By Naveen Maliakkal
Reliving the Nightmare
Going into Arsene Wenger’s 1000th match as Arsenal manager, I had a strange sense of optimism about our match against Chelsea. Unless Andre Villas-Boas is involved, matches against Chelsea always inspire nervousness and fear for me. Since Jose Mourinho became the manager, Arsenal have beaten Chelsea three times, once in 2007-08, once in 2010-11, and once in 2011-12, the only one at Stamford Bridge (trips down memory lane, anyone?). The reason for such optimism came from the fact that last year’s Chelsea had rather obvious strengths and weaknesses. At that point in the season, Chelsea had morphed into a side that defended well and effectively counter-attacked their opponents. However, this desire for defensive solidity led to players that lacked creativity and failed to occupy the right spaces in time to operate well as a possession-based side. Coming off a 1-0 victory over Spurs, in which Arsenal seemed content to sit back and hit their opponent on the counter, after an early Tomas Rosicky goal, it seemed like Arsenal had the willingness and ability to pull off the ideal strategy to use against Chelsea. Clearly, this was not the case.¹
Instead, Arsenal looked to control the match with their possession and their style played right into Chelsea’s hands. Chelsea set up in a 4-2-3-1 with a central midfield composed of Oscar up top and David Luiz + Nemanja Matic at the base. Andre Schurrle played on the right, Eden Hazard on the left, and Samuel Eto’o as the striker. The plan was simple. Press Arsenal heavily in midfield -> hopefully isolate Mikel Arteta in the process -> win the ball -> counter-attack and have this pattern in a while loop that terminated once the game was out of reach. Even though Chelsea often left their back four exposed by pushing Luiz and Matic high up to press Arsenal’s midfielders, they had reason to believe their strategy would work.
First, the long-ball route of press evasion represented a strategy with low profitability. With Bratislav Ivanovic, Gary Cahill, and John Terry against Arsenal’s one decent aerial presence, Olivier Giroud, Chelsea could comfortably win the first ball. It also meant that Hazard and Schurrle could ignore Arsenal’s fullbacks without much consequence, as crosses into the box to pick out Giroud would have little success. This allowed them to move infield, increasing the number of pressers who could suffocate Arsenal in central midfield.
With a lack of speed² in the front line, long balls into the space behind the back line did not represent a profitable strategy either. Therefore, Chelsea’s back line had more of an incentive to move up the pitch to limit the amount of space Luiz and Matic left behind when they took up advanced pressing positions.
Finally, Arsenal’s desire for verticality in midfield played right into Chelsea’s pressing. The benefit of such verticality is that it allows Arsenal to more quickly pass the ball up the field. Since passes move faster than players dribbling the ball, this means that Arsenal can move the play up the field with more speed. Against a side that predicates itself on deep-defending in an organized fashion, this approach helps to increase the probability that Arsenal find themselves attacking a disorganized defense, one that has a significant misallocation of resources given the game situation in time and space. When building from the back, pushing the midfield up can create space for one of the midfielders to drop deep to receive the ball, while the others stay high. If the player dropping deep is hard to dispossess or can anticipate danger, make a quick and effective decision, and execute it (Example: receive the ball, turn away from the presser, and then play the ball up-field, through the opponent’s defensive lines, to set the attack way), then this can allow the press to be bypassed with the ball winding up in the opponent’s half against fewer defenders, as the pressers find themselves behind the play.
Problems occur when you do not have this player. Without this player, presser can isolate the midfielder who drops deep, often forcing him to play the ball back to the center backs or goalkeeper, if a turnover does not occur first. The pressers can continue their pursuit of the ball easily, as they do not have to change direction much (they do have to keep passing lanes closed though), and apply pressure to the center backs and to the goalkeeper. This can lead to turnovers, panicky clearances, or speculative long balls. With the other two midfielders (Cazorla and Chamberlain) higher up the pitch, the midfielder in possession lacks easy passes to move the ball away from the pressure. The midfielder could funnel the ball to the fullbacks, using them as a pressure release value. However, Arsenal’s fullbacks (Sagna and Gibbs) did not provide much going forward. Often they wind up looking to pass the ball forward to the advanced wide player (Podolski on the left and Rosicky tended to be on the right) or make a simple pass into central midfield. If the former occurred, then one of the Chelsea fullbacks pressed the man on the ball. This led to a turnover, a pass back to the fullback, or a pass into central midfield.
