Olympique de Marseille’s Cross Analysis Under Marcelo Bielsa
SAMIRA KUMAR TACTICAL ANALYST
The 2014-15 season of Olympique de Marseille, under Marcelo Bielsa, was considered for this case study. Crosses (including cut-backs) that led to either of the following were considered:
- Shots (on/off target),
- Blocks from the defender after attempt at goal
- Saves from the keeper
- Goal from the cross.
Note that all the data that was used for this research was plotted and calculated manually, close to 100% accuracy.
Cross Location Zones:
The pitch is split into 9 different zones and crosses from zone 13/15/16/18 (Figure 1) were only evaluated, taking into consideration of the fact that a cross from either one of these zones would ideally stretch the opposition defenders (during build up) and would create a higher chance of scoring from the cross.
Figure 2 shows the location of all the crosses that were analyzed for this study. For all the crosses that were considered, the main criterion was that the opponents weren’t settled in their defensive shape (in defensive transition) and the outcome of the crosses were in either one of the following category:
- Goal from a shot or header.
- Shot on (Goal Keeper saved) or off target.
- Attempt blocked by the defender after the attacker has an attempt at goal.
Rare cases, for crosses whose outcome didn’t fall into any one of the above categories (cross missed the target, cross blocked before it reached the target man…etc), were still considered if the build up to the cross or the position of defenders and attackers had high significance.
From this study it was evident that crosses that were close to the penalty box (compact) were less successful than the ones from close to the touchline (stretched). The shape of the opposition defensive line and the space in between them were main factors that led to unsuccessful outcomes when crossing from near the penalty box
So how does the location of the cross have a partial/significant influence on the outcome?
In the 2 crosses that are highlighted below, one of the crosses is from near the throw line while the other is from the edge of the box. Both the crosses were taken when the opposition were in transition. The major difference is that Marseille preferred to use the wide areas in build-up for the first cross (Figure 3) and as a result, the opposition left back is stretched. This forces the LCB to provide cover for the LB and the whole defense gets stretched and also lesser number of opposition midfield players inside the box, as one midfielder goes wide to support the left back, to close down the space for the cross. This situation, inside the box, would isolate a defender 1v1 or 1v2 with the attacker, thereby increasing the chances of getting a successful outcome from the cross. In this case, the final attempt from the attacker is blocked by the defender.
In the other case (Figure 4), the cross is from the edge of the box and all the 4 defenders are much compact and closer to each other inside the box, with support from their midfielders, who are closer to them now, to win the second ball.
In a similar study of crosses from Bayern Munich’s 2015-16 season (till winter break), their large % of goal from crosses have come from crosses farther away from penalty box, with exceptions of goals from cut-backs from near the penalty box (Figure 5).
The distance between the oppostion defenders were higher for the crosses that led to a goal for Bayern Munich. Data collected for distances between defenders show that the wider the opponent’s defensive line is stretched (horizontally from the two throw-in lines), the higher the chances of creating a high quality goal scoring opportunity from a cross, provided the opponents are in their defensive transition mode and have not settled down into their defensive block.
The opponent’s defensive phase movement is a key factor here, since if they have settled down to their defensive block, even when their defensive line is stretched, there would be other players (midfielders/attackers) to close down the open space or to win the second ball and the space to attack the cross would also be very less inside the penalty box. Crosses from near the penalty box (Zone 16 and 18) would have forced the opposition defenders to drop deep towads their goal, thereby reducing the space in between the defenders and goal. In this case, Bayern Munich and Marseille preferred to use cut-backs, to attack the space in front of the defenders rather than behind them. If the cross is from either zone 13 and 15, ideally there would be space between the defenders and the goal for the attackers to attack the cross and the attacking team will make use of that space.
Figure 6 shows the cross location and the distance of the last defenders from the goal (size of the circle). The larger the circle size, the higher the distance between the last defender and the goal.
It is seen that as the cross moves towards the goal (closer to the penalty box), the defenders drop deep to cover the space in front of the goal therey increasing the difficulty in finding space to attack the cross. Crosses from wider regions have pushed the defenders higher on the pitch, creating space in between the defender and the goal (Figure 3).
