Ian Graham does not make for an especially likely revolutionary. He has a distinctly academic air: genial, whip-smart, just a touch crumpled. He is not a natural salesman. He does not particularly enjoy giving interviews. Roughly once every 10 minutes, he allows a mischievous, outré sense of humor to get the better of him. He feels this makes appearing on any broadcast medium something of a risk.
It is hard to deny, though, that he is a resoundingly successful insurgent. Twenty years ago, he was among the first to explore the idea that soccer might be able to understand itself better by examining the vast reams of data produced by every single player in every single game. He did not so much pioneer the field of soccer analytics as help to conjure it into existence.
Then, over the course of a decade at Liverpool, he acted as proof of concept. From scratch, he built a data department that came to be regarded as one of the most sophisticated in the sport. His systems, his methods and his insights turned a club that had long been a drifting, fading giant into a beacon of innovation.
There are two ways to gauge his influence. The simplest is soccer’s default: the weighing of silver and gold. In his time at Liverpool, the club was crowned champion of England — for the first time in 30 years — Europe and the world. It made the Champions League final, the sport’s biggest game, three times in five seasons.
But a better measure, perhaps, is the wake he left rather than the trail he blazed. When he joined Liverpool in 2012, the fact that an elite team might employ an actual scientist — he holds a doctorate in polymer physics, but deploys his honorific only as a joke — was seen as either outlandish or absurd.
Soccer had long been resistant to outsiders, those who had not established their bona fides within the sport as players or coaches. Insiders regarded academics with particular contempt. The sport still viewed itself as too dynamic, too fluid, too poetic to be reduced to the mundanity of numbers. The idea of a data department was still something of a novelty in itself.
By the time Graham left Liverpool earlier this year, however, it was closer to a necessity. It is widely accepted that any club serious about competing in the continent’s major leagues should consult data when signing new players and assessing performances.
Almost every major team in Europe has a data department, increasingly including someone with a scientific background. Graham would be forgiven, perhaps, for thinking that the revolution he helped to instigate was complete. As far as he is concerned, though, it has barely begun.
There are, in Graham’s mind, two reasons that soccer is more complex than theoretical physics. The first is that “hard science” — his term — has the benefit of being bound by a set of unassailable rules. The laws of physics are nonnegotiable. Particles behave in predictable ways. That is not the case in soccer. “In physics, you do not have to take into account that gravity works slightly differently in Germany,” he said.
The second is that elite sports do not provide the “huge luxury” of controlled experimentation. European soccer does not operate in sterile laboratory conditions. There is no opportunity to formulate, test and modify a hypothesis. “It’s very emotional, very reactive,” Graham said. Fans and executives alike demand instant gratification.
The long-term future extends, at most, six weeks or so. To Christmas at the latest. The one thing nobody in soccer has, as a rule, is time.
He attributes much of his success at Liverpool to the fact that he did. This was, he said, the key ingredient in the “special sauce” the club developed. “The first thing I said to the owners was that they shouldn’t expect to hear from me for six months,” he said. “That’s how long it would take to build all the structures we needed. Every time there was something more pressing, we were able to hire someone else to do it.”
That few — if any — other teams have that privilege limits soccer’s ability to make the most of the great advances made in analytics in recent years. Even Brighton and Brentford, the two English clubs that now function as Liverpool’s heirs at the cutting edge, with their fairy-tale ascensions to the Premier League powered by data, must keep pace with a field evolving at breakneck speed.
“If you look at what people are doing outside the sport, people who have the time to try things out, it’s often a lot more advanced,” Graham said. “The tools available, the technology, the data are all a lot better now. If you were to start building a system today, you’d have a much higher baseline. Inside a club, you have to stop developing at a certain level. There’s so much day-to-day work that there’s no time for research.”
That is not the only limiting factor. Clubs operate in distinct silos: The work they do on data is largely proprietary. That teams should not share knowledge or disseminate best practices makes perfect sense on a sporting level. But not only is it antithetical on a scientific one, it serves to diminish the scale of data’s potential impact.
Teams that did not have the foresight to be early adopters are, Graham estimates, “10 years behind” the likes of Liverpool, Brighton and Brentford today. Those who had the appetite but not the resources are locked out, too. “The teams who could benefit the most from it often can’t afford to do it, or at least do it properly,” he said.
It has been almost a year, now, since the 45-year-old informed Liverpool that his role there had reached “a natural end.” Working for the club he had supported as a child was his “dream job,” he said, but he felt as if he had achieved all that he could. He knew that, at least in a professional environment, he would not be able to start from scratch again.
When the news of his impending departure got out, he quickly received a flurry of offers from other teams, all hoping he could do for them what he had done for Liverpool. Graham did not find the prospect appealing. The systems he had designed for Liverpool were now the club’s intellectual property; he did not particularly want to build something for someone else. “I felt like I’d done it,” he said. “It would have been crazy to work for just one club again.”
Instead, he set his sights on helping soccer as a whole to become just a little bit smarter.
Over the past couple of months, Graham has met with a succession of owners, and prospective owners, of soccer teams. They are — largely, though not exclusively — extremely wealthy Americans, often executives from private equity and venture capital firms, all of them keen to acquire the services of Ludonautics, the firm he established after leaving Liverpool, for the clubs they have bought or the clubs they hope to buy.
The appeal is obvious. In a sport chronically lacking in time, Ludonautics has the feel of a shortcut. Graham’s résumé is compelling. So, too, is that of Michael Edwards, the feted, publicity-averse sporting director who worked with him at Liverpool and who is now engaged by the company as a “sporting consultant.”
The pitch, though, is not that they can repeat the success they had at Liverpool; it is that they can expand upon it. Graham no longer has to work according to the strictures and demands of an individual team. He can, instead, use the full gamut of modern technology at his disposal to build something new, something better, and to drive the sport’s next great leap forward.
In time, he said, that may even allow him to attain what he regards as the “holy grail” of analytics: assessing the actual significance of a manager. “That’s very complicated,” he said. “It tends to be conflated with who has the best players, the best team. There are a lot of second-order effects. It’s very hard to know exactly how good any manager is, and what sort of impact they have on results.”
What has struck him most in his recent meetings is how little soccer still knows about itself. It is not just that complex things — how much of a team’s performance can be attributed to luck, how much it is spending for each point it has acquired — remain a mystery. The simplest building blocks often do, too.
Most pressing is that, in many cases, teams do not know what should be regarded as success. Ludonautics has seen sale prospectuses for teams in which the values of the squads are little more than finger-in-the-air estimates. That, Graham said, represents more than just a little sales sleight of hand; it has a tangible and detrimental effect.
“In terms of performance, they often do not have a systematic way of knowing who they are and where they are,” he said. “They do not have a sense of the underlying strength of the team. Without that, how do you know where you should be finishing? How do you know if coming fifth is good or bad? And how do you hold people accountable?”
As far as he is concerned, that is in the sport’s interest as a whole: The more teams that know the simple things as well as the complex ones, the better the sport becomes. “There’s a quote from John Keats about Isaac Newton using the prism to explain the colors of a rainbow,” Graham said. “But knowing why it happens doesn’t make a rainbow any less beautiful.”