Anytime the ball made its way to one of the two advanced midfielders, Matic or Luiz would press the ball, the other would press the other advanced midfielder. The key to this approach, outside of Luiz and Matic pressing properly, was Oscar. If either Cazorla or Chamberlain received the ball with their back to goal, under pressure, Arteta could provide an easy passing option. Arsenal could then restart their process of possession in an attempt to break down their opponent. Oscar looked to block this passing option, sticking to Arteta like glue (though Schurrle sometimes took up this role). This meant that Arsenal’s midfielders had to try to make the right turn towards goal to get away from the initial presser. Obviously, this tended not to work. Once the turnover occurred and the ball reached the first passer, with Schurrle, Hazard, Eto’o and Oscar in advanced positions and close enough to combine their movements, due to their defensive positioning, Chelsea could rip apart the defenders Arsenal happened to have in front of the ball. This pattern kept occurring and within 20 minutes, the game was over.³
Some Potentially Good News (?)
Chelsea ran that plan perfectly, in large part, because they had the right players on the pitch. The question come Sunday, if they wish to operate in a similar manner, is whether they still have the ability to execute with the players they have now, particularly with Cesc Fabregas. As I said previously, Oscar played a key role as a defensive No. 10 in Chelsea’s win last season. If we substitute Fabregas in for Oscar, Chelsea suffer a significant drop in tactical discipline and defensive intelligence. If the members of the attacking front are given zones to mark, then this could lead to Flamini or Wilshere finding himself in to receive passes from advanced areas more often. This would improve Arsenal’s ball retention, giving them more chances to circulate the ball, looking for an opening to exploit, before a potential turnover.
In response to this, Mourinho could look to have Schurrle or Willian (whoever he decides to start) either tuck in or man-mark Flamini. This could prove effective if the Frenchman stays in the area one expects a deep-lying midfielder to occupy. However, if Arsenal rotate in midfield or drop another midfielder deep more often, then they could more comfortably build from the back. If Chelsea opt for a man-marking style, Flamini swapping positions with a more advanced midfielder (probably Wilshere) has the potential added bonus of moving one of Chelsea’s counter-attacking threats further away from goal4.
If Oscar retains his place as an advanced disruptor, then Fabregas could find himself in David Luiz’s position of aggressively pressing with Matic. For all of Fabregas’ attacking ability, he does not have the strength, overall athleticism, and aggression of David Luiz to crunch into Arsenal’s more advanced midfielders. This could leave a soft spot in Chelsea’s pressing game for Arsenal to exploit. If Chelsea employ a similar strategy as they did last season, exploiting Fabregas’ defensive shortcomings could find Arsenal attacking an unprotected Chelsea back line.
However, in Diego Costa, Chelsea have signed one of the best defenders at the center forward position. In addition to his abilities in attack, his defensive qualities could prove pivotal in this match. He could operate as an effective high presser, pressuring center backs and the goalkeeper while keeping the passing lane in his shadow. He could work to deny easy passes into the midfield, particularly passing that bypass the first midfield line, funneling Arsenal’s play towards the midfield pressers. Finally, he could drop into midfield and look to mark Flamini out of the game himself, clogging the midfield even more so, forcing Arsenal to rely more on their center backs and their fullbacks to move the ball forward. If Chelsea employ a pressing strategy against Arsenal, regardless of what it is, Costa will probably play a key role5.
