How one team’s data operation paid off, and the player signed sight unseen: An excerpt from ‘Net Gains’
Alejandro Moreno explains why Brazil are the clear-cut favorites to win the World Cup in Qatar. (1:33)
The following is an edited excerpt from Ryan O’Hanlon’s forthcoming book, “Net Gains: Inside the Beautiful Game’s Analytics Revolution.” The book comes out tomorrow (Oct. 18), and you can order a copy here or purchase one from your local bookseller. This excerpt comes from a chapter about the ongoing efforts — and continued failures — to truly quantify the value of midfield play.
“It was something new and foreign to me,” Tim Sparv said. “I was bought because of stats and data.”
It would not be wrong to call Sparv the Sergio Busquets of Finland. He’s a year older and two inches taller than the Spaniard, but they played the same role at the base of midfield. At six foot four, he’s all gangly arms and legs, intense eyes, and an intimidatingly dense beard. He moved to England at age 16 to join the academy of Premier League club Southampton, where he was teammates with the likes of Champions League winner Gareth Bale and England internationals Theo Walcott and Adam Lallana.
– How Barca and Real Madrid became the same team (E+)
– How do we measure creativity in soccer? (E+)
– Why your favorite team won’t win the World Cup (E+)
Sparv never made an appearance with Southampton’s senior team, though he didn’t expect to: “I don’t think I was ever thinking that I was good enough to play for Southampton’s first team,” he said. He left in 2007 and has since played in Sweden, the Netherlands, Germany, Denmark, Greece, and Finland.
Where you might have seen him: playing for the Finnish national team. After all, you can’t miss him, the bearded beanstalk with the neon- colored armband. As the captain of Finland’s national team, Sparv did something no other Finn has done before: led the country to its first- ever major tournament.
At Euro 2020, however, the greatest moment in Finland’s footballing history — a 1-0 win in their first match — coincided with the near tragedy of Denmark’s Christian Eriksen collapsing on the field and going into cardiac arrest. After the match, Sparv, in particular, spoke thoughtfully, not just about the difficulty of finishing the game and his concern for Eriksen’s health, but about the shock that everyone else had experienced, too.
“You start thinking about your family, for me it’s my girlfriend and five-month-old daughter,” he told the Guardian. “I contacted them immediately after what happened to Eriksen and told them how we were feeling. Seeing it up close, the kids in the stadium, kids watching on TV, it can be a very traumatic experience. I hope anyone who needs it will get the help to deal with this.”
Seven years prior, Sparv had just joined Greuther Fürth, a small German club in northern Bavaria. Fürth, the town, was first mentioned in the year 1001, in a document written by Henry II, the 15th emperor of the Holy Roman Empire. The club was founded about 900 years later and won German championships in 1914, 1926, and 1929.
Then, the rise of Adolf Hitler, World War II, and the subsequent partitioning of the nation threw the German-soccer landscape into disarray. Whatever organizational structures and competitive advantages had been built up before the war were completely erased. Greuther Fürth didn’t make it back to the first division until 2012 — only to be relegated right back down the following season, which is when Sparv arrived.
In 2013-14, Greuther Fürth finished in third place in the German second division. This earned them a two-game playoff with the 16th-place team from the first division, Hamburg, for the final spot in the first division the following year. They drew the first match, 0-0. The second game was also a draw, but this one ended, 1-1, and the match was in Fürth. You know how this goes now: The tiebreaker was away goals, so the result kept Hamburg in the Bundesliga. A moderately successful season for Greuther Fürth had ended in failure, and no one outside of Nürnberg would care. At least, that’s what Sparv thought.
The folks at Midtjylland did care. Owned by professional sports bettor Matthew Benham, the Danish club had an in-house predictive model that thought that, based on their underlying performance, Greuther Fürth were better than just the third-best team in the German second division. No, it saw them as equivalent in quality to a lower- level Premier League side.
They, of course, couldn’t prove this, but there was enough cross-pollination between countries in the Champions League and in the Europa League that the model could make estimates the club was confident in. From there, they decided to take another leap. The club wanted to acquire a central midfielder, and they saw that Sparv played more minutes than any other central midfielder for this team they thought was much better than anyone realized.
– Stream on ESPN+: LaLiga, Bundesliga, MLS, more (U.S.)
