How Europe Does Progressive Passing

And what it tells us about the Champions League knockouts.

Once upon a time, in the dark ages of soccer stats, people counted passes. We knew even back then it wasn’t a super useful hobby because there are a lot of different kinds of passes, some harder than others, and nobody outside Catalonia’s going to get all hot and bothered if a third-string Barcelona defensive mid completes a million five-yard dumps to a center back. So people came up with a category of passes that seemed harder and more valuable than others and started counting those as their own thing. We called these special passes “progressive.”

The dirty secret of progressive passes is nobody really knows what they are. I mean yeah, the idea is they make some kind of progress toward the attacking end, but how and by how much depends on where you’re getting your numbers from. American Soccer Analysis uses a simple definition—any pass that moves the ball 25% of the remaining distance to goal—that I like because it’s clean, intuitive, and scales down as the ball moves upfield and the defense tightens up. A progressive pass could be a longball from the keeper, a linebreaking pass in midfield, or a short cutback to the penalty spot. Anywhere on the field, at any angle, a pass is progressive if it gets you a quarter of the way from where the passer is to where the team’s ultimately trying to get to, dead center of the goalline.

The other week Eliot McKinley tweeted a viz that separated ASA’s progressive passes into 40 types, as determined by a k-means algorithm that he and Cheuk Hei Ho have used over the years to group similar-looking passes into clusters:

Eliot’s thing was cool because it shows how ASA’s goals added model expects each type of progressive pass to impact the scoreline. (Notice that the model thinks even incomplete longballs out of the back are probably a good thing on average.) We don’t have g+ for leagues outside the United States, but there’s a lot of other interesting stuff you can do with these progressive pass types for the top European leagues. You might start, for instance, by looking at who does what:

No surprise that good teams dominate the pass types that end closest to the opponent’s goal while bad teams do a lot of the ones that start around their own box. What’s fun is picking out particular clubs’ signatures, like Liverpool’s unique diagonals from Trent Alexander-Arnold’s flank to Sadio Mané and Andrew Robertson on the opposite wing, or Manchester United playing a lot of short balls around the top of the box where Bruno Fernandes lurks. Barcelona’s thousand little daggers into the box are nothing new, but their dominance in the left halfspace is: Lionel Messi’s made a point of popping up over there to attack from new angles this season, and Pedri provides an attacking midfield threat they haven’t had since Andrés Iniesta shipped off for the Bill-Murray-advertising-Suntory-whiskey stage of his career. As for Atalanta, Gian Piero Gasperini’s attack may have slipped a gear or two since the Papu-Iličić glory days but it still tops four different progressive pass clusters in the final third, more than any other club.

Or maybe teams aren’t specific enough for you. Maybe you’d rather see top individual passers:

Jack Grealish has been undressing England with square balls from the left side of the penalty area. Jesús Navas, quarterbacking a Champions League attack from a wingback spot at a sprightly 35 years old, dominates two different progressive clusters on the right wing. Mats Hummels breaks lines up the middle for Dortmund, while Jadon Sancho spent most of the season splitting defenses from the right corner of the box. Fred and Granit Xhaka like to sneak balls from midfield into the heart of the defense. Toni Kroos loves his big switches. But the standout name on this viz has got to be Borussia Mönchengladbach’s underrated center back Matthias Ginter, who’s completed more than anyone else in Europe of three different kinds of progressive passes up and down the right side. Guy’s a one-man buildup.

But birds-eye data surveys of the top five leagues are so last week. This week you want to read about ...

The Chaaampions

A quick glance at progressive passing among the clubs who’ll play Champions League knockouts this month (sans Porto, sinto muito) will tell you that even though progressive passes are generally a very good thing, how much a club does them is as much a stylistic choice as a measure of quality. If you remember our Seven Styles of Soccer, you’ll notice that teams in the Aggressive group do more progressive passing than the Possession style. Man City has the third-highest offensive rating out of the fifteen clubs in FiveThirtyEight’s model but ranks dead last for progressive passes per game; Lazio averages less expected goals per game than Freiburg, per FBref, but does more progressive passing than Barcelona and Juventus. Same story with completion rates, which are more suggestive of which kinds of progressive passes teams attempt than how good they are at attacking. Liverpool doesn’t really care if TAA’s diagonals hurt their possession stats—they care about risk and reward.

