with lots of goals. Arne Slot’s current version of Liverpool might be effective, but they are no longer producing anything like Klopp’s heavy-metal football.This is a statement I’ve rarely been able to write: seven Premier League teams are pressing more aggressively than Liverpool so far this season.Despite an xG differential that’s closer to Leeds than Liverpool, Chelsea’s rating isn’t far back of the defending champs. That’s because of their massive, talented squad, and an aggressive pressing-and-possessing style that has tended to produce results in the Premier League.We’ll see if it continues under new manager Liam Rosenior, whose team had less than 50% of the ball and attempted only six shots in a fortunate 2-0 win over Brentford last weekend.play0:31Rosenior admits he has to ‘earn’ support from Chelsea fansLiam Rosenior speaks after Chelsea’s 1-0 win over Pafos FC in the Champions League.The recent win over Man City was huge and impressive, but I’m not sure we actually learned that much about Manchester United — or, more specifically, Man United under Michael Carrick.City came into the match with what I’d say was a fourth-choice backline at best, and United sat back and exploited those four players over and over and over again. That approach isn’t going to be an option in most remaining matches, and when it is, those opponents are going to have a much better group of defensive personnel than City did.That said, this team was simply just pretty good under Ruben Amorim. If Carrick can keep it going, this is probably one of the five best teams in a league that is likely to get five Champions League places next season.The one advantage Newcastle have over Manchester United in the eyes of our ratings is that they’ve been more controlled in possession while producing similar-quality performances with similar levels of pressing intensity.Man United are completing their lowest percentage of pre-final-third passes in any of the past nine seasons. Eddie Howe’s side, meanwhile, are completing a higher percentage of passes outside the attacking third than in any season over that same stretch.It hasn’t borne itself out in results or even performances, really, but Brighton are playing like a team that tends to win a lot of points: a league-low PPDA and a buildup-pass completion percentage that’s higher than everyone other than last season’s top four and Aston Villa. They also have a theoretically deep and talented squad.I’m not sure I buy it — not every manager and group of players can turn pressing and possession into points — but if you squint hard enough, you can convince yourself that the Seagulls have a higher ceiling than anyone outside of the six teams ahead of them in these rankings.If you’re still wondering if the Premier League is bad or if all of the Premier League teams are good, then allow me to alert you to what’s happening in North London.That’s what Tottenham did against West Ham, at home.And then they did this against Borussia Dortmund, at home:Their underlying numbers in the league are truly horrific — 16th-best! — and yet they’re currently in fifth place with one game to go in the Champions League’s league phase. They’re one point ahead of the PSG, the defending European champs.I’m considering this a win for the ratings.A … win? A win?!?! This team is tied for second place and you’re telling us that a homemade algorithm that puts them behind Brighton and Spurs is doing a good job? It’s rating them as the most average team in the league!Well, to begin the argument with my italicized self: Power rankings are pointless if they just match the standings. If you want that, then … just go look at the standings! We’re trying to isolate the underlying factors that drive winning and then use them to identify the teams that are most likely to win points in the future.At the same time, just looking at xG would put this team all the way down in 12th. Villa are more talented than that, and by highlighting their weird, City-lite approach of patient possession without pressing, I think we’re picking up on some of the reasons why Unai Emery’s teams tend to outperform their underlying numbers every season.play2:42Why Aston Villa are ‘not title contenders’ after loss to EvertonFrank Leboeuf and Craig Burley review Aston Villa’s 1-0 defeat to Everton in the Premier League.This is one of those situations where, if I just looked at the numbers and watched the games, I would tell you that Brentford are a much better team than Tottenham. But if my life was on the line and Brentford and Tottenham were playing each other on a neutral field, I would probably pick Tottenham? And I would probably die? I don’t know.Brentford have been fantastic this season, but they’ve done it with a style that’s hard to consistently turn into dominant performances.By every metric other than points, this is a slightly-below-average team: they press like one, they pass like one, they create and concede chances like one, and they have the talent like one.Six teams in the league are completing less than 80% of their passes outside of the attacking third, and five of those teams have a PPDA in the bottom-six of the league. That’s pretty common: you sit back, absorb pressure, and then attack with vertical, higher-risk passes. It’s been the underdog playbook since someone in Scotland recognized that it wasn’t against the rules to pass the ball to a teammate.As for the other team of the six, though? It’s Andoni Iraola’s Bournemouth, who are pressing more aggressively than everyone other than Brighton and Chelsea.Marco Silva’s side gets a boost because of how secure they are with the ball outside of the attacking third. They’re below average everywhere else, but they’re completing a higher percentage of their buildup passes than all but six other sides.Silva is getting more out of this side — an aging team without much, if any, high-end talent — than he should, and I think the unique tactical profile helps explain some of that.Had we started these rankings earlier, I think they would’ve predicted the current Palace fall off. Not, you know, the cultural collapse, where they let their captain leave for a smaller fee than they were offered over the summer and then their manager announced he’d be leaving after the season and then lit the club’s team-building process on fire in a post-game press chat.But despite some really impressive play over the first couple months of the season, Palace were doing it without any control: of the ball, of opposition territory.If a big club is going to hire Oliver Glasner, they need to figure out whether he’ll be able to do something different with a better set of players and a much more demanding performance threshold.They’re up to 16th in the table, and they’re currently closer to 10th place than the relegation zone. My “Daniel Farke for Manager of the Year” campaign cannot be stopped.This one surprised me, more than any other entry on the list. Everton are in 10th! Jack Grealish! Three points back of fifth! Just beat Aston Villa! And yet these ratings have them close to equivalent to Wolves.Why is that? Their xG differential is only 16th-best and they’re still playing pure, distilled David Moyes ball, without much pressing or controlled possession.I think I still have Wolves in my “three worst teams” in the league tier, but their position here tells us a couple things: they’ve been able to press and control possession in a way that other relegation-level teams typically don’t, and their underlying performances have been much better than their historically terrible point total suggests.One reason there probably won’t be a relegation battle is that the team currently in 18th place is actually the 18th-best team in the league.Another reason there probably won’t be a relegation battle is that the team with the second-worst xG differential (Sunderland) is currently just two points behind the team that won the Premier League last season (Liverpool).Here’s how I would sum up the Sunderland situation: they currently have a better chance of qualifying for the Champions League than being relegated, but they’re also going to be one of the favorites for relegation before next season starts.Data analysts have often talked about the so-called Messi test: you could determine the validity of whatever player-value metric you created by seeing if Lionel Messi ranked at the top. For these ratings, I did something similar: as long as Burnley were at the bottom, I knew I was doing something right.
已发布: 2026-01-23 13:43:00
来源: www.espn.com
