Thursday, September 30, 2010

Villarreal's South American-European fusion:

Villarreal prove that it is possible to play 4-4-2 across four lines

Above figure 1: Villarreal's 4-4-2/4-2-2-2 shape.

If Jonathan Wilson's explanatio
n as to raison d'être of the 4-2-3-1 formation is true (affording licence to playmakers and dribblers in an age of increased physicality), then little wonder it first became popularised in Spain, that country that produces a phalanx of ball-players; players who would be miscast if they were to operate as traditional box-to-box dynamos in a 4-4-2. Witness, for example Roy Hodgson's struggles to impart lessons on Liverpool's more adept ball players, or more pointedly, Joe Cole's entire history as a young footballer.

Another reason perhaps why 4-2-3-1 is advantageous to such players, is that it enables them to defend higher up the pitch. When their team recovers possession, the creators are in closer proximity to both their own forwards and to their opponent's goal. Closer, in other words, to their natural habitat, to their comfort zone. Clearly you would not want to see the likes of Joe Cole or Theo Walcott vainly huffing and puffing inside their won final third.

It is a staple of tactics-centred discussion that formation is not an absolute but a neutral template subject to the holistics of exploiting player characteristics. From this principle can be extrapolated a further observation; the formation (based on average starting position) is not the shape a team will adopt during defensive phase, with teams switching to one or even two seperate configurations depending on the mome
nt of play. As Rafa Benitez said of 4-2-3-1; it can transform itno a 4-3-3 (4-2-1-3) or a 4-4-1-1 depending on the human material available and the manager's preferred model of play or his overriding specific objectives for the game at hand. Clearly some formations are more congenial to certain transformations than others due to the simplest of positional adjustments, and so 4-3-3 can become 4-1-4-1 or 4-2-3-1, 4-4-2 invariably morphs into 4-4-1-1, 4-3-1-2 tends to become 4-3-2-1 and even 4-2-2-2 can be transformed into a 4-2-3-1 through the application of one of the strikers.

Internazionale under Jose Mourinho played a 4-2-3-1 (that became 4-4-1-1 in the defensive stage) but that, under a different coach with a different set of proirities and the same group of players, may well have been rendered it a 4-2-1-3 (becoming 4-2-3-1 in defensive phase). [This seemed to be the intention that Rafa Benitez was entertaining at the start of the 2010-11 Serie A season; a more pro-active Inter stationed further up the pitch which in theory would alleviate Samuel Etoo' and Goran Pandev of much defensive burden. This plan of course has been overturned by an unseemingly concession on the part of Benitez towards positional autonomy and tactical assymetry largely at the behest of Samuel Etoo, with the result that Inter now resemble a leftward-inclining 4-2-2-2 whenever Milito dislodge
s the Cameroonian international towards a nominal wide role].

All of the above permutations are in adherence to the near-universal truth that top-flight teams nowadays can no longer afford to work less than nine men behind the ball (Fabio Capello dixit), the dilemna is one of how to configure these nine men in an effective manner without causing total distortion of their natural game. For certain players, there must be a compromise made between standing around idly and tracking back all the way to their own corner flag.

We have seen in recent yea
rs, the past three seasons to be specific, the disturbance rendered by teams trying to not just tweak their formation, but thinking they can do so without overhauling their model of play. Arsenal, in a post Vieira-context, come to mind. That their 4-3-3 and 4-2-3-1 variants have not made them less vulnerable to conceding goals is due to a lack of intensity in their pressing, and not necessarily ferocity in the tackle. The football is arguable more expansive than ever but they are playing with the purpose and effectiveness of the lacsadaisacal Barcelona side of 2007-2008, going through the motions, usually comfortable but never quite convincing. This is largely borne of a side and a manager whose platform of play was built around defending relatively deep, absorbing pressure and then hitting teams on incisive ball-to-feet counter-attacks.

Yes, the Arsenal of the 2001-02 or even the Invincibles vintage could elaborate play to the point of dizzying opponents but this was tempered by two factors:
firstly, Arsenal's most convincing wins when coupled with poised and deliberative build-up play came at a time when in the Premier league many more mid-to-lower table teams played a very open, vertical and naive form o
f football. These green pastures abounded before inferior teams began to employ their essentially negative take on 4-5-1 and set-up camp around their own final third. Put simply, teams are no longer pouring forward and affording the Gunners ample space to cut through them on the occassion of counter-attacks.