Pre-Cross Build-up Analysis:
Out of the 35 crosses that were analyzed, only 9 crosses involved a 1v1 dual in either Zone 16 or 18. Only 4 crosses, out of these 9, were delivered directly after a 1v1 dual. In the other 5 crosses, after the 1v1 dual was won, the ball was passed backwards to the nearest zone before being crossed: If the 1v1 is won in Zone 18, the LB and the opposition midfielder drop deep to stop the cross, creating space for one Marseille player in Zone 15 as seen in Figure 7.
The Marseille attacker takes on a 1v1 dual with the opposition left back, forcing the opposition left midfielder to drop deep and provide cover for the left back. This creates free space for the Marseille attacker in Zone 15 to receive the pass back and deliver a cross into the box. This scenario depicts the role of the free man in the free zone and using him to cross under less pressure. There were higher instances of defenders not applying enough pressure to block the cross (Figure 7 and 8).
Figure 10 shows a situation where there’s a 2v2 situation in Zone 18. The Marseille attacker, at initial stage of build-up, is 1v1 with the left back and drags him wide. The opposition midfielder drops back to provide support to the left back, thereby creating a 1v2 overload against the Marseille attacker. The ball is then passed back to the nearest free man, who’s under less pressure before delivering the cross.
The 2 defenders don’t close down the space and the attacker crosses under less pressure. The scenarios that were discussed above were present in the crosses that were considered for this analysis. Marseille’s main objective in build-up to a cross is to use minimum number of players in build-up and overload the cross attacking zone, create a free man who could deliver the cross under less pressure.
Analysis on Defenders:
“You see defenders today who follow an opponent everywhere. The defender’s point of reference should never be his opponent but a team-mate.” Arrigo Sacchi
The distances between the defenders, the defensive shape (flat/cover) and the distance between the defender and the goal (height of defensive line) were considered while quantifying defenders position and their impact to the success of a cross.
According to Arrigo Sacchi, the ideal distance between 2 defenders, while defending, would be between 18-25 metres. Defenders, who were near the zone of cross, were stretched more than the other defenders since they had to cover more distance to block or stop the cross, while the other defenders drop deep, to mark the attackers inside the box (Figure 11).
The LB in the above situation is dragged away from the goal as he had to close down the space before the cross. The LCB follows the LB to provide cover (13.1 metres). Meanwhile, the RB and RCB drop towards the goal (3.9 metres) to mark the 2 Marseille attackers. In this research, the average distance between the LB and the LCB was 11.11 metres, which indicates that Marseille preferred to cross from Zone 15 and 18. Figure 12 shows the distances between the defenders for all the crosses that were analyzed.
The Red, light blue and the dark blue bars are representing the LB-LCB, LCB-RCB and the RCB-RB respectively. The average horizontal distance between the defenders were found out to be approximately 26.5 metres. On a similar research conducted on Bayern Munich, the average distance between the defenders, of the team they faced in 2015-16 season (till winter break), was 29.5 metres. This could be because of the nature of the leagues they play in and the method of build-up to the cross. While comparing Marseille and Bayern Munich for successful outcomes (goals), defenders facing Bayern Munich had an average horizontal distance of 30.5 metres between them (Figure 13), whereas it was 23 metres in the Marseille research.
In the above situation, the LB is stretched away from his nearest defender. So the defensive shape for this cross is found out from the side which is stretched. The LB is 15.12 metres from the goal line, while the nearest defender to him, LCB, is 7.5 metres from the goal line. Hence it is considered that the LCB is providing the cover to the LB, based on their vertical distance from goal line. Similarly for the other 2 defenders, since the difference in their distances is less than 2 metres, they are considered as flat in shape. Note that all these are just calculated in order to have a more structured evaluation of the defenders based on their position with respect to each other.
Figure 15 shows the Horizontal distance between the defenders (bar chart) vs. Minimum Vertical distance between the defender to the goal (line graph), for all the analyzed crosses. The defender who’s closest to the goal defines the height of the defensive line (low/medium/high line). In case of figure 14, LCB is the defender closest to the goal. Therefore his distance from goal is considered as the height of the defensive line.
