Science and the World Cup: Big data changing football

  As the World Cup heats up, researchers are using their expertise to help soccer coaches train players and formulate tactics.
  The leading international scientific weekly “Nature” published an article stating that at the World Cup in Qatar, players will use a “more evidence-based” way to show their coaches their value on the field. Within minutes of the game, tournament organizers send each player a detailed analysis of their performance on the field.
  This is the latest application of digital technology in football. Data analysis can provide guidance for everything on and off the field, from player transfers, training intensity, to formulating targeted tactics, giving real-time best passing lines on the field and so on.
  What the players are faced with is basically an astronaut-level data review. Wearable vests and wristbands sense motion, use GPS to track location, and record shots from each foot. Cameras set up from multiple angles can capture everything from the number of times the ball is held to the time it is held. To make sense of this information, most top teams employ data analysts, including mathematicians, data scientists and physicists poached from computing application giants such as Microsoft and CERN.
  In turn, insights from analytics experts are changing the way football is played. Strikers are now less likely to shoot from distance, wingers are more likely to pass to teammates than to cross, and managers are more obsessed with pressing high up the pitch – these tactical changes all have solid evidence to back up the coach’s own instincts.

  ”Big data has ushered in a new era in football,” says Daniel Memmert, a sports scientist at the German Sports Academy in Cologne. “It has changed the way teams think about tactics, how they behave, how they analyze their opponents, how they develop talent and discover players. ”
  Liverpool FC’s data team, which includes physicists formerly at CERN and Cambridge University, built a model specifically to assess whether a player’s actions on the pitch are more likely to lead to a goal. Sports scientists at the University of Lisbon, in collaboration with La Liga giants FC Barcelona, ​​published an analysis last year looking at how long different types of passes have on the pitch.
  Matthew Payne, the developer of the Euro 2020 prediction model and a doctoral student at Oxford University, said that the accuracy of using statistical methods to predict the outcome of the game exceeded many people’s expectations. For this World Cup in Qatar, Payne’s modeling shows Belgium as the most likely to lift the World Cup, followed by Brazil.
  ”The most useful thing we do is the pregame report,” Payne said. “We’ll look at the attributes of the players on the other team and draw up charts to show how they’re playing, how they’re moving. I’m going to Give some tactical suggestions or changes based on this.” Before the last game, Payne analyzed and found that the opponent’s left back’s header data was poor. He said: “So I suggested that the tall striker play on the right.” Oxford City won in the end. It’s something that seasoned scouts can see with their eyes. But Payne said, “Data is more objective than people.”