Secondly, there is the small matter of Europe: can it be purely coincidence that Wenger's most successful foray into Europe came only in 2005-06 after switching to 4-2-3-1 for European encounters? It was seen at the time as a concession towards negativity, but in reality with an extra playmaker between the lines it helped Arsenal masticate possession as well as defend more effectively.

That older Arsenal (2000-2005) was a side which tended to switch from 4-2-2-2 to a 4-4-2 in defensive phase. The strikers were exempt of defensive duties whilst the wide playmakers were expected to align themselves with the rock-solid central midfielders in a line of engament, but in order to avoid space between their back four and midfield opening it was necessary to move the two banks close together; effectively Arsenal had two lines of defence which were usually stationed deep-to-medium high, which means that the midfield line (the first one to pressurise opponents) would either set up on the cusp of its own final third or else just inside its own half, respectively. Of course, the amount of times this would happen was barely noticeable since the proportion of time spent without the ball by Arsenal, who maintained possession much better than
did their domestic opponents anyway, was miniscule.
And even when forced into such a position, the side was blessed with quick passers and fast runners like Pires, Ljunberg and Henry so that the counter-attack was surgically effective and quick.

Those sides were also predicated on a defensively solid double pivot in midfield drawn from two of Vieira, Edu and Gliberto Silva. The departures of Vieira and Edu was monumentous in its implications for the sides playing style from 2005 onwards, for although Wenger would spend the next three seasons mai
ntaining the same formation (4-2-2-2), his promotion of Cesc Fabregas changed the complexion of the midfield. Shorn of Vieira's industry, the team was left with only one fetching midfielder (Gilberto or Flamini) covering now for three ball-players instead of two.
Of surpreme importance here was the defensive contribution and positional intelligence of Robin Van Persie from about 2007 until 2009. Even when playing as a strike partner to Emmanuel Adebayor, he consistently would drop off to form a first barrier of defence in the midfield; which meant that Arsenal could switch to 4-4-1-1 in defensive phase. More enticingly, the Gunners could even afford to allow their wide playmakers to partipate in pressing higher up the pitch (now more medium than low) in what was effectively a advanced midfield band of three (Hleb/Ljunberg on the right, Van Persie central and Rosicky/Nasri on the left); so 4-2-2-2 could become 4-2-3-1 when necessary. Arsenal were demonstrably more comfortable with this medium pressing during this transition period of 2007-09 than their confused attempts in a Barcelonaesque guise since 2009. Perhaps a return to this intermediary template might be recommended if the experiment with 4-3-3 continues to prove anaemic.

For a side who look effortlessly assured in a 4-2-2-2, one need not look further than Villarreal. When this team met Arsenal during the March 2009 Quarter-finals of the Champions League, both were playing lar
gely identical formations. At time I recall saying that Manuel Pellegrini's men would essentially be facing a more athletic version of themselves and it largely panned out. But the larger corpus of Pelle
grini's work at Villarreal from 2004-2009 offers encouragement for those who wish to build an side who will elaborate the ball but away from the popular 4-2-3-1 and 4-3-3 boilerplate.

Manuel Pellegrini's template was restored by current coach Juan Carlos Garrido following a brief parenthesis under Valverde who tried to introduce a Sacchian 4-4-2 based on high defensive line and hard running; didn't really suit the players' style. Basically, under Valverde they resembled a Premier League but they were foresaking their heritage; born of a South American style at a continental European tempo, like a Wenger side but without the physical intensity nor explosiveness. Like a traditional Wenger side they didn't initiate pressing until midfield or sometimes even deeper so as to invite opponents forward and leave space for counter-attack. Like Arsenal, two wide midfielders provided pause and playmaking, this figure rising to three whenever they switched to 4-2-3-1.

There is a tactical reason as to why South Americans find playing in Villarreal to be such a soft landing.