“They were looking for a central midfielder,” Sparv said. “In their eyes, the league was undervalued compared to others, so they felt like, ‘OK, we can find some gems in this league.’ We were doing really well back then. I was playing and doing fine myself. They could see, ‘OK, this is someone we like,’ and also that when I was in the team, we were winning more than when I was not playing.”
***
Across every sport there exists a kind of positional spectrum-players who are inherently more valuable than others simply because of the position or role they’re able to play. In football, a quarterback is more valuable than a punter, and all of the other roles slot in somewhere between. While basketball has moved toward a more fluid positional approach, wing players — those who can create their own shots and create shots for others — hold more value than big men.
And in baseball, well, baseball obviously already has this all figured out. In WAR calculations, there’s a positional adjustment baked into the numbers: Catchers get the biggest boost, followed by shortstops. Second basemen, third basemen, and center fielders all get smaller positive bumps. And then, corner outfielders, first basemen, and designated hitters all get significantly dinged because of the comparative ease of occupying those roles.
Omar Chaudhuri, Chief Intelligence Officer at the consultancy Twenty First Group, wonders if the same kind of idea shouldn’t apply to soccer, too. It’s simple. Soccer is such a low-scoring game that goals — and the players who can score them and create them –should have outsize value compared to the other players on the field. Plus, execution in the final third requires a much higher degree of precision than anywhere else on the field.
This idea, however, goes against soccer lore. The best youth players end up playing central midfield because it’s the position where you touch the ball the most. It’s the thinking man’s favorite position — the one who pulls the strings. One theory suggests that midfielders make the best managers because they have to be aware of everything happening on the field, on a 360- degree spectrum.
The theory is certainly not disproven by the fact that Pep Guardiola, the most successful coach of the 21st century, was a midfielder. Midfielders tend to give the best quotes, too. “I’m always looking. All day, all day,” Xavi told the Guardian. “Here? No. There? No. People who haven’t played don’t always realize how hard that is. Space, space, space. It’s like being on the PlayStation. I think shit, the defender’s here, play it there. I see the space and pass. That’s what I do.”
As a former mediocre Division 1 midfielder myself, I’ve struggled to come to terms with this idea. I prided my game on my positioning, my ability to get out of pressure, receive passes away from the pressure, and make the pass before the pass before the pass that leads to a shot. There are few things I enjoy more than watching an elegant center mid, gliding through traffic and playing pass after pass, just a microsecond before the 21 other players even recognize its possibility. But when I shut off my romantic brain and try to dial into sporting efficiency, I’m left with the nagging possibility of this: Sure, it might be hard and it might look cool, but midfielders are secondary actors! They play too far away from the goal.
“By that logic, the better players should be playing in more dangerous areas,” Chaudhuri said. The market tends to agree with this idea. In addition to Paul Pogba, Barcelona‘s Frenkie de Jong and Real Madrid‘s Aurelien Tchouameni are the only other midfielders in the top 30 of the most expensive players of all time. And according to Chaudhuri, coaches verify the idea, too.
“Even if you just look at the way players are selected, it seems like coaches either consciously or subconsciously kind of agree with our hypothesis,” he said. “When attacking players move to better clubs, or get older, they tend to move backward on the field, not forward.” The best example of this is Victor Moses, who was a winger for a number of mid-table clubs and then became a starter for Chelsea when they won the Premier League in 2016-17 — only after he started playing as a wing defender. “He’ll maximize his earnings at Chelsea so playing a more defensive role makes more sense,” Chaudhuri said.
I suggested to Chaudhuri that perhaps it’s just a data issue. Expected goals and expected assists put numbers on just how much of an effect attackers are having on a team; could it be that we just haven’t figured out a good way to measure how midfielders drive winning? “I certainly think there’s things we aren’t quantifying in midfield,” he said, “but I’m not sure even if we could quantify them they’d make them more valuable than attacking players.”
***
Julien Laurens likens Amadou Onana to Paul Pogba and Aurelien Tchouameni ahead of his proposed move to Everton.
Rather than trying to solve the mystery of midfield or write it off as a position that wouldn’t move the needle, Midtjylland decided to simplify things. Greuther Fürth were a good team. Tim Sparv played a lot for Greuther Fürth. Therefore, Tim Sparv must be doing things that help make Greuther Fürth a good team.
