The Premier League's next evolution: is the obsession with possession coming to an end?
In Michael Cox's "The Mixer," he tells the story of Manchester United winger Andrei Kanchelskis (1991-95) muttering "English football is s---" as he stormed off the practice field, protesting what he felt were pointless crossing drills. Kanchelskis wasn't wrong. In the years following the Heysel ban, as English teams were reintegrated into European football competitions, it had quickly become clear that English football was stuck in the dark ages. No team advanced beyond the second round of the European Cup from 1991-93, and when the competition became the UEFA Champions League and introduced group play, nobody made it out of the group stage in 1994 or 1995.
While the game was evolving into something more controlled and creative in other parts of Europe, English football was still pretty reliant on long-bomb passing, hopeful crosses, and hard tackles.
Alex Ferguson, Kanchelskis' manager at the time of "s---," took these failures to heart. Long considered a "man-manager" (the soccer way of saying "players' coach") more than a master tactician, his biggest strength ended up becoming his adaptability. He brought more creativity to the club and he expanded his tactical repertoire. Both he and United evolved. They not only dominated the Premier League (eight titles between 1993-2003), but in 1997, they broke through to the Champions League semifinals. Two years later, they won the whole thing.
United served as a bellwether of sorts for the rest of the Premier League, giving other clubs a template from which to work, and the combination of tactical awareness and increasing money from TV rights -- which allowed top clubs to import choice playing and coaching talent from all over the globe -- would turn the league into a juggernaut. Liverpool won the Champions League, with Chelsea reaching the semifinals, in 2005. Arsenal made the finals in 2006, and three English teams made at least the semifinals in 2007, 2008, and 2009. United won its second title in 2008.
But innovation never stops. It is a constant race to stay ahead. Ferguson was kept from two more European crowns by Pep Guardiola and Barcelona in 2009 and 2011, and by the end of Ferguson's United tenure in 2013, the Premier League had again lost its edge. From 2010-13, English teams made only two combined semifinal appearances (United lost in the 2011 finals, Chelsea won an unlikely title in 2012) and in 2013, no team made it out of the Round of 16. Ferguson's retirement left a void not only for United, but also for this rich but increasingly directionless league.
It was somewhat symbolic that Barcelona was the team to hold Man United back. Those Barca teams, along with the Spain squads (loaded with Barca players) that won Euro 2008, the 2010 World Cup and Euro 2012, did more to redefine the game of soccer than any in recent times. Barca showed the world the platonic ideal of an ultra-patient, possession-first, kill-the-body-and-the-head-will-follow squad. By the time of Ferguson's retirement, Guardiola had left Barcelona and was preparing for three seasons at Bayern Munich. And by the time Guardiola came to England to coach Manchester City in 2016, the league's re-evolution had taken hold.
It has only picked up speed since.
Here are a few stats for the average Premier League team in different ranges of seasons:
*Direct Speed is an Opta measure that defines how quickly you are attempting to advance the ball vertically, presented in meters per second.
The average Premier League team is far more pragmatic than it was a few years ago. There are fewer wasted possessions (and fewer possessions overall), fewer low-percentage shots and fewer low-percentage passes. There are fewer long balls, too, and more commitment to passing triangles and retention. The data was already trending in this direction before Guardiola's arrival and it went into overdrive after. In short, English teams - in particular, the Big Six (Arsenal, Chelsea, Liverpool, Manchester United and Tottenham) -- have more control over the ball than they used to.
Taking Guardiola's average-skewing City out of the equation, average possession for the five other members of the Big Six only went up from 57 percent before Fergie's retirement to 58 percent. Still, while possession itself didn't change, teams' tendencies in possession did: the Big Six Sans City have gone from 503 passes over 103 possessions per 90 minutes before Fergie's retirement to 557 in 97 since. Their percentage of forward passes has dropped, and their direct speed has fallen from 1.8 meters per second to 1.5. (For context, 1.8 would have ranked 10th in 2013 but would rank second today. The league is far less "vertical" now.)
Adapting in this way, and continuing to spend on big talent, helped open the spigot for European wins again. After claiming just two of 16 Champions League semifinal spots from 2014-17, Liverpool made the finals in 2018, then Liverpool beat Tottenham in the finals in 2019. (City, though dominant, have fallen short thanks to back-to-back quarterfinal upsets against Premier League rivals: Liverpool in 2018, Spurs last year.)
It's too early to know what to make of this Premier League season; the teams are not yet through half of their 38 games and there's plenty of time for regression to the mean. But things have changed this fall.
Here is the same chart as above, but with 2019-20 data included:
Until this fall, almost any English team attempting to play the possession game like the "big boys" was rewarded with defeat. Over the past five seasons, only 11 teams outside of the Big Six have enjoyed a rate of possession over 50 percent: Everton (four times), Southampton (three), Bournemouth (two), Leicester City (one) and Swansea City (one). They averaged a ranking of 10.5 in the final league table, but never finished ahead of any Big Six teams while doing it.
Granted, no strategy was likely to regularly get you ahead of those six, but a lesser form of the possession game didn't always do much for these teams in the standings. Bournemouth finished 16th in 2016, for instance, and Southampton 17th in 2018. Meanwhile, the 10 most successful non-Big Six clubs of these five seasons (according to average points won per match), included the direct, vertical and 43 percent possession Leicester squad that won the league in 2016; the 2018-19 Wolves squad that had 46 percent possession, played bend-don't-break defense and allowed almost no clean shots on goal; and the definitively old-school 2017-18 Burnley team that enjoyed 43 percent possession, threw bodies in front of every opposing shot and attempted mostly forward passes, take-ons and crosses on offense.
Possession has primarily been the favorite's tactic, not the underdog's. But that has not been the story in 2019-20.