So if the defenders were horizontally stretched, they could still be in a low or medium defensive block (lower vertical distance between the last defender and goal) and stop the cross. The average distance (vertical and horizontal) when Marseille missed or their attempt being blocked by a defender, was 9 and 27 metres respectively. For their successful crosses (Goal or Keeper save), the distances were 11 and 25 metres respectively. This indicates that their opponents were 2 metres closer to the goal and were less horizontally stretched (more compact) when Marseille had their unsuccessful attempts. The location of the cross has an influence on these distances between the defenders. The average distance between the last defender and the goal (vertical distance of the defensive line) was 10.3 metres. Figure 16 and 17 shows the average Horizontal and Vertical distance for the entire cross outcomes of Marseille.
Attacking Player’s Analysis:
Marseille’s strategies to attack the final cross are analyzed in this section based on the position of the attacking players with respect to the defenders near them. Figure 18 shows the player location for a Marseille cross from Zone 15. The 2 attacking players, inside the box, are close to their defenders and the distance between the defenders and attackers are measured. If the distance between the attacking player and defender is less than 5 metres, then it is considered that the attacker is “Man-to-Man” with the defender or else he’s in a “Free” role.
The idea of evaluating player-defender distance was mainly due to the fact that most of the crosses had similar patterns of attacker-defender getting isolated inside the box with a 2v2 and 2v3 scenario (Figure 19).
In the above graph, the circles represent the players attacking the cross. The green region represents the players who were under the “Man to Man” category while the blue region represents the players in “Free” category. There were very few instances of only 1 player attacking the cross. The players in free role were largely near the edge of the box, positioned to win the 2nd ball or were attacking the cut-back cross.
Figure 20 shows player location of few crosses. The second defender, behind the first stretched defender, will be isolated. The last 2 defenders (RB and RCB in case of figure 20) were isolated and targeted to attack.
The cross would be aimed at the zone where Marseille had overloaded the defenders. So in order to achieve this, Marseille, during the cross build-up, would try to provoke the opponent into committing more men into the crossing zone, thereby forcing the opponents to have less men inside the box.
Jed Davies, in his upcoming book on Marcelo Bielsa, explains the concept of “Overload to Isolate” (Figure 21) where one zone/channel is overloaded and then play is switched to the opposite isolated zone. https://www.youtube.com/watch?v=jHyHk0VCDAE
The build-up to the cross was restricted to the wide channels near the throw line, since the attackers have to move closer towards the goal and space has to be created near the goal (Figure 22).
The wide Marseille player would ideally come inside and make a blindside run behind the last defender. If the cross was from closer to the penalty box, the Marseille midfielder (not involved in the build-up) would place himself on the edge of the box to open up a cut-back option. The movement of Marseille’s wide attacker behind the opposition full back caused problems to the opposition defenders.
Cross Location Zones:
- The location of the cross has a considerable significance to the outcome of the cross.
- The opponents were in transitional phase when defending the cross.
- Crosses that were away from the goal had higher success rate compared to the crosses from near the penalty box, although exceptions like goals from cut-backs were present.
- The location of the cross influenced the horizontal distance between the defenders, which in turn influenced the final outcome of the cross.
Pre Cross Build-up:
- The build-up pre cross involved minimal number of players, with maximum of 3 Marseille players.
- There were 1v1 duals in Zone 16 and 18, which were used to create a free man in the nearby zones to deliver the cross under less pressure.
Analysis on Defenders:
- Defenders were evaluated based on their vertical and horizontal distances.
- The average distance (vertical and horizontal) when Marseille missed or their attempt blocked by a defender, was 9 and 27 metres respectively. For their successful crosses (Goal or Keeper save), the distances were 11 and 25 metres respectively.
- The distances between the defenders had significant impact on the final outcome of the cross.
- The defenders were also categorized into “Cover” or “Flat” in shape based on their position with respect to their adjacent defender.
Attacking Players Analysis:
- The attacking players were evaluated based on their distance from their nearest defender.
- They were categorized into “Man-to-Man” or “Free” role.
- Marseille’s preference to create overload inside the box to attack the cross was evident.