What is noticeable is just how
important an element is proximity in the passing game of sides like Villarreal.They like to triangulate when going forwa
rd, always in the vicinity to one another. Pellegrini said that all five channels of attack must eventually be occupied, but not by designated personnel

Brazilian journalist Rodrigo Leitão from the coaching website Universidade de Futebol has called this Wenger/Pellegrini shape a "U-shaped midfield" as a sort of compromise between the flat midfield four and the pure box midfield of Brazilian and Colombian sides; the former is more conducive to defensive cohesion whilst the later privileges the twin central playmakers. Pellegrini's model allows the wide playmakers to converge centrally when the side has possession, but they must return to defend the flanks during defensive phase. Brazilian playmakers on the other hand have no such obligations, usually opting to loiter centrally just behind their two strikers whilst leaving the grim work of assisting the exposed full-backs to the defensive midfield duo behind them (hence the prevalence of 'broken teams' in Brazilian football).
Apart from the ever-prevelant 4-3-1-2 enganche set-up, the Villarreal model is the most widely imitated interpretation of 4-4-2 across the South American continent. This is true of sides like Nacional of Uruguay, Lanús of Argentina and it was true of the River Plate coached by Pellegrini back in 2003 who caused a furore by moving D'Alessandro out to a wide position. Similar charges of sacrilege were laid at the Ricardo LaVolpe upon his return to Argentina in 2006 to coach Boca Juniors. The arguments laid forth by LaVolpe were practically self-evident: that without a singular fixed enganche, it becomes less easy for opponents to mark such a crucial player out of a game, and since the playmakers will spend much of the match swapping flanks and rarely static, assigning man-marking duties will be a big ask anyway. All that
is asked of the playmaker in return is a little more mobility and a willingness to work behind
the ball, as lesson which Deco and Luka Modric (and even a re-born D'Alessandro playing in Brazil) have taken on board, alas something which Pellegrini clearly failed to impart upon an intransigent Juan Roman Riquelme in 2007.

Garrido's current format heralds a return to the successful and entertaining ways of old. In Pellegrini's final year the wide playmakers (interiores-mediapuntas) were two from Robert Pires, Ariel Ibagaza and the injury-ridden Santiago Cazorla. Now Borja Valero joins a rehabilitated and delectable Cazorla for the 'wide' postions. When in possession they are rarely far from each other (even if one stretches the play, the other will invariably shift across to a central area much like Pires and Ljunberg did at Arsenal).

Since a central tenet of the Villarreal philosophy is that width be fluctuating and never permanent, mobility across the front line is as essential as the presence of attacking full-backs. At least one striker ought to be comfortable drifting out to the flank (as per Thierry Henry) and Garrido is blessed to be able to call upon two of the quickest strikers around; Giuseppe Rossi and Nilmar who has proved a perfect replacement for Nihat. That both forwards have doubled as wingers for their respective national sides gives an idea as to their pace. Perhaps the Yellow Submarine has been lacking the Plan B of a more muscular, fixed reference in front of goal since the departures of Joseba Llorente, Guille Franco and Jozy Altidore, but now they have a mobile and technically-giifted quarted in front of holding midfielders Marcos Senna and Bruno. Another option is to replace Senna with the more lightweight yet technically gifted Cani.

Figure 2 below: Villarreal attack - the wide-midfielders move into central areas, whilst a striker takes up a wide position.

It is a common sight to see one of the interiores, a full-back and a striker all whirling around along one flank, only to switch the play if and when the attack becomes congested. The maxim remains the same - three players must be in proximity in order to exchange passes at all times- and it is as much as defensive as an offensive principle; the higher the chances are off ball retention.

Below: Fig 3: Villarreal in defensive phase: pressure begins inside own half and 4-1-3-2 (Senna pressing alongside widemen), before becoming two flat banks of four stationed deep when necessary.

In defensive phase, the team can afford to keep both strikers relatively high since the midfield tends to drop off, conserve energy and only press the ball once inside their own half (whilst arming a potential counter-attack). Nilmar and Rossi are such livewires that they can be entrusted to find sp
ace along any of the five channels in the opponents' half regardless of whereabouts across the pitch they find themselves, thereb
y justifying the deeper positioning of the two banks of four.


  1. Roberticus, I love reading your articles. Nice work again.

  2. Interesting that you talk about five channels in the opponent half. Do you mean the gaps between and outside defenders in a back 4? That could provide some interesting theories around that for exploiting those gaps. I've always thought of attacking set ups looking at how to drag defenders out of position, but never thought about how to expoit the gaps and use those channels, since those channels are almost always there.

