I was hoping to find a Bundesliga-related request to fulfill this week, and sure enough, I’ve got one. From Kartsen: “Would love if you could find the time to write about Eintracht Frankfurt some time.” The time -- consider it found.
Last year, Eintracht Frankfurt was remarkable. Not necessarily because of how they played. No, they finished seventh in the league table, and the quality of the chances they created and conceded didn’t suggest they deserved much more than that. They lost in the Europa League semi-finals on penalties to Chelsea -- a nice run, but no more than that.
Rather, they were remarkable because, despite being an above-average German team whose deep cup run came in Europe’s second-tier competition, they somehow managed to put together one of the best attacking trios anywhere in the world. In terms of straight per-minute goal production -- non-penalty goals+assists -- Frankfurt had two of the top five players in the Bundesliga: Sebastien Haller in third (0.87 NPG+A/90) and Luka Jovic in fifth (0.84). (Minimum 1500 minutes played, data from Stats Perform unless otherwise noted.) Both contributed more goals per minute than the Bundesliga’s two undisputed superstars, Timo Werner and Robert Lewandowski. Meanwhile, Ante Rebic, he of dunking-on-Willy-Caballero fame, performed impressively in a tertiary role: 0.59, 17th in the league. The only club with a higher-flying trio were Borussia Dortmund, with Jadon Sancho in first, Marco Reus fourth, and Mario Gotze ninth. Some Harry Nilsson fan at Bundesliga HQ went so far as to refer to Frankfurt’s forwards as “The Magical Triangle”:
This year, Eintracht Frankfurt have been remarkable once again, but the table won’t provide any evidence of that. They’re currently in 12th, closer to the relegation zone than a qualification spot for the Europa League. But once you strip out the randomness inherent to kicking or heading a ball past a keeper who’s allowed to use his hands, well, Frankfurt have been as good as they were last year -- and they’ve done it without Haller, without Jovic, and without Rebic.
Last summer, the 21-year-old Jovic went to Real Madrid for $66 million, the 25-year-old Haller went to West Ham for $44 million, and the 25-year-old Rebic jumped over to Milan on a season-long loan. Jovic has had an awful year. There were rumors that his new manager, Zinedine Zidane, wanted to sell him before he played a single game for the club, and since then he’s only given the young Serb 399 La Liga minutes this season. Jovic was also criticized by Serbian politicians for returning home from Spain during quarantine. His response was special: “It's a shame others were not professional in their work and did not give me concrete instructions on how I should behave during isolation.” And then, somehow, Jovic also fractured his heel while training alone and will now be out for at least seven weeks. Not great!
Haller, meanwhile, is West Ham’s leading scorer, but that seems more like an accident of extended playing time on a bad team rather than a special achievement. He’s 47th in the league in NPG+A/90 (0.34), behind multiple West Ham teammates, including 31-year-old Scottish journeyman Robert Snodgrass. Rebic is having the opposite experience of his former teammate: six goals for Milan, but he’s only started six matches.
Back at Frankfurt, not a single player sits in the top 20 of the NPG+A/90 leaderboard. The club didn’t really reinvest in reinforcements for the departed trio: a couple small deals for the aging Bas Dost and a teenage Serbian striker named Dejan Joveljic (who was then sent on loan to Belgium), plus a loan for Portuguese striker Andre Silva from Milan. Dost and Silva have scored when they’ve played, but neither one has featured a ton. So, without the star threesome from last season and without any clear, consistent replacements, the team has managed to hum along at roughly the same level they were at last season. The attack has declined ever so slightly (1.73 expected goals per 90 to 1.66), but the defense has picked up the slack: 1.42 xG allowed, down from 1.56 the year before. They’re running cold on both ends of the field -- fewer goals than xG, more goals allowed than xG allowed -- but a relatively easy run-in (just one game of the final nine against a team in the top five) seems likely to help even things out.
So, uh, what the hell? Aren’t goal-scoring and then goal-assisting the two most important things a player can do in such a low-scoring sport? Yes, but all of the other stuff matters, too. It’s just a matter of how much. That’s the impetus behind the cool new project from the folks at American Soccer Analysis: a statistic called “goals added” that aims to calculate the value of every on-ball action that occurs in MLS. Here’s John Muller:
Soccer analytics has always had a problem between the boxes. Thanks to expected goals, we’ve gotten good at valuing shots, but shots won’t tell you much about the ninety-plus-minute scramble that produces just 26 total chances over the course of your average MLS game and maybe three goals if you’re lucky. Shots make up about three seconds of action for every four minutes of soccer. Grading the sport on that alone is like assigning GPA based on how well students walk across the graduation stage.
Jovic, Haller, and Rebic were all stars when it came to walking across the stage; their impressive goal contributions were all backed up by great shot-getting and chance-creating. Except, they might not have been all that great at all the other stuff. Over at the site Smarterscout -- created by Dan Altman, former senior advisor for football operations at Swansea City -- they rate players on how likely they are to make their team score and prevent their team from conceding. As their definition reads:
NYA uses two mathematical models to evaluate players' attacking and defending: a shot creation model and a ball progression model. Both models calculate expected goals (xG) generated and conceded by teams during games. Then algorithms divide the models' estimates of xG for (xGF) and xG against (xGA) into credits and demerits, respectively, for individual players.
These ratings (out of 100) are adjusted by position and standardized to the Premier League. Haller’s attacking performance last season rated as a 66, Jovic was a 57, and Rebic was a 41. In terms of all the things that come before the shot, Frankfurt had a different star last season. Filip Kostic led the team in FBRef’s shot-creating actions stat (“The two offensive actions directly leading to a shot, such as passes, dribbles and drawing fouls.”), and he was also the leader in progressive passes (“Completed passes that move the ball towards the opponent's goal at least 10 yards from its furthest point in the last six passes, or completed passes into the penalty area. Excludes passes from the defending 40% of the pitch) and the vertical distance he carried the ball with his feet. Kostic primarily features as a wingback, so he’s on a different scale than these other guys, but his Smarterscout attacking rating was 77 last year, and it’s up to 87 this season.
The unfortunately named Adi (Adolf) Hutter deserves some credit, too. He came to Frankfurt off of Red Bull Salzburg’s progressive-coaching conveyor belt. In 2017-18, Niko Kovac had the club playing a patient, relatively conservative style -- about 11 shots attempted and conceded per match, and about 1.26 xG both for and against per match. (Pretty weird that Bayern Munich hired him, huh?) Since Hutter arrived, the defense has weakened a bit, but the attack has improved by a much bigger margin. They cross the ball less frequently than they did under Kovac, but the biggest change is the pace of play. The average sequence (an interrupted possession) under Kovac was 7.2 seconds, while the average under Hutter has dropped down to 6.6. Kovac’s teams moved the ball toward the opponent's goal 1.69 meters per second, and that’s shot up to 1.99. With Kostic now the team’s centerpiece and with the departure of three guys who excelled around goal but not elsewhere, the team has slowed down a bit and inched toward a more controlled approach that seemingly values all the other stuff: they have more possession, the ball doesn’t move up the field quite as quickly, they win it higher up the field, they cross even less frequently, and they play more passes into the penalty area. The results don’t show it, but the performances suggest a team that’s adapted well to its changing personnel.
Given their standing in the table, I wouldn’t necessarily recommend any newcomers adopt Frankfurt as their Bundesliga team for the rest of the season. But if you’re looking for a long-term commitment, then they might not be a bad choice. If they ever effectively replace Haller, Jovic, or Rebic, then watch out.