1. There are a few identity crises within the Big Six. I have mostly spoken in this piece about the six richest clubs as a monolith, deploying the same strategy and equally sharing the top six spots in the table among each other. That, of course, is not how things have played out, especially since Guardiola's second season. The 2017-18 and 2018-19 City squads dominated the Premier League at a level rarely seen, with Jurgen Klopp's Liverpool joining them in an all-time, one-on-one race last spring. Their runaway success appears to have led to identity crises and increased managerial churn at their immediate rivals
Arsenal just fired manager Unai Emery and were, until Monday's win at West Ham, closer to the relegation zone than fifth place. Spurs ran out of gas under manager Mauricio Pochettino and recently handed the reins to former Chelsea and United manager Jose Mourinho. They are seventh. Manchester United finished sixth last season and have had to deploy a solid recent run of form to get back to that level now. City have lost some of their form, too.
It's a bit of a chicken-or-egg situation here. Did slippage from a few big clubs open the door for others, or did familiarity and the rise of certain clubs and tactics speed up the slip? Or is it both? Teams definitely adjust over time no matter what your style: you have to get better at the style to retain the same level of success. Some members of the Big Six most certainly haven't gotten better.
Also, the possession game has been a bit more egalitarian this fall. At 58 percent, Brendan Rogers' Leicester have the best possession rate for a non-Big Six team since Southampton in 2013-14, while Brighton, Everton, and Norwich City are all at 50 percent or higher. That none of those last three are higher than 12th in the table tells us something about possession's slipping correlation with winning, though Leicester are second heading into the festive period. They're a special case, perhaps reaping the financial dividends sown from Champions League participation and the sale of some of 2016's star players (N'Golo Kante, Riyad Mahrez, Harry Maguire).
2. Teams appear more capable of following the "Beating Man City" blueprint. Guardiola's side look downright mortal in his fourth season. While Liverpool has managed to somehow upgrade last year's form (2.9 points won per match currently), City are 14 points back. Granted, City's "mortal" is most clubs' "fantastic." They are averaging two points per match and have once again advanced to the Champions League last-16 with ease. Still, slippage is slippage.
So what's gone wrong? Offensively, not much of anything. City has scored 44 goals in 16 matches, which projects to 105 for the season. Only one club (2017-18 Manchester City) has scored more. They are taking possession in their opponent's defensive third more than ever, and more sequences of play are leading to shots than ever. While their 65 percent possession rate is their lowest in three years, it's still the highest in the league.
No, their main problem is in defense as Premier League opponents appear more capable than ever of exploiting the vulnerabilities of Guardiola's high-line system. They've given up at least two goals in all six league "non-wins" (two draws, four losses), and their 1.19 goals allowed per game belies that of an eighth- or ninth-place team, not a contender. These issues have been twofold. First, despite having more money at their disposal than most nation-states, City have personnel problems. Club icon Vincent Kompany is gone, center-back Aymeric Laporte is out with a long-term knee injury and another center-back, John Stones, missed more than a month with a muscle injury. The closest thing to a steadying presence in the back is Nicolas Otamendi, and he's three years past his peak.
Beyond that, the back line is always going to be the most vulnerable spot for a Guardiola team. Back in his Barca days, he famously said "without the ball, we are a disastrous team, a horrible team, so we need the ball."
The keys to beating a Guardiola team are, and have always been, transition opportunities and set pieces, and while opponents are generating only slightly more chances via set pieces, they are advancing the ball far more directly and effectively. Whether opponents have adjusted to take better advantage of this (both in terms of tactics and personnel) or whether City just isn't as good at transition defense, the table has shifted a bit out of their favor.
Let's look at some sequence data. Opta defines sequences as "passages of play which belong to one team and are ended by defensive actions, stoppages in play or a shot." (They differ from possessions in that possessions can make up a series of sequences.)
- Sequences ending in a shot: 6.2 in 2017-18, 6.3 in 2018-19, 7.3 in 2019-20 (to date)- Sequences ending in a shot and featuring fewer than 6 passes: 5.6 in 2017-18, 5.4 in 2018-19, 6.3 in 2019-20- Sequences featuring only one player: 1.9 in 2017-18, 1.6 in 2018-19, 2.1 in 2019-20- Average sequence time: 12.8 seconds in 2017-18, 12.4 in 2018-19, 11.7 in 2019-20
So what happens now? Is this a shift toward a new, evolved tactical landscape or a brief, confused interlude? For all we know, City buys a couple of A+++ defenders in the upcoming transfer window and never loses again. They still have all the money in the world, after all, and from an expected goals perspective, they have been a bit unlucky: their xG differential is an otherworldly +1.87 per match compared to their current goal differential of +1.56. The universe could right itself pretty quickly.
If nothing else, though, we may have caught a glimpse of what said landscape could resemble.
Is Liverpool the template? They benefit from talent most other clubs won't have: one of the greatest managers in recent history (Jurgen Klopp), a high payroll, and perfect-for-the-system players like tireless ball-recovery masters Sadio Mané and Roberto Firmino.
Is Leicester the template? Expected goals suggest their form could regress soon (their goal differential is +1.81 goals per match, but their xG differential is just +0.89), but Brendan Rodgers has crafted a nice balance of possession and generally combative play (lots of ball recoveries, lots of take-ons), with strong execution on set pieces.
Is the template overseas? In Spain, where possession play has reigned for a longer period of time and almost everyone is pretty proficient, more old-fashioned tactics like winning duels and aerials (and executing on set pieces) has begun to correlate more strongly with wins and losses again.
In conclusion, teams and leagues are always evolving. City took possession play to extremes and in response, the Premier League shifted into a more modern, possession-based league. Now, the top teams are again championing speed in transition to catch opponents out. Something will shift this status quo. We just don't know if it will happen first in England or elsewhere.