Hint: It’s the goals. But that's not the whole story.
- By Daniel AltmanDaniel Altman is the owner of North Yard Analytics LLC, a sports data consulting firm, and an adjunct associate professor of economics at New York University’s Stern School of Business.
With just eight games left, the 2014 World Cup in Brazil has already been dubbed one of the best ever. Ignore the fact that the tournament has only existed in its current format since 1998; from the last-minute goals by Switzerland, Greece, and Argentina to the near misses by Algeria, Chile and the United States, this definitely has been an exciting World Cup. But why exactly has it been so enthralling, and can FIFA repeat the feat in 2018?
When you ask a typical fan what makes a sport entertaining, you’re likely to hear two answers: "close games" and "lots of scoring." In general, soccer has plenty of close games but not so much scoring. Only one of these has changed in this World Cup.
It’s not the close games. Say we call a game "close" if the score is tied or the two teams are within one goal of each other. If you look at first 56 games of World Cups since 1998 — that is, the group stages and the Rounds of 16 — then the closest tournament was 2010. The average game had just 12.2 minutes where the score wasn’t close, in a World Cup that was supposedly the most boring of all time. In 2014, the average for non-close minutes per game is 12.7.
There are more goals this year, however. The first 56 games of the 2010 World Cup averaged just 2.2 goals, as opposed to 2.8 in 2014. That extra goal every two games really could have affected results, especially since the score was close most of the time. But why were there so many goals in the first place?
It wasn’t the number of shots. According to FIFA, teams have taken 1,513 shots so far in Brazil. Four years ago, they had 1,810 shots in the entire tournament, of which 1,622 were in the first 56 games. Shots have actually fallen. But the shooting percentage in the World Cup has gone from 7.6 percent in 2010 to 10.2 percent in 2014 — a stunning increase.
How should we interpret these data? The fact that the number of shots dropped suggests that team defending in 2014 has been no worse than in 2010. Rather, it seems that the attacking players have simply performed better. The question is whether their higher shooting percentage has come from luck or skill.
This is a central issue in soccer analytics. One way to approach it is to ask whether the strikers were taking shots with a higher probability of scoring. In principle, better dribbling and passing sets up better opportunities to beat the goalkeeper; this is indicative of skill — and not just the striker’s skill — rather than luck.
Michael Caley has been tracking the likelihood of scoring for each shot during this World Cup. He throws out own goals and goals from penalty kicks. So far, the number of goals he would expect the teams to have scored, given the characteristics of their shots, is 143. That’s six more than the actual number of non-penalty non-own-goals so far. If anything, teams have been unlucky in their finishing. (Or they simply ran into a great goalkeeper; Michael says the difference between the statistics was just three goals before the United States met Belgium.)
So it’s not the number of shots that has contributed to the goal glut in Brazil; it’s their quality. Whether this change is down to the skill of attackers or defenders is a matter for debate and videotape. It’s also possible that shooters were especially unlucky in South Africa four years ago. Yet the fact that the number of shots did not rise hints at a slightly different conclusion.
When a team takes possession of the ball, it usually has to move the ball through the midfield before shooting. Midfielders can stop an attack before it starts; once the ball goes past them, the chances are much higher that a shot will occur. The quality of that shot may depend more on defenders — did they keep the attacking players as far away from the goal as possible, and did they cut down the most attractive shooting angles? Strikers may have shown exceptional dribbling skill, too, but the eye test suggests not many have.
World Cup teams have taken fewer shots than in 2010, so the midfielders may well be doing their jobs. But the quality of those shots is higher — indeed, much higher — so perhaps defenders are not. So if FIFA wants to make the 2018 World Cup in Russia just as exciting as this one, it only has to send a simple message to the national team coaches: Leave your best defenders at home.