Premier League reranked: Why Man City won’t beat Arsenal

We have a title race, we may have a fight for fourth, we could have a chase for fifth and, although it looked dormant a few weeks ago, the relegation battle might even be back from the dead, too.

With the Premier League about to go into hyper-speed and teams to play approximately 4,000 matches each from now until the New Year, it’s time to take a quick step back from the action and assess the state of the league from top to bottom. And to do that, Ryan O’Hanlon and Bill Connelly are back with the fourth edition of their Premier League power rankings.

Once again, they have each ranked all 20 sides, 1-20, and then combined their rankings to produce one master ranking. The criteria: Who we think would win a match if any of these teams played each other on a neutral field in the near future.

The updated rankings are below, followed by some analysis of the most notable changes (or non-changes) from the previous edition of the rankings.

The updated rankings

December and November’s rankings — the combination of Bill and Ryan’s individual rankings — are listed along with the new adjustments, plus each team’s present points totals on the Premier League table and goal differential:

Arsenal is the best team in the league at this moment

As it says above, we’re ranking teams based on who we think would win a match if any of these teams played each other on a neutral field in the near future, but does that mean today? Because if so, both Everton and Bournemouth probably have justification for being ranked in the top five. Does it mean “at some random point over the next few weeks?” If so, Manchester City could get healthier, shift into gear and begin its seemingly inevitable title push.

Splitting the difference, however, gives us Arsenal in the top spot. The Gunners have simply been the best team in the league so far. They top the table, and some of their best performances have come in recent weeks.

Basically, Arsenal manager Mikel Arteta has crafted a team that combines the best of Liverpool with the best of Man City. They’re either first or second in most of the categories that Liverpool cares most about. They’re first in high turnovers from open play (40.3) and high turnover margin (+18.5 per match in open play), and Liverpool’s second in both categories. They’re tied for second with Liverpool in passes allowed per defensive action (9.7). They’re first in progressive passes per match (72.4), and Liverpool’s second.

Meanwhile, they’re also either first or second in most of the categories that City cares most about, from possession rate (60.1%, second only to City), average possession time (33.9 seconds, second only to City), total touches in the attacking third (237.1, second only to City), total touches allowed in the attacking third (102.1, first with City second). Throw in some last-ditch tenacity — they’ve blocked 33.4% of opponents’ shots, most in the league — and you’ve got an absolutely ferocious team, one with better fundamental stats than last year’s: Their +1.11 xG differential per match tops the +1.03 they had managed after 17 matches in 2022-23.

Arsenal’s 2-0 win over Brighton on Sunday sent quite a message. Against a Brighton team that controls the ball as well as anyone in the league not named Manchester City or Arsenal, the Gunners were in control of basically everything. Over the first 80 minutes, they outshot the Gulls 23-1 while producing more touches (628-574) and pass completions (438-396), and even dominating loose balls (ball recoveries: 32-25). When Brighton finally established some semblance of attack over the last 10 minutes, Arsenal put the game away with a counter-attack goal from Kai Havertz.

We knew that the Gunners were capable of offensive bursts — see: six goals against Lens, five against Sheffield United, four against Luton Town, etc. — and considering they allow the fewest goals and by far the lowest xG (expected goals) of any team in the league, we knew they were capable of domineering defense. But the way they told a solid Brighton team, “We’re going to do our things better than you, and we’re going to do your things better than you” … well … that’s something title winners do. — Connelly

Manchester City: fourth in the table, tied for second in our rankings?

Remember when Manchester City used to always have the ball?

Since 2017, only six Premier League teams have controlled more than 63% of the possession in their matches. Those teams were:

• Manchester City in 2017-18: 71%

• Manchester City in 2018-19: 67.9%

• Manchester City in 2021-22: 67.9%

• Manchester City in 2019-20: 66.2%

• Manchester City in 2022-23: 64.7%

• Manchester City in 2020-21: 63.5%

This season, City still lead the league in possession, but they’ve dropped down to their lowest rate since Pep Guardiola took over as manager: 62.2%. That’s still the 10th-highest rate in England since 2016, but it’s below the Liverpool team that won the title and the one that nearly won the quadruple, along with the one season when Maurizio Sarri managed Chelsea.

