With the World Cup now over, we look back, through the lens of data and advanced metrics, to review what’s happened on the pitch over the last month. We will analyse the winners, some individuals who deserve recognition, how all 32 teams compared and some of the surprises of the tournament (looking at you, Morocco).
Argentina’s victory over France was hard fought, but they were the more dominant team overall, especially during the first half. The Match Dominance chart below helps us visualise this. The chart is based on Expected Threat, which increases as the ball is moved into dangerous positions (from which a team is more likely to score a goal). Argentina had more of the ball than France (53% vs 47%) and they also had it in more threatening areas of the pitch. France were not helped by not having a shot until the 68th minute!
Although Messi was the star of the show, his supporting cast were crucial to Argentina’s victory. The scorer of the second goal, Angel Di Maria, caused France real problems as he attacked down the left flank. His team looked for him constantly for the 64 minutes he played, where he provided both a goal and an assist. In central midfield, Enzo Fernandez covered almost every blade of grass as he worked tirelessly with and without the ball. He kept the ball moving as he played and made more passes than anyone else on the pitch (88, 76 successful) and rarely played backwards (9 of 88).
Messi and Mbappé
This World Cup saw arguably the greatest player of all time, Lionel Messi, lift the trophy and claim the Golden Ball, whilst his PSG teammate Kylian Mbappé (no doubt a future great, if not already) went home with the Golden Boot.
They dominated the attacking stats for the tournament, having the most goals, shots, shots on target, assists, key passes and successful take-ons. As well as scoring themselves, they also sit in the top two places for Expected Threat (for the whole tournament), showing they consistently passed, carried and dribbled the ball into dangerous areas more than anyone else in Qatar.
The Golden Glove
Goalkeeper and Golden Glove winner Emi Martinez was no doubt crucial to the team’s shoot-out victories over the Netherlands and France (saving three of the nine he faced), but in open play he fared less well, conceding seven goals in open play versus an xGOT of 4.71. Penalty saving can be a big advantage in knockout competition, but perhaps Croatia’s Dominik Livakovic can feel unlucky to not have taken the award (see more later).
After a surprise loss to Saudi Arabia in their opener (where they were unlucky based on an xG against of 0.18 and an xG for of 2), the South Americans rallied, as shown by some of their tournament stats. They had more passes, key passes, shots and shots on target than anyone else and also generated the highest xG. They used the ball well, putting it in good positions to generate and take goal scoring opportunities.
They also worked hard off the ball, effectively pressing their opponents, but we’ll have more on that later.
With so many teams playing so many matches, it’s possible to start looking for trends in terms of playing styles and approaches. Sequence metrics are a good way to do this, since they consider how many passes and how quickly a team moves the ball forward, on average, per sequence. When we look at all the teams in the World Cup, we have Spain at one extreme (highest passes, slowest direct speed) and Iran and Australia at the other (fewest passes and highest direct speed). Of the quarter-finalists, six were above average for passes per sequence and below average for direct speed (we will include Portugal since they were only out in terms of direct speed by a tiny margin). The Netherlands were also comfortably above average for number of passes, which leaves Morocco as being the most unique of the final eight sides, as they played in a more direct, counter attacking style compared to the other seven.
We can also use team-level metrics to measure how effectively teams are pressing their opponents. By counting how many shots a team concedes on average and the number of passes a team allows their opponent to play per defensive action (PPDA), we have a proxy for pressure effectiveness. When we look at these, six of the quarter-finalists are above average, with the Netherlands and Croatia being below average. We can also see that winner Argentina allowed the second fewest average shots of any team. Costa Rica are a real outlier, with their 7-0 loss to Spain in their first group match (where they had 19% possession being a big contribution to this).
Although there is no set recipe for success, we can say that on average the teams that play with a more possession-based, higher pressing style of play tended to go further into the tournament than those who looked to move the ball in a more direct style and sat back.
Some other things worth mentioning…
Returning to Morocco, it could be said that they did well to reach the semi-finals, given they faced an xG against of 6.54 (3.72 xGOT) but only conceded twice. No doubt they were helped by some effective last ditch defending (three of their defenders are in the top five for tackles made) and solid shot stopping from their goalkeeper Bounou, but on other occasions they may not have gotten so deep into the tournament.
Third-placed Croatia arguably had the best midfield in the tournament (helping the team generate the second highest number of passes and most successful forward passes), but they also had an inspired man in goal. Dominik Livaković conceded six non-penalty goals from an xGOT against of 9.72, meaning he prevented almost four more goals going in than you’d expect.