Finally, Arsenal have the quickness and the intelligence to make runs in behind Chelsea’s back line. If Arsenal take a page out of the classic 4-3-3 book, they could look to use two or all three of Oxlade-Chamberlain, Alexis Sanchez, and Danny Welbeck to push Chelsea’s back line deep and spread them out, creating more space between the midfield and the back line. This could be space Arsenal could exploit or the potential of leaving that much space unoccupied may limit Chelsea’s enthusiasm to send men forward to press Arsenal at all vertical levels, like they did last season.
Crosses and (Some) Set-Pieces are Not Arsenal’s Friends
The inability for Arsenal to score off set-pieces is well known, and against Chelsea, it should only continue. With Gary Cahill, John Terry, Diego Costa, Nemanja Matic, and Branislav Ivanovic all likely to start on Sunday, Chelsea have a significant aerial advantage over Arsenal, both defensively and offensively. Chelsea will have little regret conceding set pieces in an attempt to foul or win the ball. Chelsea will probably have little issue conceding the flanks if Arsenal plan on consistently crossing from wide areas. In this game, a cross in the air, into the box, is probably as good as a turnover. Therefore, when Arsenal get the ball wide, they will probably find more success trying to make a pass into the interior, into the ball-side half-space (which implies that an Arsenal player needs to get into that space) than trying to successfully cross the ball.
On offensive corners or free kicks where the flight of the ball will probably lack depth, Arsenal may look to win the first ball and the near post and hope that the second ball falls to the right guy. Otherwise, they may want to play the corner short. This way they avoid the resource allocation problems Arsenal tend to have on corners. Sometimes they send plenty of resources forward, but allocate them poorly, much like a Keynesian policy of stimulus spending or temporary tax cuts to revitalize an economy, leading to low success and too much of a risk of getting countered for such little potential reward (footnote: Like most of government spending, these Arsenal efforts on corners are an enormous waste with damaging consequences). Sometimes they do not send many men into the box, looking to avoid getting counter-attacked, but they still put the ball into the box. Once again, the probability of winning the first ball, let alone scoring, is so small that we might as well consider this a turnover, unless they do well to win the second ball. Off a corner, fast opponents have plenty of space to run past Arsenal’s defenders, so unless Arsenal understand how to press off the corner or win the second ball, they still face too large of a risk of getting countered relative to the potential reward.
Therefore, unless Arsenal have come up with some brilliant ways to allocate players on set-pieces against a side as big as Chelsea, they may want to treat corner kicks and set-pieces with little potential for the flight of the ball to have depth like opportunities to make unpressured short passes.
Since Arsenal Probably Won’t Sit Deep, Press High
When you are the underdog, you tend to choose one of these strategies to get a result—sitting in a compact formation or pressing (or both). You have to exert some control of the game out of possession to reduce the amount of control that the superior team has in possession. Arsenal did not sit deep against Chelsea last season when it seemed like an ideal strategy. With Costa and Fabregas helping to increase Chelsea’s effectiveness at breaking down a deep-defending side, it makes it even less likely that Arsenal will sit deep at Stamford Bridge. This means that Arsenal have to consistently press Chelsea, to both create attacking opportunities and to protect the back of their defense.
We have seen Arsenal utilize a high pressing strategy in matches against Manchester City and Galatasaray. We have also seen Arsenal fail to employ any form of pressing against a side like Tottenham. As is often the case, predicting Arsenal’s strategy without the ball from match-to-match can be frustrating. On Sunday, Arsenal may be pushed into a pressing game due to Mathieu Flamini, and not Mikel Arteta, at the back of midfield. Ironically, Flamini’s lack of ability on the ball (or to do anything productive) may incentivize Arsenal to use a defensive system that limits the amount of plays he has to make, especially on the ball.
Unlike some of Arsenal’s attempts at pressing high, Arsenal probably need to maintain consistent pressure on the center backs and even the goalkeeper. This belief comes down to two reasons. One, Arsenal do not have the sweeper keeper or two Raphael Varanes at the center back positions to properly manage a high line. Essentially, they do not have enough of an ability to quickly shift defensive resources when the ball is played beyond the back line. However, if Arsenal do not push their back line high up the pitch, a high pressing strategy can leave a large amount of space in front of the center backs. Chelsea have players who can exploit both spaces—the space behind the back line (Schurrle) and the space in front of the back four (Oscar, Fabregas, Hazard, etc.).