Of course, possession doesn’t equal domination, but City’s finishes in those six 63-plus-possession seasons were as follows: one, one, one, two, one, one. City’s current table position, this first time they’ve dipped below 63: fourth.

That’s not to say that there’s some magical kind of threshold where, if you surpass 63% possession, you automatically win the league. I repeat: City lost in 2019-20, and very nearly lost in both 2018-19 and 2021-22. Barcelona continues to lose out on LaLiga titles despite breaking the 63% mark nearly every year. Both Napoli and Bayer Leverkusen had one-offs where they got there and didn’t win their respective leagues. And the last time Paris Saint-Germain lost Ligue 1, they did so with 63.1% of the ball.

Rather, for City so far this season, the drop-off just seems to be a symptom of their larger struggles instead of a direct cause. The easiest explanation for what’s gone wrong: Erling Haaland has missed 15% of the season, and Kevin De Bruyne has played for 22 total minutes.

If you look back at all of those possession totals for City, you can see that there was a sudden decline from 2021-22 to last season: more than three percentage points. In order to get the most out of Haaland — a player who isn’t going to drop deep and combine with teammates, and also someone who simply isn’t going to touch the ball that often — City had to get a little more comfortable without the ball. And to get more comfortable without the ball, Guardiola adjusted by playing more traditional defenders than he had in the past.

Doing so, in turn, put a bigger burden on De Bruyne to generate opportunities for his teammates. In the past, City created loads of opportunities no matter who was on the field. They were dominant in 2018-19 when De Bruyne nearly missed the whole season and never really suffered in the matches he missed in the following seasons, either. Now, though, it just doesn’t work anywhere near as well without him and without Haaland turning this smaller load of possessions into even higher quality opportunities.

Throw in the departure of Ilkay Gündogan, a few games missed for Rodri and the arrival of players like Josko Gvardiol and Jérémy Doku, who aren’t as secure on the ball as previous City players, and you have a team that’s controlling less of the ball than ever before and worse at turning it into chances whenever they do get it. Through 18 matches, City are creating 1.7 non-penalty xG per match — roughly equal to what Manchester United did last year, tied for 15th-best mark since 2017, and only the third-best rate in the league this season, behind Newcastle and Liverpool.

It’s wild to say this about a team with a striker who scored a league-record 36 goals last season — at 22, in his first season in England — but if Manchester City doesn’t win the league this year, it looks like it’ll be because they don’t score enough goals. — O’Hanlon

Sean Dyche has Everton playing its Dychian best

Most Premier League points, past 12 matchdays: Aston Villa 29, Arsenal 26, Everton 25, Liverpool 25, Newcastle United 23.

Twelve matches isn’t enough to give us definitive conclusions about anything — what do we think this is, college football?! — but it’s a healthy sample. It’s about one-third of a Premier League season, after all, and it spans about a three-month period. Three months is a lot of sleep! And over three months’ worth of sleep, Everton has generated the third-most points in the best league in Europe. See all those effusive words about Arsenal above? The Gunners have generated only one more point.

With each passing day, Everton looks more and more like the team of Dyche’s dreams. They’ve outscored four December opponents by a combined 8-0. They’ve got the lowest possession rate (39.3%), the most direct attack (1.69 meters per second) and the lowest average number of passes per possession (3.7) of any team that isn’t currently in the relegation zone. They block 32.2% of shots (second in the league), and only 6.5% of opponents’ attempts are worth 0.3 xG or more (first). Consequently, only 28.0% of opponents’ shots end up on target (first). James Tarkowski (first in the league with 25 shots blocked) and Jarrad Branthwaite (second in total defensive interventions) might be the best center-back duo in the league.

