advertisement
advertisement

PAID CONTENT

hey2
PAID CONTENT

Beyond goals: Inside the Beautiful Game’s data-driven transformation

How new analytics-driven strategies are changing the way global football is being played

Beyond goals: Inside the Beautiful Game’s data-driven transformation

The use of data in football—well, soccer to some of us—has exploded in recent years. World Cup hopefuls are looking for new ways to apply data analysis to gain a competitive edge, and FIFA, the sport’s international governing body, employs a team of technical experts and data scientists to help teams improve performance and advance the game.

advertisement

IBM and Fast Company recently cohosted “Innovation on the Pitch,” a virtual conversation with FIFA insiders to learn more about the role data plays in modern football—and find useful parallels in the business world. Here are four takeaways from the discussion.

1. If you’re only watching the ball, you’re missing most of the action.

Traditional football statistics—such as goals, assists, steals, shots, blocked shots, and saves—tend to focus only on a player’s direct interactions with the ball. That’s leaving a lot of data on the table, according to Chris Loxston, group leader of football performance analysis and insights at FIFA. Modern data sets include the actions players take when they’re off the ball as well. And it turns out those actions can be important—even decisive.

“At the Club World Cup in Abu Dhabi, we awarded ‘Man of the Match’ to a center back who was on the ball for 21 seconds of a 97-minute game,” Loxston says. “If we had only looked at on-the-ball events, we couldn’t have told the full story of that player’s performance.”

advertisement

2. It takes human interpretation to make data useful.

Pascal Zuberbühler, senior technical expert for FIFA and a former professional goalkeeper and coach, noted that the amount of data players and coaches can use to improve performance and strategy has grown massively since his days as a pro player. But he stressed that it takes a player’s (or coach’s) perspective to put that information to use.

For example, Zuberbühler believes that goalkeepers should be actively supporting their teammates. So when he’s reviewing mountains of performance data, he zeroes in on how many times the goalkeeper offered to receive the ball and whether they were communicating with the rest of the team. The data can measure the interaction, but it takes human skill to take the next step and determine whether that data might hold critical and actionable insights.

3. The best data tells a story.

If Zuberbühler is dissatisfied with the level of interaction between the goalkeeper and the rest of the team, he takes the information to the goalkeeper coach to see how they’re coordinating with the head coach on player training. Often, he finds, they aren’t. With data in hand, Zuberbühler can connect the dots—telling the coaches a compelling story about how disconnection in their training habits can lead to disconnection on the pitch.

advertisement

The ability to use the data to tell a story is critical if you want the coach to listen, Loxston says. He recalled a situation where a team was struggling to build an attack that starts from its backfield. Loxston compiled data from several games to root out and identify the problem. When he met with the coach and the team, he led with video showing instances of when the team struggled and the times they got it right. Then he used the data to tell a logical story.

“When we weren’t having problems, it was because our fullbacks were five meters deeper. So we had an out on the side,” Loxston told them. “In the games where we were struggling, our fullbacks were too far over the shoulder of the opposition. They could easily intercept the passes we were trying to play out there.”

4. Data needs a common language.

If a global organization like FIFA is going to bring data to its operations, it needs to make sure the data is equally intelligible to football clubs at every tier and across all languages. That’s more difficult than it might seem. When FIFA hired five different data providers to record stats from eight matches at the Club World Cup, the only measures all the providers agreed upon were the number of kickoffs and the number of goals scored. Corners, red cards, and free kicks were all in dispute.

advertisement

FIFA decided it would create its own standard definitions—and collect its own data. Loxston, for example, described an “offer to receive” as clear, deliberate, and obvious indications from players to receive the ball either by pointing behind them, signaling in front of them, or changing the orientation of their bodies. Standard definitions like that create meaningful data sets that can be compared across the globe.

Zuberbühler also noted the need for data scientists and coaches to speak a common language. Which is to say, data scientists need to better understand the game of football—and coaches need to better understand data.

“I can deliver 8,000 data points to a goalkeeper in another football country,” Zuberbühler says. “But they may have no idea how to use it. That’s why we need a group like Chris’s—analysts who understand the game.”

advertisement
advertisement
advertisement

About the author

FastCo Works is Fast Company's branded content studio. Advertisers commission us to consult on projects, as well as to create content and video on their behalf.

More