Chelsea also have players who are adept at winning the first ball out of the air, controlling it or flicking it/knocking it down to a teammate. Obviously Diego Costa can perform this role, but so can Branislav Ivanovic. If Arsenal do apply heavy pressure, Ivanovic may move higher up the pitch to provide a secondary outlet for long balls, using his aerial superiority to dominate Arsenal on the flanks. And in addition to providing a wide outlet, Ivanovic may be the off-the-ball attacking fullback in the Premier League, other than Pablo Zabaleta. In addition to his athleticism and aerial ability, he has the intelligence to make the right runs, particularly those underlapping runs (runs to the inside of the advance wide player) in the half-spaces that defenses find difficult to defend.
Therefore, Arsenal need to make it difficult for Chelsea to deliver accurate long balls to their target man or into the spaces around the back line. However, an aggressive high press is only as effective as the pressing side’s ability to limit the gaps they leave behind when they charge forward. That could be an issue.
This brings us to the fact that this will be Arsenal’s third game in eight days with the first two matches being a North London Derby and an important Champions League match. In particular, the fitness of Mesut Ozil and Jack Wilshere, if he starts on Sunday, has to be a concern. Although Mesut Ozil played quite well in both matches, he did look tired against Galatasaray. When it comes to Jack Wilshere, the recent ankle injury, combined with the physical play Chelsea will likely display, could mean a high rate of attrition for the young Englishman and/or a high risk of injury/aggravation of an old injury. With these two players probably playing as the advanced central duo in Arsenal’s 4-1-4-1, there may be too much of a risk that a high pressing game puts too much of a tax on them. It is games like this that Arsenal may miss Aaron Ramsey the most, both for his willingness to drop deep to help build from the back and his stamina/athleticism.
So, Arsenal could look to press consistently throughout the match. Or they could look to play like they did in the first leg against Bayern Munich last season. In that match, it appeared the plan was to press like mad for the first 15-20 minutes, hope to get a lead, and then sit back. The problem, outside of the lack of proper coordination in their pressing, was the period ended without that all important lead. We could see a game plan where Arsenal look to blitz Chelsea at the start, emptying a lot of the gas tank, searching for that lead. If they get that lead, then they will have a real chance at getting something out of this match. If they fail to get the lead, they could be in for a real hiding. Such is the case with high variance strategies.
¹Though if Giroud scores his chance at around the 3:45 mark, we may have seen Arsenal implement a more cautious strategy.
²Going with Tim’s recent motif, the word “pace” is used more appropriately to describe salsa that cowboys will not substitute for “stuff” made in New York City compared to its use in our discussions on football.
³This style of build-up also proved to be a problem against Dortmund with Arsenal’s desire for verticality leaving Arteta often isolated in possession. Maybe Arsenal will have greater patience in their build-up and stretch the pitch horizontally on Sunday, but I cannot remember a time where Arsenal tried to attack a pressing side in this manner.
4Some real problems with this for Arsenal are that Flamini is not particularly intelligent in his movement or great on the ball, so not having Arteta is a huge blow. Wilshere dropping deep is also problematic since his first instinct seems to be to dribble when pressed. In this match, in deeper midfield areas, this will probably lead to turnovers, where Arsenal do not have the bodies to press and slow Chelsea’s counters. While the reward of beating the individual marker is a positive, it will probably come too infrequently and is too low leverage (still need to move the ball 50ish meters to get into a good scoring position) to make the risk-reward proposition worth it, in most cases.
5I think we can all admit that Mourinho is making way too much out of Costa’s hamstring in an effort to convince Spain to not play him during these international breaks. That way he can reduce the chance of an actual injury to the only center forward he has that can do what he wants in his system, for 90 minutes, game-in, game-out.