This roster doesn’t seem to possess much genuinely elite talent — the market values at Transfermarkt rank them 15th in the league, with only one player’s services (defensive midfielder Amadou Onana) valued at more than €30 million — but Dyche has them playing brilliant Dycheball. And if not for the 10-point deduction that I waited as long as possible to mention, the Toffees would currently be in 10th place, ahead of Chelsea, tied with Brighton and only three points behind sixth-place Newcastle.

No matter how things play out from here, the point deduction will end up defining Everton’s season. There’s no way around that. Either they continue to play well and we continue to talk about how close to a spot in Europe they might be if not for the penalty, or their form starts to fade and, despite a mid-table overall performance, we talk about how close they are to the relegation zone. After all, even with their current torrid form, they’re still only seven points away from danger. Play poorly for two weeks, and that gap could be cut in half.

But so far, the deduction seems to have been strangely freeing. It likely gave Dyche an extra “us against the world” card to play — managers love those cards — and seems to have made Goodison Park once again one of the toughest places in the league to play. Everton has averaged a finish of 12.6 in the table over the past five years and could be on the way to something similar this season. But it would be just about the most encouraging and energizing 12th- or 13th-place finish you’ll ever see. — Connelly

Bournemouth continue to trust the process

Much to the surprise of, well, pretty much anyone who watched the Premier League last season, Bournemouth fired manager Gary O’Neil over the summer and replaced him with Andoni Iraola. O’Neil had saved the Cherries from seemingly assured relegation after taking over for Scott Parker — Iraola had never managed a team with a positive goal differential in a major European league.

The first few months of the season seemed to showcase the stupidity of the decision. O’Neil took over at Wolverhampton, and his side were tied for 11th at the end of October after flirting with relegation the year before. Bournemouth, meanwhile, were in 17th, just one point clear of the drop zone. Not only that, the performances were awful, too. Only Sheffield United had a worse expected-goal differential through October.

This could’ve been the moment when the club cut their losses, parted ways with their inexperienced foreign manager and replaced him with, say, Sam Allardyce or Alan Pardew or Steve Bruce. Someone, as they say, who knows his way around a relegation scrap.

Instead, Bournemouth hung onto Iraola — and it’s working out brilliantly. Since the start of November, only Everton, Aston Villa, Arsenal and Liverpool have won more points than Bournemouth. And really, only the Gunners have played better soccer since the end of October.

Here’s the non-penalty xG-differential table since the start of last month:

Bournemouth were somewhat fortunate not to be relegated last year since, by xG differential, they were easily one of the three worst teams. That — or some version of the idea that they couldn’t afford not to improve — drove them to replace O’Neil with Iraola, who, in his two seasons with Rayo Vallecano, produced proactive sides that finished mid-table in LaLiga despite a bottom-of-LaLiga payroll. Imagine what he might be able to do with a roster awash in all that Premier League TV money.

Well, this is what his team has done so far, compared to last season:

• Non-penalty xG: 1.35, up from 1.05

• Non-penalty xG allowed: 1.66, down from 1.71

• Touches in opposition box: 25.5, up from 19.6

• Final-third possession (field tilt): 43.4%, up from 35.9%

• Passes allowed per defensive action (PPDA): 11.86, down from 15.71

They’re creating more, conceding less, controlling more of the ball and pressing much more aggressively. Oh, and they’ve done it all with mostly the same roster as last season. Their three €20 million-plus summer signings — midfielder Alex Scott, forward Hamed Junior Traore and midfielder Tyler Adams — have combined for just 236 league minutes so far.

We shouldn’t write off their awful start, but this really looks like the case of a club finally catching up with its coach’s new ideas. If Bournemouth keep it up, Iraola should be a candidate for Manager of the Year. — O’Hanlon

Spurs, Brighton, Brentford find out that talent rules all

Brentford and Brighton are the two smartest clubs in the Premier League — both with top-down mandates from ownership to base decisions on objective information to the maximum extent such a thing is possible. Knowing a better way to play the game and a better way to identify players has allowed both teams to overachieve.