On July 3, 2014 — the day before the quarterfinals of the World Cup in Brazil began — Midtjylland purchased Sparv from Greuther Fürth for $330,000. “Back then, statistics for me was like goals, assists, possession and stuff like this,” Sparv said. “So I was not sure what they were talking about. It’s a different, very inspiring approach.”
It quickly became clear that it was the right move, for both sides. In Sparv’s first season with the club, Midtjylland won their first-ever Danish title. In his second season, Sparv played every minute of the team’s eight Europa League matches, captaining Midtjylland in a famous 2-1 victory against Manchester United in February of 2016.
Across Sparv’s six seasons in Denmark, he played more than 10,000 minutes and appeared in 140 games. He was signed for little reason other than an assumption based on some readings spit out by an algorithm: Tim Sparv makes his team more likely to win. And the results are hard to argue against. With Tim Sparv on their team, Midtjylland won the league title in three out of six seasons.
Sparv, meanwhile, became fascinated with the way Midtjylland tried to measure everything. Almost everyone else in this book has entered soccer from the outside, beyond the confines of traditional thought; Sparv, though, became a convert from within. When a team signs you from out of nowhere and they tell you it’s because of stats, then you’re probably a little more likely to open your mind to whatever those stats have to say.
“In meetings, for example, we could see our expected-goals trend line, how we’d been performing for a period of time,” he said. “And you can see that it wasn’t only a coach’s subjective opinion. It wouldn’t be only the tape we were seeing from the previous game. We were also seeing statistics incorporated into that message. I thought that was really cool. For others, it didn’t really make a difference, but I thought it was a good way of making their message clearer. It was this objective factor and not only what we were seeing with our eyes.”
At Midtjylland, most of the data and technical jargon was hidden from the players. Deeper data and analyses were there for the players who wanted it, but only if they wanted it. Rather, the numbers just informed the guidance that was being given to the players. Most notably: Don’t shoot from there!
“At some point we started talking a lot about taking shots from good positions instead of shooting from range because it doesn’t statistically make sense,” he said. “So we talked for a period of time about playing a pass because if your teammate is in a better position, the chance of scoring goes up. We were shown this chart as well and we could see where goals were scored from. So at least players had a picture in their head when they went out to train.”
“And then maybe someone at the edge of the corner of the box is trying to take a shot and then he gets this image in his head, ‘No, oh yeah, this is stupid shooting from here. I’m actually going to look for a pass instead.’ At least it made me think when I saw it. Nowadays when I watch TV and somebody takes a shot from 35 meters, I always think to myself, ‘That is not how we score.’I love these long-range efforts that actually end up with a goal. There’s not a sweeter goal in the world, but it just doesn’t make sense.”
In the American sports vernacular, there exists this type of player who sacrifices his personal success for the greater good of the team: players who eschew individual stats in order to do the things that make their team more likely to win. At a loss for words, you’ll hear an announcer say, “This guy? He’s just a winner.” In a sense, this is an indictment of the things we’re counting.
If a player is better off doing things that we’re not counting — if other little actions and decisions have a bigger impact on the end result of a game — then we’re counting the wrong things. The people who are employing analytical thinking in the right way are searching for their own version of the same thing: a way to quantify a player’s true contribution to winning. When I asked Sparv how he tries to assess his performance, he gave the same kind of selfless answer you hear from so many athletes.
“I’m not so concerned about me having good stats after a game,” he said. “I think when I was younger, I was more selfish, more looking after my own performance, and even if my team won, and I had a bad game, I was disappointed. Now my mood is the same as the team’s mood: If we win, I’m happy. If we lose, I’m sad. So I don’t look so much at my personal stats actually. It’s more the team stats and how we’re progressing as a team.”
Of course, this was the same reason he was signed by the most statistically fluent soccer club on the planet. If he’s playing a lot of minutes and the team as a whole is playing well, then he’s probably playing well, too, whether he realizes it or not. It’s all still very theoretical, though. Might we ever figure out a way to isolate the things that Sparv does that make his teams better? He, at least, has an idea: “Maybe in the future we’ll somehow be able to measure a person’s organizational skills, or someone’s communication skills”.
Excerpted from NET GAINS: Inside the Beautiful Game’s Analytics Revolution by Ryan O’Hanlon. Copyright © 2022 by Ryan O’Hanlon. Published and reprinted by permission of Abrams Press, an imprint of ABRAMS. All rights reserved.