Who does he play like?
What do we mean when we compare players to each other? The weekend’s paid letter is about similarity scores, those sketchy algorithmic player comparisons you see all over the internet. I built a playstyle map to see if I could find stat-based stylistic comps that made sense, and here’s how it sorted attackers:
Does this mean anything? Is it useful? Could you tell your buddies that the players near each other are similar without getting laughed out of the groupchat? I’m not sure, but that Messi-Neymar-Grealish island looks promising, at least. I checked the results by comparing top MLS prospects to their famous European counterparts and wrote about the whole thing at length for paid subscribers.
Thanks to everyone who subscribes to keep space space space going.
Acceleration due to gravity
After weeks of delays and exhausting hype, on Sunday it finally happened: “Donda” dropped. Oh, also Lionel Messi played for PSG. Both debuts were disappointing, disjointed, baggy, improvisational, rhythmless. Kanye rhymed “Messi” with “messy,” which is about the level of artistry he’s operating at these days; PSG proved him right.
Listen, this team is obviously going to be very good. Messi’s a tough player to slot into a team and expect everything to click right away — just ask any Argentina coach ever. He’s a right winger who almost never plays right wing. He comes deep in the buildup, central in possession, and lately all the way over to the left halfspace whenever he feels like it. He uses his gravity as a weapon. He drags defenses around without even touching the ball. But the shifting structure creates puzzles for his teammates to solve, too, and putting it all together at speed takes practice.
If there was a winner when Messi came on for Neymar — and of course this could all work differently if we ever see PSG’s front three together — it was less often Kylian Mbappé than the right back, Achraf Hakimi. Even late in the game Reims wanted to press PSG’s freeform 4-3-3 from a 5-3-2 defensive shape. When Messi emptied out the right wing, it drew the left center back forward and made space for a teammate to run into:
It’s been a long time since Messi had a right back with wheels like Hakimi to hit that space on the run. He’s also never played with a striker like Mbappé, and it’s fun to watch them try to figure each other out in live time. The most promising moment came right at the very end, when Messi again used his positioning instead of his passing to try to put a teammate in behind:
Another forward might have headed up the wing when Idrissa Gueye turns on the ball with room to run, but Messi backpedals toward the sideline instead, opening his body like he’s looking to receive and quarterback the transition himself. If Gueye so much as cocks an eyebrow in Messi’s direction, the center back bites, leaving a lane wide open to put Mbappé in on goal. Instead he broadcasts his pass and gets picked off. You’ll get ‘em next time, champ.
Sarri plays Skyball
Maurizio Sarri’s 2017-18 Napoli was a famously possession-happy side, completing more passes that season than anyone but Manchester City. When they got a free kick between their defensive third and the halfway line, unless they were closing out a game, they did what short-passing teams do and restarted with a safe sideways tap:
Lazio last season under Simone Inzaghi were a far cry from a Sarri side, and they weren’t shy about looking for Ciro Immobile over the top, but they still didn’t airmail a lot of long free kicks into the final third. It’s just not a normal thing for good teams to do.
When Lazio hired Sarri this summer to replace Inzaghi, they bought a brand, a style, a promise that with a little time and dedication they, too, can play beautiful short-passing soccer. So you can forgive their opponents, Spezia, for being caught off guard by a 60-yard dead ball routine five minutes into Saturday’s game:
Where the hell did that come from? Who knows! But it reminded me of something I wrote not too long ago about Barnsley’s extremely un-Sarriball style in England’s second division:
Any other club would tap the ball sideways and get back to the business of building up. Not Barnsley. Almost every foul or offside flag outside the defensive third is an excuse for them to stack eight men in a tiny rectangle at the top of the opponent’s box and have a center back or goalkeeper lob an artillery shell into the mixer. Four Tykes charge into the box to stretch the defensive line; four stand ready for a second ball or a chance to counterpress. It’s such a simple, brilliant way to make a mess in the most dangerous part of the field that you sort of wonder why everyone else is doing free kicks wrong.
I do not believe that a Maurizio Sarri side is going to start playing like Barnsley. On the other hand, I don’t think it would be a bad idea if even the most idealistic short-passing teams — maybe especially those teams — occasionally used set pieces and a good pressing structure to play for field position rather than pure possession. Even the best soccer is still a dumb game with a bouncy ball, and if you cause enough chaos around the opponent’s box it just might bounce to you.
The German exchange rate
According to Ryan O’Hanlon, only four players in the top five leagues have averaged more than one non-penalty goal or assist every 90 minutes since 2018:
- Kylian Mbappé: 1.30
- Lionel Messi: 1.27
- Robert Lewandowski: 1.09
- Jadon Sancho: 1.02
So sliding Sancho into the Manchester United lineup should be an easy way to score more goals … right? Yeah, about that. Here’s how many non-penalty goals or assists Sancho is averaging in his first three Premier League games: 0.00.
