The Evolution of Bayern Wingers
With Robben and Ribery both missing the most crucial games of last season and Bayern once again losing a Champions League semi-final, the decision makers at Bayern saw a need to invest in the wide areas of an already deep squad, to prepare for one last assault on the crown of European football under the management of Pep Guardiola. But while fans dreamt of star players like Eden Hazard or Ángel Di María, Bayern signed the Brazilian Douglas Costa for 35 million from Ukrainian Premier League side Shakhtar Donetsk, a team they beat the previous season in the Champions League to advance to the quarter-finals. While Costa arrived in time for the first training sessions, Bayern added French youngster Kingsley Coman from the Italian champions and Champions League finalists Juventus in late August. Surprisingly Coman was only signed on loan, with the contract tying him to Bayern until the summer of 2017, leaving an option for the German side to purchase him permanently afterwards.
Immediate impact
As Ribery and Robben missed most of the first half of the season, both Coman and Costa got a lot of playing time. Costa in particular carried the team offensively during that stretch and recorded twelve assists in his first twelve Bundesliga games for Bayern. The duo, soon christened “CoCo” by the German media, was feared by opposing defences because of their insane speed, and rapid changes of direction. However, Costa had to sit out the final four games in December and, while the Frenchman Coman kept playing, his form dipped. It soon became clear that he was still inexperienced, and some of his play seemed both immature and rash.
After the winter break, the Brazilian Costa seemed slow off the mark and Bayern’s game depended more on the recovered Robben and Ribery. Coman then had two more great games. Against an overwhelmed Bremen side he exploded into life, grabbing three assists, and just four days later his substitution was the much-needed spark for Bayern to overcome his former club Juventus in the first knockout-round of the Champions League. The youngster set up the goal that would eventually send the game into extra-time and then scored the final goal of the match.
As the season is over, and both new arrivals have had their ups and downs in their first year, it is now time to look at how differently they interpret their roles in comparison to Robben and Ribery. For this purpose, the two duos are compared with each other, with a younger version of “Rib&Rob” used, in this case from 2009/10 – the first and arguably best season of Arjen Robben in Munich. And let’s not forget about the ever-unpredictable Thomas Müller, who plays in various roles throughout each game, from time to time as a right-winger. To compare, an analysis of the two respective Bundesliga seasons.
Statistical comparison
First up, an overview of the most important statistics. The respective stats are either per 90 minutes or per game in order to have a fair basis for comparison. Of course Ribery’s and Robben’s numbers this season are less representative just because they played rather few games. It has to be mentioned that stats for the current season are from early April of 2016, and the last matchday which is covered is the game against Stuttgart. However, this should not alter the message.
Player | Shots / 90 | Assisted Shots / 90 | Assisted Shots per Shot | Key Passes / Game | Dribbles / Game |
---|---|---|---|---|---|
Costa | 3.0 | 3.4 | 1.1 | 1.8 | 3.3 |
Coman | 1.8 | 2.3 | 1.3 | 1.5 | 3.5 |
Ribery | 1.5 | 4.6 | 3.1 | 1.8 | 2.2 |
Robben | 4.3 | 2.9 | 0.7 | 1.8 | 1.8 |
Ribery 09-10 | 2.5 | n/a | n/a | 1.5 | 2.8 |
Robben 09-10 | 4.2 | n/a | n/a | 1.8 | 4.0 |
Müller | 3.6 | 3.2 | 0.9 | 2.0 | 1.2 |
If you look at the shots per 90 minutes you can see that Robben, despite not playing up-front, even tops Müller in shots in both seasons. This underlines his unmatched drive to score.
Unfortunately, data on assisted shots for the 2009/10 season was not available. If we take a look at that particular stat for this season, we can see that both Costa and Coman assist more shots than they take.
This is also reflected by the assisted shots per shot. These extra created shots are a huge boost for the genuine strikers like Müller and Lewandowski. Lewandowski has a league-leading 139 shots through Matchday 28 and is on his way to win the “Torjägerkanone” awarded to the player with the most goals in a single season, while Müller has beaten his previous best of 14 goals in a single season. Ribery has assisted an incredible 4.6 shots per 90 minutes so far this season, but, while this is still remarkable, those numbers are boosted by his limited minutes. It has to be pointed out that Costa, with his high numbers in both shots and assisted shots, is a constant threat to opposing defences and looks to score more often than not. However, this is also his weakness, as he often forces a shot when one or two smart passes would have created a better opportunity.
