Rory Smith reports in the New York Times, image by John Sibley, Reuters:
Ability is no longer the only currency in the transfer market. Information, and the skill to interpret it, is just as important now, too. Manchester City appointed Laurie Shaw, an academic with a Ph.D. in computational astrophysics who previously advised the British government. Models do not just allow (them) to assess players or performances. It lets them analyze how a coach played. It predicted young players’ development, based on historical parallels. It helped discern whether a player was shining because he was a part of a good team, or because he had some special talent.Markus Krösche’s first summer in his new job was a frenetic one. Within a couple of months of taking up his post as Eintracht Frankfurt’s sporting director, he found himself with not only a manager to replace but a couple of star players, too.
Krösche, 41, hired from RB Leipzig, got to work. He hired Oliver Glasner as coach. He completed a complicated deal to sign Rafael Santos Borré, a Colombian forward, from the Argentine club River Plate, and acquired two young wingers to complement him. In total, he brought in 11 players that summer, and sold or loaned out a dozen more.
The acquisition that may prove his most significant, though, passed by almost unnoticed. Quietly, Krösche returned to his former club to hire Bastian Quentmeier, a bookish former hockey player with unruly hair and a fisherman’s beard, as his new club’s head of data analysis.
Even in the relatively small world of German soccer, few had heard of Quentmeier. Even fewer knew exactly what his role in Leipzig — data scout — actually involved. Those who did, though, regarded him highly. “He has something unique,” said Ralf Rangnick, the all-purpose visionary and current Manchester United manager who had approved Quentmeier’s hiring at Leipzig. “Really good, and unique.
The appointment did not generate headlines. Quentmeier’s arrival was so low-key, in fact, that Eintracht did not even see the need to confirm it on the club’s website. There was no announcement beyond a subtle update to Quentmeier’s personal LinkedIn profile.
That modesty belied its importance. Krösche had pulled off a coup — strengthening his club’s hand while weakening a rival’s — in an arms race so new that it is still just taking shape. He had hired Quentmeier not merely for his skills but for something of increasing value to clubs across Europe: his knowledge.
Sea Change
Perhaps the best way to gauge the speed with which soccer has embraced data is to compare Quentmeier’s circumstances in his new job with those in his previous post. At Eintracht, he is in charge of a team of three analysts: another, full-time staff member, plus two students in support roles. There is nothing unusual about that.
When he arrived at Leipzig, in 2016, it was a little different. He had joined the club, initially on a part-time basis, after bumping into Johannes Spors, its chief scout, at a conference in Munich. At the time, Quentmeier was working for a subsidiary of Scout7, a company that provides data and video footage to clubs.
“We had games from all these leagues around the world,” he said. “We had to pay for all of them, but we found that there were some leagues that the clubs, our clients, were not interested in.” His job was to track which leagues were being watched by teams that had signed up for the service, and which were not.
Quentmeier thought nothing of the meeting until, a few months later, Spors got back in touch. Despite its corporate backing, Leipzig had always cultivated a deliberate start-up energy, and Spors was interested in finding out how best to use data to help with signing players. He asked Quentmeier if he would take on the “mini-job” of advising the club which data providers might be most useful.
The pay was hardly lavish — a few hundred euros a month, Quentmeier said — but the trial was successful. A few months later, Spors and Rangnick asked if he would like to join the club permanently.
Officially, he would be a data scout — one of only a couple employed in Germany at the time — but that did not capture the full extent of his role. Quentmeier was not joining an established staff to be trained. He was in charge of a department of one. “There was nothing, really,” he said. His job was, in effect, to find out what the job of a data scout would be.
He spent the first few months trawling through the various data providers, working out which ones gave him the best quality of information. He spoke to Opta, Wyscout and InStat, three well-known providers, and then branched out, outside soccer, picking the brains of anyone he could think of who worked in data analysis.
Mostly, though, he tried to work out the sorts of questions a soccer team’s data system needed to answer. “Every coach and every sporting director and scout has their own idea,” he said. He knew his model needed to be flexible enough to adapt to individual tastes. It was not enough just to compare defenders, for example. “It had to distinguish between someone who could play the ball and someone who was more a warrior,” he said.
Designing and building the system occupied most of his first year. “It was easier to do it on my own, from the start,” he said, rather than simply buying an external system and trying to tweak it to suit Leipzig’s needs.
The model Quentmeier built did not just allow him to assess players or performances. It let him, and his superiors, analyze how a coach played. It predicted young players’ development, based on historical parallels. It helped him discern whether a player was shining because he was a part of a good team, or because he had some special talent.
Most of all, it gave RB Leipzig another edge. “Leipzig spent a lot of time and money to be at the head of the curve,” Rangnick said in a telephone interview last year. Those sorts of things do not stay secret for long in soccer. Quentmeier believes that when he started, only a couple of other teams in Germany were investing in data.
Now, he said, it is “normal” for clubs to have a team of data analysts. That means it is normal, too, for teams to do all they can to make sure they have the best data analysts. That can mean looking outside the sport for expertise. Or, increasingly, it can mean taking the approach that brought Quentmeier to Eintracht, and plucking someone from a direct rival.
Shopping Lists
Like Quentmeier, a vast majority of data scientists — even at the game’s most decorated clubs — remain essentially anonymous. Only occasionally, when a team makes a particularly significant or an especially unusual appointment, do their names drift to the surface.
Manchester United’s hiring of Dominic Jordan, in October, as its first director of data science was greeted as a major step forward for a team hidebound by conservatism. Last year, Manchester City’s appointment of Laurie Shaw, an academic with a Ph.D. in computational astrophysics who had previously advised the British government, seemed sufficiently exotic to attract attention.
The picture inside the sport, though, is different. “There is much more knowledge of smart people, people doing good work, people making waves at other clubs,” said Omar Chaudhuri, the chief intelligence officer at the data-led consultancy Twenty First Group. “Executives will know them by name. They are much more likely to have them on their shopping lists.”
These are not, in most cases, easy appointments to make. Krösche knew Quentmeier from their time at Leipzig; he could vouch for his work firsthand. Not everyone has that benefit. Clubs are reticent to share knowledge and information that they consider proprietary. Few, if any, are prepared to make public the work performed by their data departments. That makes establishing the credentials of any individual member of a staff extremely difficult.
“Sometimes, proof of their success is enough to convince people,” said Chaudhuri, noting that clubs perceived to be doing well will find their staff in demand from others, eager to acquire a little of the magic. Even then, it can be hard to know exactly where credit should go.
“Executives who aren’t experts in data do not necessarily know what good work looks like,” he said. “Sophistication of analysis and sophistication of presentation aren’t always the same thing.”
In some cases, that has led clubs to the likes of Twenty First Group and Nolan Partners, a headhunting firm based in London that specializes in sports, to establish who is doing what and who is doing it well.
“We have provided a picture of what exists in that space for a few teams,” said Stewart King, Nolan Partners’ lead for Europe. Twenty First Group has been commissioned to run recruitment processes, too, sitting in on panels and devising practical tests for potential candidates, Chaudhuri said.
Often, there is an emphasis on communication: Quentmeier has found that analysts and the models they use have to be able to predict, and answer, the sorts of questions coaches and scouts are likely to ask. That, Chaudhuri said, is the thing clubs are looking for above anything else.
Both expect these sorts of acquisitions to become more common over the next couple of years, as clubs scramble to keep pace with rivals or forge ahead. Ability is no longer the only currency in the transfer market. Information, and the skill to interpret it, is just as important now, too.
0 comments:
Post a Comment