He’s taken no shots. He’s played two key passes with an expected assist total of 0.1. In Sancho’s 119 minutes on the pitch, United has mustered 0.7 xG to their opponents’ 2.7. The most exciting Premier League signing since Paul Pogba is in the same lineup as him and they’re getting outplayed by Southampton and Wolves. What gives?
The obviously correct answer is that 119 minutes on a new team is nothing, even the best players have off days, United’s still figuring some things out, yada yada, settle down. But that’s boring. The just-plausible-enough, way more entertaining answer is that Sancho’s getting smacked by the Bundesliga tax.
Over the summer, Tony ElHabr and Aditya Kothari (who you may remember from the letter on measuring defensive space) got their hands on some VAEP data and tried to calculate how much an average player’s on-ball output changed when moving from one league to another. They found that a player transferring from Germany to England could expect to contribute 16-17% less to his team’s goal difference in the Premier League. That put the Bundesliga startlingly low in the league strength rankings, below Portugal:
Should Ed Woodward be worried? Nah, come on. Even if these quick Twitter estimates were right, 83% of Sancho’s insane production at Dortmund would put him comfortably among the best attackers in England, and he’s still just 21, younger than guys like Kai Havertz, Mason Mount, and Ferran Torres. You should feel extremely safe predicting that Jadon Sancho will score and assist a lot of goals in England. But it’s probably worth keeping in mind next time we compare stats across leagues that not all goals and assists are created equal.
In an effort to be a more sportswritery sportswriter, I’m going to try something I swore I’d never do at space space space: I’m going to write about Arsenal. No, I’m going to answer a mailbag question. Here’s FC Cincinnati data scientist Hayden Van Brewer:
Even with the Statsbomb podcast gone, I have James Yorke’s voice telling everyone to chill out with three-match takes on the Premier League. Still, how do you set expectations and measure progress in Arsenal’s shoes? What do you think Arsenal see as their path to rebuild success?
I’m sorry, I’m really trying to sportswriter this up, but the most exciting take I can muster on Arsenal is: they’re probably fine. They finished eighth in the Premier League last season, seventh by goal difference, sixth by expected goal difference. Their squad has the league’s sixth-highest Transfermarkt value, a nonsense number that’s still a decent proxy for the fact that they make more money than all but ten or so clubs in the world and invest a reasonable amount of that money in their roster. Lately they’ve even been spending it on young, potentially good players like Martin Ødegaard and Ben White instead of whatever washed-up Kia Joorabchian client is looking for a multimillion-dollar retirement fund. They’re not exactly a good team, but they play like good teams. Even after the catastrophic start, FiveThirtyEight projects Arsenal to finish tenth, and the bookies have them ninth to win the league at 200 to 1.
Like I said: probably fine.
But Arsenal fans are not fine with being fine, and they’re extremely not fine with losing 5-0 to Manchester City, even if three of those goals came against ten men, because they can remember a time when they were the best club in the world and City was cannon fodder instead of the other way around. I get that. I’m a Barcelona fan. Nostalgia’s a hell of a drug.
So what does progress look like? More of the same, honestly. Arsenal wrapped the first half of the 2020-21 Premier League campaign with 27 points and a +4 goal difference, then Mikel Arteta shook things up and they posted 34 and +12 the second time around. Kids like Bukayo Saka, Emile Smith Rowe, and Ødegaard settled into the lineup while the senior citizens wandered off into the sunset. Even Nicolas Pépé finished the season on a tear, with six goals in his last six games (on 1.7 xG, but still).
Yes, it sucks to play two of your first three games of the new season against last year's Champions League finalists. Yes, it’s painful to finish those games at the absolute bottom of the barrel while Tottenham’s perched at the top. But this is a problem of perception more than reality. Whether we’re ready to admit it or not, Arsenal’s a midtable club now, and midtable clubs have to take their whippings from the megarich just like everyone else. Say it with me: everything’s going to be fine. Probably.
Played on a spreadsheet
(Gonna shamelessly rip off Harper’s Index, one of my favorite magazine things.)
Number of goals by which Norwich City has been outscored through three games: 9
By which Arsenal has been outscored: 9
Aerials contested by Greuther Fürth striker Branimir Hrgota: 17
Aerials won: 0
Hrgota’s height, in centimeters: 187
Average minutes played by West Ham subs: 4
Expected goals per Rayo Vallecano shot: 0.02
Goals scored by Augsburg against Bundesliga opponents: 1
Against Augsburg’s own goal: 2
Source: FBref ❧
Sign up for space space space
You'll get free weekly letters about soccer (and never any spam). If you want to get more letters and read the full archive, you can become a paid subscriber any time.