With 48 games now being played, let’s use some of Bepro’s Advanced Metrics (and event data) to find some interesting and surprising stories from the World Cup group stages.
Some Big Teams Miss Out
Belgium’s squad offered a lot of experience, but only offered one goal, the equal lowest of any side at the tournament. They had a lot of shots (35), with nine hitting the target. Their total accumulated Expected Goals (xG) value of 4.742 shows that you would expect them to score more based on the quality of chances they had, however with a significantly lower Expected Goals on Target value of 3.239 (this measures the probability of shots on target resulting in a goal), they were clearly struggling to find the back of the net.
Unfortunately for Lukaku, one of modern football’s most prolific scorers, he had a tournament to forget. He was unfortunate to start the World Cup sidelined with injury, but Belgium’s three biggest chances of the group stages all fell to him (total xG 1.711), but he was unable to score any of them.
Germany came close to reaching the knockout stages, but perhaps can see their inability to take advantage of their relative dominance during their opening game to Japan. This contributed to them losing the match and Japan leaving with three crucial points.
They controlled the first half, as shown by the Match Dominance chart below. This is based on Expected Threat which measures the probability that having the ball at a point on the pitch will lead to a goal. Despite their dominance they were unable to score from open play (their goal came from a penalty converted by Ilkay Gündoğan. They generated goal scoring chances (with an open play xG of 2.273), but were thwarted by a solid goalkeeping performance from Gonda and ineffective finishing.
Over the three games they played, Germany generated 8.59 xG and 5.48 xT showing that they were creating goalscoring opportunities, but the final results did not go their way.
An Unlikely Survivor
Australia are the only team to qualify having conceded more than they’ve scored (three for, four against). They managed to score those goals from just 1.175xG and you’d expect they will need to keep that form going if they want to keep progressing.
Interestingly they are the most direct team left in the tournament, when considering their sequence data. With 2.86 passes per sequence and a direct speed of 11.85 km/h they are out on their own, as shown by the below chart.
A Theme for Progress?
Looking at those that qualified in the above chart, we can say that those teams that qualified had a lower direct speed and a higher number of passes per sequence on average, compared to those that did not make the last 16.
If we see the same chart for those that didn’t make it, they are on average 0.7 km/h higher in terms of direct speed and played 0.73 fewer passes per sequence. There’s also more spread in terms of style for these teams, from Belgium's high number of average passes per sequence, to Germany's slower direct speed.
An Expert Chance Creator
Returning to Australia, they will come up against Argentina, who have the individual player who has generated the highest Expected Threat (xT) during the group stages in the form of Angel Di Maria. He’s generated 0.86 xT, demonstrating his ability to still create and put the ball in dangerous areas. His xT map for the World Cup group stages shows how his ability to cut in from the right, using his dominant left foot, is causing opponents real problems.
Press, Press, Press
Perhaps unsurprisingly Spain have the fewest passes per defensive action (PPDA), which tells us that they are the team pressing the most effectively, in that they are limiting their opposition to a low number of passes per sequence (6.02). Interestingly there is not a great difference between the PPDA of the teams who qualified and did not when you remove Costa Rica, who really are an anomaly! It will be interesting to see if this does offer any insight as the number of teams in the competition continues to reduce.
Korea Leading the Line
South Korea had the highest average starting sequence line, showing that they started their sequences higher up the pitch than any other.
There was a slight difference between those who did and did not qualify for the knockout stages, with those who did qualify starting their sequences on average 1.9 metres higher up the pitch.
Interested in having access to this type of data for your team? Get in touch with us here!