Spurs, meanwhile, seemed to experience the restorative powers of a special manager to start the season. The arrival of Ange Postecoglou immediately transformed the team’s style of play and vaulted them up to the top of the table despite the departure of Harry Kane over the summer.

Set-piece savvy, cost-effective recruitment and managerial brilliance all matter, but they can only take you so far. The midseason struggles of all three clubs point to a much simpler, much more important driver of success: just having good players.

Brentford have continued to exploit the basic analytical tenets of the sport. They take the highest quality shots in the league. They conceded the lowest quality shots in the league. And they create more value from throw-ins and set pieces than anyone in the league.

The stat expected possession value (xPV) calculates how much more likely each individual action makes a team to score, based mainly on where the ball moves to on the field. By looking at throw-ins and set pieces, we can begin to assess which teams are playing these situations most optimally. And through this lens, it’s Brentford, way ahead of everyone else:

Despite a sound process, they’ve struggled this season — especially recently — because their two best players, Ivan Toney and Bryan Mbeumo, have started a combined 15 games due to Toney’s gambling-related suspension and Mbeumo’s recent injury.

Brighton, meanwhile, aren’t moving their players or throw-ins into dangerous areas, but that was all fine when their center-backs were putting the soles of their feet on the ball, and it was eventually leading to goals. Roberto De Zerbi’s team is still dominating possession — not far off from City’s levels this season. It’s just not going anywhere this term. Their xG differential last season (+0.69) was top four, but this year it’s more than halved, all the way down to +0.3. The biggest reason why: They lost Alexis Mac Allister and Moises Caicedo, €150 million worth of midfielders, over the summer and still haven’t found a way to replace either one.

As for Spurs, they’ve still managed to score 34 non-penalty goals, second most in the league, after Kane’s departure. But the brilliance of Kane was that he was an attack all unto himself — his ability to create opportunities for himself and his teammates allowed for more defensive decisions to be made elsewhere in the lineup. With him gone, though, Spurs have had to completely destabilize the defense — 26.22 non-penalty xG allowed, seventh worst in the league — in order to produce that much attack.

In addition to Kane’s exit, Spurs lost their new star, James Maddison, to an ankle injury at the beginning of November. He hasn’t played since the 4-1 loss to Chelsea. Tottenham’s xG differential with Maddison on the field: plus-0.52 per 90 minutes. Without him? It’s minus-0.56. — O’Hanlon

Nottingham Forest is determined to give us a relegation fight

Since the initial August season preview, we’ve updated these power rankings three times. In all three instances, the three recently promoted teams — Burnley, Sheffield United and Luton Town — have occupied the bottom three positions. All you have to do is look at the table to understand why, of course: They’ve won a combined six of 50 matches thus far, four of 46 that weren’t against each other. No Premier League team has survived the drop with fewer than 34 points, and none of these three teams are on pace to top even 22. It’s looking pretty dire.

But for those of us who enjoy the pure morbidity of a good relegation fight, Nottingham Forest’s current form offers hope. In the past three months, they’ve taken just seven points from 12 matches. They’ve scored the third-fewest goals from shots worth the fourth-lowest xG, their best attacker (Taiwo Awoniyi) is dealing with another medium- to long-term injury, and only reasonably decent defense — they’ve allowed the seventh-most goals from shots worth the 10th-most xG, and center backs Willy Boly and Murillo have rather comfortably been their most consistently strong performers — has allowed them to secure three wins and enough draws to stay safe for now.

Over their next seven league matches, Forest will play both Newcastle and Bournemouth, one of the league’s most in-form teams, twice each. They also face first-place Arsenal and seventh-place Manchester United. If they don’t beat Bournemouth at home Saturday, their next most likely win might be a Jan. 20 trip to Brentford.