In key passes, as in passes that lead to a shot on goal by a teammate, there is almost no difference between the seven. Both new wingers are able to keep pace with the numbers set by the established players.
However, this picture changes for the better if you look at dribbling numbers. Both Costa and Coman attempt one dribble more a game than Ribery and Robben. Their numbers are on par with those of Robbery in the 2009/10 season. They both tend to feint inside, then use their speed to get around the outside of the defender and look for one of the strikers inside the box. As defences figured out to have a second defender covering on the outside, the success rate of these dribbles took a dip in the second half of the season. Furthermore, especially Costa’s crosses seemed to be ill-advised from time to time. This is an area where both players need to improve in the future. Müller, meanwhile, isn’t able to produce such a high dribbling rate. This is not only caused by his playing style but also because he spends a lot of time in a more central positon.
The correlation between crosses per 90 minutes and dribbles per game is depicted in the illustration above. All analysed Bayern players are highlighted by red dots, while another 20 players from the Bundesliga playing similar positions are highlighted by grey dots.
Two players have separated themselves from the rest – Costa and Coman, who are top-notch in both categories. They each attempt more than four crosses per game – the only other to do so is Filip Kostic – and they both register more than three dribbles a game – the only others to do so are Robben from his 2009/10 season and Karim Bellarabi. The huge number of crosses is quite surprising as Guardiola teams usually do not depend too heavily on crosses, but try to play their way through the centre of the pitch and inside the penalty box. You can legitimately draw the conclusion that the new wingers look for a cross after every dribble rather than a shot, as Robben does. This may be explained by the fact that Guardiola lines Costa up on the left side and Coman on the right, where it is more difficult for them to take a shot as their strong foot is on the outside. Ribery and Robben both play as inverted wingers, where they can take a shot after beating their opponent on the inside. Another graphic that displays this is shown below.
In this graphic the percentage of shots on target over the percentage of team shots taken by a single player are depicted. This stat shows how many shots one player takes for himself out of all the shots by his team during his time on the pitch. It has to be mentioned that, due to a lack of more specific data, an average distribution of team shots per game had to be assumed.
Robben from the 2009/10 season is just off the charts in both these categories. His numbers are even higher than those of Lewandowski for the current season, despite him taking close to five shots a game. Unsurprisingly the most memorable moments from that specific season are those typical Robben’s goals from outside of the box against Fiorentina in the Champions League knock-out round or inside the penalty area against Schalke in the DFB Cup semi-final. Robben, in both seasons, and Lewandowski are the only two Bayern players to take more than 20 percent of their team’s shots. The only non-Bayern player to do so this season is Andre Schürrle. While Ribery’s number can be seen once more as a statistical abnormality, and Costa is on level with not only Müller and 2009/10 Ribery, but basically all of the Bundesliga competition, Coman shows a severe lack of shots.
But it’s not only the lack of shots taken that is concerning. Less than ten percent of his shots are on target. This is one of the worst percentages of all players analysed. This is the area where he needs to improve a lot, as most recently seen in the final minutes of the Champions League semi-final versus Atlético Madrid. However, with him being 19 years old, this is still an excusable weakness, albeit one that he needs to constantly work on, as the Bayern attack will have to rely more heavily on him in the future.
Conclusion
Time for some final words. The stats show that both Costa and Coman can contribute to a team, but in a different way than Ribery and Robben did and still do. More crosses and fewer shots – these are the main conclusions to draw here. With all four wingers healthy, Bayern is definitely a tough side to prepare for, for every opponent, because each of them has their individual strengths. Robben with his shots and will to score, Ribery with lots of assisted shots and dribbles towards the centre of the field, Costa with electric dribbles out wide and fierce crosses when lined up on the left, but also nice shots from outside of the box when lined up on the right, Coman with blistering pace and enormous potential, and lastly Müller who’s everywhere at all times. Both Costa and Coman will, however, have to develop a lot before they can take the role of Robbery, especially on the European stage, something that was not covered by stats in this article.
Some days after reading an idea striked me again: you had Goalimpact data before in some analysis.
They (Goalimpact) gave (selected) data to publishers (for free?), you used them also. These data might be super interesting here. Any chance to get and add them?
btw: will a German translation follow? If not, a cross promotion would be great to get more audience and feedback over here. :)