Ryan O’HanlonMar 3, 2026, 06:53 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.”Follow on XMultiple Authors
Marcotti brands Arsenal’s win over Chelsea ‘garbage to watch’ (2:07)Gab Marcotti hopes the Premier League clamps down on “grabbing” from set-pieces after being bored by Arsenal’s win over Chelsea. (2:07)
Why it’s time for the Premier League to embrace the back three
The league offered nothing unique from a tactical or talent perspective. The best soccer was being played in Germany, Spain, and even Italy. The likes of Bayern Munich, Real Madrid, Barcelona, and Juventus were, inarguably, better than anyone in England. As if to prove the point, Leicester City went out and won the Premier League in 2016.
Everyone else was forced to adapt or die, and the next 10 years may have been the peak of English soccer: an era that married technical and physical skill with on-field results. The teams were great — and they were great to watch.
Through 28 weeks, Premier League teams have combined for 505 open-play goals — the fewest since the 2020-21 pandemic season. And if we remove that one season in the history of England’s top flight when there were no fans in the stands, then this is the lowest-scoring season from open play since 2009-10.
Teams have only put 1,659 open-play shots on target so far this season — by far the lowest in Opta’s 17-season dataset, and 300-plus shots fewer than in either of the past two seasons.
Prefer pretty passing to goalmouth action? Well, you’ve been bored to death this year, too: teams have completed 48,248 open-play passes in the attacking third. That’s the lowest since 2011-12 and nearly 10,000 fewer than we saw either last year or the year before.
Fixing this — and it needs to be fixed, unless you think soccer is the most popular sport in the world because “watching a lot of corner kicks” is our universal language — will require rule changes and new modes of on-field enforcement. But it will also require a coach or a club willing to do something that might break the sport free of its current stalemate.
To any prospective trailblazer out there, I have a suggestion: Do something no one else is doing right now and fully embrace the back three.
– Top 50 USMNT players of 2026, ranked by club form – Do Premier League summer transfers need more patience? – Premier League Power Rankings: Chelsea up, Liverpool down
“Um, your big idea is that more coaches and clubs should do what, uh, Ruben Amorim was doing before he got fired? Have you seen Manchester United’s record since they stopped playing with a back three?”
First off, Amorim had the team in sixth place when he got fired. Plus, I’d argue that Manchester United were the one team in the league that was actually playing risk-taking, wide-open soccer. Their matches featured a ton of shots — at both ends of the field. It’s not like he got fired because the team wasn’t doing well. He got fired because he was a pain to work with, as his pre-firing news conference made clear.
“Well, he was a pain to work with because he refused to play anything other than his stupid back three!”
That brings us to the main problem with quantifying the effect of playing with a back three: so many people have given it a bad reputation.
Back in 2022, soccer data analysts Pascal Bauer, Gabriel Anzer, and Laurie Shaw wrote a fantastic paper titled “Putting team formations in association football into context” for the Journal of Sports Analytics. Bauer works for the German FA, Anzer with RB Leipzig, and Shaw recently joined Liverpool after leaving City. These are three of the more accomplished and refined analysts in the sport, and it’s obvious in the paper.
Formation notations, of course, are largely meaningless. Games are dynamic, player movement is fluid and unpredictable. “No 4-4-2s are created equal” and all that. So, to define a team’s formation, the trio looked at how a team positioned their players in the buildup phase — once they’ve settled possession, the opposition has settled into its defensive shape, and the challenge becomes: “How do we move the ball up the field?”
By using tracking data from seven Bundesliga seasons, they identified that most teams tend to build up with either two or three defenders as the deepest line of players — the former indicating a back four, the latter indicating a back three. Most teams also defended the buildup phase with either a back four or a back three. They compared the success of the various formations against each other by looking at the average expected goals created in each matchup.
“The conclusion is that the three-defender build-up formation appears to be more easily countered than the two-defender formation while showing less of an upside benefit against other formations,” they wrote. “Building up with two defenders is significantly more popular amongst Bundesliga teams than building with three defenders; our results indicate that the latter does indeed appear to be a weaker option.”