    Another unrelated thought...
    In response to "When in possession they are rarely far from each other (even if one stretches the play, the other will invariably shift across to a central area much like Pires and Ljunberg did at Arsenal)":

    I'm sure my enduring memory of Pires at Arsenal was of him hugging the touchline, either waiting in space or immediately after the transition into counter-attack. I know he came off the wing as he neared the opponent's goal, but I don't really remember him, or Ljungberg/Wiltord/etc, playing centrally enough to merit describing Arsenal's formation as a 4-2-2-2. Certainly no more than a standard wide midfielder would be expected to move laterally as the ball goes wide on the opposite side.

  3. AnotherJames,

    thanks for your insight on that Pires-era Arsenal.

    I didn't mean to convey that Pires and Ljunberg were under instruction to cut inside. When defending, they had to take up positions on the left and right of midfield respectively - but they were given freedom to either advance down the flank or move inside; ditto Villarreal.

    You're correct to point out that, at least in Ljunberg's case, the wide-player would stay quite wide unless making late runs into the box (standard for any wide midfielder, sure) - I also wonder however, would this be a natural consequence of Ljunberg being a right-footed player stationed on the right flank. Whereas Pires was right-footed but stationed on the left; therefore the temptation for him to cut inside much earlier was perhaps as much inclination as design.
    Pires' drifting in centrally (especially when Arsenal were patiently building up play as opposed to launching swift counter-attacks) provided the team with vision that neither Vieira nor Gilberto/Edu (for all their other qualities) were capable of.

    I also hope that I've made clear, or at least intimated, the distinciton between this kind of 4-4-2 and the Brazilian box midfield (which is unmistakably a narrow 4-2-2-2 during defensive and offensive phases), by stating that the Invincibles and Villarreal players are obliged to drop either side of the defensive mids.

    As for your first question; by mentioning 'five channels' I am referring to the outside-left, inside-left, centre-fw, inside-right and outside-right vertical bands; stemming from the days when attacking play was divied up between five forwards occupying those respective spaces. As tactics have evolved and the number of forwards diminished, the more dynamism and varied movement is required of remaining attackers and support players. Pellegrini's ideal was to have somebody stationed across the width of all five bands whenever Villarreal managed to arrive at in the opponents final third.
    Hope that was clear, apologies if not.

  4. fascinating piece as usual! Given what you wrote, it would have been even more fascinating and tempting to see Pellegrini at Liverpool than Hodgson. Kuyt would have been an ideal forward for Pellegrini to drift wide, despite his tactical limitations. Gerrard and Cole could have probably excelled in the mediopunta roles where they could drift wide or move central and whoever would have remained in midfield likewise.

  5. very very interesting article, Villareal are a very interesting team, I would be very interesting to learn more about their scouting systems, this is the same for Seville, how do they find the players from South America who will settle in Europe especially at the prices they pay.

    Good article


  6. Mahdi,

    With Pellegrini at the helm, Liverpool fans would have been entertained with his style of football. It would be a complete mirror image of Hodgson's; perhaps not quite intense enough to get them places in the PL though. But but nice, slow and flowing passing.

    Kuyt as the mobile second striker? Perhaps; though I could picture Pellegrini asking Babel or Jovanovic to do this. Perhaps he'd even convert Kuyt into a full-back or a holding midfielder.

  7. Khanage,

    Look up some reading material on Monchi, Sevilla's sporting director.

  8. Great article Roberticus as always.
    Interesting snippet on Valverde who i am watching closely at Olympiakos right now (with Ibagaza/Fuster)

  9. fantastic as usual. my only wish is for more regularposts, but I gess when you go that in-depth you can't just crank it out every week...

  10. Nik, how is Valverde doing in Greece? HOw does he have them playing?

  11. Josef,

    I'm glad you appreciate the posts, only I'm sorry I can't publish in greater number and more frequently, mainly due to other working committments.

  12. Very enjoyable article. Couple of questions -

    Do you think this is what the U.S. was trying to do in the World Cup? If so, where did they come up short?

    Also, when play is being built up the side the players are well positioned to defend in the event that they lose the ball. The only players in space for the opponent would be the far side flank players, and forcing a team to hit cross field passes under pressure, while by no means impossible at that level, is about as good as you can do. But what about when the ball is central? Are they extra vulnerable to counterattacks building on the flank, with both the opponents fullbacks able to quickly move into space? I haven't seen how Villarreal deals with the opponents in these important moments of saida.

  13. Space Ghost,

    Sorry for leaving it disgracefully late to reply to your post; I haven't checked in on the blog in over a week.