Over in the player columns, you’ll notice the progressive passing leaders are heavy on fullbacks and wingbacks, who are active in every phase from the buildup to the final pass, while the leading receivers are mostly strikers, who are always an option ahead of the ball. Only Neymar leads his team in both progressive passing and receiving, although Messi comes close (wide creators are great if you can get ‘em). As you’d probably expect, the midfielders in charge of breaking lines for their team are Bayern’s Joshua Kimmich, Real Madrid’s Toni Kroos, and City’s Kevin De Bruyne (pretty impressive considering he’s really been more of a false nine this season, with fewer bodies in front of him). Simone Inzaghi probably shouldn’t brag about the fact that a 38-year-old goalkeeper is Lazio’s most frequent progressive passer, but the manager’s love of longballs could be an interesting test for Bayern’s increasingly vulnerable high line.

If you like colorful pitch maps more than tables and decimals, I’ve got something for you. The viz below compares Champions League teams to each other across the 40 progressive pass types and highlights which kinds a team does more often than everyone else. Each color represents one cluster of passes, with several examples of each cluster plotted on the pitch to give you a better feel for what kind of passes it includes.

Again, this is about style, not the quality or even quantity of a team’s progressive passing. A club that lights up in a lot of different clusters probably has a playstyle that’s a little unusual among the elites (looking at you, Lazio and Sevilla), while a club like Chelsea can play an average number of progressive passes but barely show up on the plot because they don’t favor any particular type. Frank Lampard’s managerial tenure wasn’t famous for its style.

If you’ve spent enough time watching these teams, you’ll probably recognize some patterns here that’ll be worth watching in this week’s Round of 16 games. Liverpool-Leipzig will be a battle for the flanks, as Jürgen Klopp’s team runs through its star right back and Julian Nagelsmann’s side favors incisive passes from Marcel Halstenberg and Angeliño’s left side into the center of the attacking half. PSG probably won’t aim quite as much for the left wing without the injured Neymar sharing space with Kylian Mbappé, though with Ángel Di María out too it’s anyone’s guess how Mauricio Pochettino will attack Barcelona. Sevilla-Dortmund should be fun for pitting strength against strength: Julen Lopetegui’s team basically only has one idea—get Navas up the wing—that might not pair well with Dortmund’s direct balls in the other direction from Raphaël Guerreiro to Sancho, who’s recently moved over to the left.

We can also run the same analysis the other direction, highlighting pass types that each team does a lot less of than other Champions League sides:

Newsflash, Barcelona doesn’t pass long. Gladbach is comically one-sided, building short up the right through Ginter and eschewing the left wing (probably good news for City, who’ve been playing without a right back). Atlético Madrid likes to attack quickly through its wingbacks, Kieran Trippier and Yannick Carrasco, but unlike Liverpool they don’t look for early crosses into the box. Sevilla hasn’t been big on breaking lines in the middle, at least not before Papu Gómez’s recent arrival from Atalanta to diversify their attack. Once again, Chelsea stand out as styleless wonders, though Thomas Tuchel’s doing his best to change that.

The other team that’s a curious blank on the progressive pass maps is Pep Guardiola’s Man City, who’ve decided they’d just prefer to build slow. “Passing 50 meters has a greater risk to lose the ball than 5 to 10 meters,” Guardiola explained on Monday Night Football a few years ago, “and when you lose the ball passing 10 meters, the transition is always shorter. It’s easier to get the ball back.” Progressive passes are good, but so is counterpressing, and sometimes those priorities compete. There’s more than one way to break down a defense. Which I guess is as good a reminder as any that, though we’ve made progress on passing analytics since the dark ages, there’s always a little further to go. ❧

Thanks for reading space space space! This letter's free, so you're welcome to share it. Please consider becoming a paid member to get more letters like last week's on Man City's asymmetrical tactics and one on the Champions League coming later this week.

Further reading:

• Cheuk Hei Ho and Eliot McKinley, Using K-Means to Learn What Soccer Passing Tells Us About Playing Styles (American Soccer Analysis)

Image: Bridget Riley, Untitled (Wave)