The one caveat to their findings, the authors note, is that if there were a preference among stronger or weaker teams to favor a certain buildup structure. The paper was published in March 2023, and the back four was the formation of choice at Bayern Munich, who would win their 11th-straight title just a few months later.
And a year after that, the Bundesliga winner could go undefeated for the first time ever — except, this time, it wasn’t Bayern Munich. It was Xabi Alonso’s Bayer Leverkusen, and they played with a back three.
Who has been the biggest overachiever in the Champions League over the past five years? Inter Milan’s annual revenues usually make them the 13th or 14th richest team in the world, and they’ve made it to two of the past three finals in the European cup — something no other team in the world can say.
The most surprising Champions League winner of the past 10 years? Frankly, the only surprising Champions League winner of the past 10 years? That would be Chelsea in 2021 — the same season they fired Frank Lampard and finished in fourth place in the Premier League. When Thomas Tuchel was brought in to replace Lampard midseason, what did he do? He switched to a back three. Oh, and the last time Chelsea won the Premier League, in 2017? They played a back three.
When RB Leipzig made a run to the Champions League semifinals in 2020, they were playing a back three. Remember how Atalanta used to flirt with Serie A titles every season? For most of that time, they played a back three.
But it’s not only upstarts catching everyone else off guard. Juventus used to win Serie A every year and make deep runs in the Champions League. And while their recent decline has much more to do with blatant corruption and club mismanagement than the formation they play, they’ve also moved away from the back three that brought them so much success.
Here’s what their pass map looked like in the first half of the Champions League final against Inter Milan:
It’s been good enough to win the treble, make multiple Champions League finals, and go undefeated in the Bundesliga, but it’s still not good enough for anyone in the Premier League … yet.
Per Opta’s designations, here’s the frequency with which every formation has been used in the league since 2009:
Formations are fluid and not all built the same, so plenty of caveats apply, but it’s clear that the back four, in its three different guises — first the 4-4-2, then the 4-2-3-1, followed by the 4-3-3, and now back to the 4-2-3-1 — is king.
And, well, the results would seem to support these choices. Here’s the collective goal differentials of all of those formations:
But as the authors of the Bundesliga study asked, how much of that is because of the true efficacy of the back four and how much of that is simply because the best teams in the Premier League happen to play with a back four?
Now, I think we should give the likes of Guardiola and Klopp some credit — that they, and other top managers, favored back fours because it was a more effective base arrangement to build from. And I think it probably was. If you were able to pin the ball high up the field, circulate possession up, back, around, and through your opponent, and only have to occasionally snuff out counter-attacks, then it made sense for you to only have two nominal centerbacks on the field.
To be an utterly dominant team like Manchester City and Liverpool were — or Bayern Munich are — then the back four is probably the optimal arrangement. Why? Simply because it puts more attack-first players on the field and asks fewer players to cover the areas you’ll occasionally need to defend.
As such, these teams can’t be pressed off the field as easily as they were in the past. (Plus, the top teams are playing so many games that Klopp-style gegenpressing might be a physical impossibility anyway.) And when they do get pinned back, the quality of the players who are “parking the bus” is way higher than it used to be, as is the quality of the players who will launch the occasional counter-attack out of the deep block.
That all leads to where we are now: the ball rarely ends up near the goal, and the only way you can consistently score goals is from set pieces. Liverpool’s season was on the brink of falling apart, then they fired their set piece coach, scored seven set piece goals in a row, and now they’re three points back of third. This is just how it works right now — and it’s not fun.
To acquire one of these guys, you have to get really lucky and, say, happen to have a stadium in the same town that Trent Alexander-Arnold was born in, or you have to spend hundreds of millions of dollars in transfer fees just for the chance of acquiring one of them.
But what if you could play in a way where you didn’t have to chase any of those player types? Wouldn’t that be a massive advantage on the transfer market?
In, say, a base 3-5-2 system, none of those roles are really necessary. With three center backs, you’re not asking one or two guys to cover an entire half of the field by themselves. The wingbacks don’t have quite the same defensive responsibilities as top-level fullbacks. The midfielders need not be as rangy since there’s more defensive support behind them. And the attacking onus shifts more toward strikers and attacking midfielders, rather than wide forwards.