    In reply to your questions surrounding the U.S. in the World Cup, I'd say that there were indeed some similarities. Donovan and Dempsey were ideal for those advanced midfield positions a la Cazorla, Borja Valero and co. (Coinicidentally, over recent fixtures Borja has filled in for Marcos Senna in a central role - and performed admirably, adding slightly more creativity if not so much muscle - whilst Cani took a wide-creative birth).
    That the U.S. should adopt what is a quintessentially South American model of play seemed at odds with what was arguably their strong suit - their British-school pace and physicality, but kudos has to go to Bob Bradley for encouraging a different way of playing beyound their already well-executed wait-and counter game.
    As to why they didn't do so well, perhaps they simply didn't have the right kind of players in other positions. I mean especially in central midfield where I felt the two holders were both too destructive and not a Senna-like figure amongst them.

    In terms of Villarreal and that potential vulnerability following the loss of possession in a central area, I can only profer that one of the defensive midfielders is expected to come to the aid of whichever full-back is being threatened, so in theory the full-back shouldn't be abandoned to a 1 v 1 situation. But the threat in itself is as much an opportunity; for if the opposing full-back advances into space, one of the Villarreal strikers can move out to occupy that space he has abandoned, so that were Villarreal to break up the counter-attack, they would be in a position to launch their own counter and with an attacker in acres of space inside the opponents' half. That alone might even give the opposing full-back second thoughts about abandoning his post in the first place.

  14. regarding the USA, it was the shakiness of the back four, not inadequacy in midfield, that was their undoing. They also didn't have the forward that drifts wide, as Altidore and Findley simply weren't up to it. They really could have used New Jersey born Rossi though!!!!!! It played to the strengths of their two best players though (Donovan and Dempsey).

  15. Hi Roberticus.

    I really like your knowledge and your view on football. So I was wondering what you would think about playing this U midfield formation with two false nines who moves out to the side in the same way as nilmar and Rossi does.

    Which means you wold play sort of 4 2 2 2 but the outside players being the one strechting the play. As Barcelona plays with fab and messi in the middle.

    There will allways be an inside and an ouside(Overlap/diagonal) movements of the false 9 and the 10 when switch the play.

    In defending the two who are wide will mark the wingers, And make it Bielsa, defending, 4 4 1 1 Sacchian, or deep counter 4 4 2 like Villareal used to.

    In offensive face, I would like the wing back and the cm on the strong side to join the attack a lot like the old 4 2 2 2 formation.

    Wb D D wb
    10 M M 10
    F9 F9

    You win the ball becomes: 10 moves in, F9 Out,

    Wb D D WB
    M M
    10 10
    F9 F9

    Switch of play to the right. F9 in, 10 out, Wb and Cm joins.

    D WB
    10 10
    F9 F9

    If the wide player stays wide, then the false 9 position make foward run. This important to get the space for the midfielder.

    The differnce from Villareal/arsenal formation is not that much, but you will be able to use 4 media puntas up front with a lot of movement, and no fixed winger.

    In barca case, I dont think the are to relyint on Messi, but I do think the way they are playing, they are not getting the best of every player. In this set up, Messi will not have the perfect role for him, but the others will improve, and you can use the best players in my oppinnion aswell.

    Brazil, could play a fluid front four of Kaka, Ronaldinho, Lucas, Neymar. No cm midfield destroyers, but two who are good at both defending and attacking. Marcelo and Alves, will take in turn to attack.

    Or in Barcas example, They can use, Fab, Iniesta, Messi and Villa as a front four with more freedom in role to two of them than staying wide.

    I Would Actullay use Fab and Xavi in central midfield. Like Fab played with Flamini , in the U Shape at arsenal. But without having a destroyer and a Playmaker, I would rather have Xavi and Fab take in turns to go up to play, and to stay back. ( Like the wing backs) And a little like the Sacchi Idea that a Player should do both.

    Another good thing is that The 4 mediapuntas up front will then take in turns to make the midfield 4 depending of who is wide. And this will also give them less defensive burden, since it is not all the time the same person on the two position out wide defensively.

    Stephen R

  16. Defensive:
    Wb D D wb
    10 M M 10
    F9 F9

    You win the ball becomes: 10 moves in, F9 Out,

    Wb D D WB
    ......M M
    ......10 10

    Switch of play to the right. F9 in, 10 out, Wb and Cm joins.

    ......M...... 10