Thursday, November 7, 2013

In Defence of the Defence



Is it human nature to blame the unfamiliar? To distrust the different? I'll leave the behavioural questions to the social scientists, but that certainly seems the case when it comes to Liverpool Football Club and tactics.

I'll get to all the whining and grumbling in some quarters that has followed Brendan Rodgers' recent shift to a three man defence in a minute. But it's worth noting that we've been down this road before.


It's barely nine years ago that Rafael Benitez swept in to Melwood with his Spanish ways and started the dastardly deed of switching Liverpool from man-to-man to zonal marking at set-pieces.

For all the hue and cry that followed, at least from those talking head analysts on TV, you'd have thought zonal defence had just become number eight on the list of deadly sins. It would have been understandable had the new system suddenly caused Liverpool to leak tons of goals from set pieces. It didn't. And even if it had, you would expect that after a couple of years, once the benefits of the new system had been evaluated, the finger pointing would abate.

Yet, four years later, as Rafa's team reached its peak, every goal Liverpool conceded was still blamed on zonal marking. Not its execution or individual error, but the system itself.

Granted, there could still be a valid basis for such an indictment. Except that the evidence did not bear it out. First, in two of his six seasons at Anfield - 05/06 and 06/07 - Benitez's team conceded fewer goals from set pieces than any team in the Premier League (Opta) and were 4th best in three other seasons.

Second, not one pundit ever blamed man-to-man marking - as a system - for any goals conceded by teams employing that tactic, even though they often conceded even more goals than a zonal marking team.

Zonal marking was unusual and different, so it had to be wrong, regardless of evidence to the contrary.

Then there was the double pivot - two "defensive" midfielders in front of the back four. It was Rafa Benitez, again, that first brought this to Liverpool, when he started pairing the Argentine Javier Mascherano with Xabi Alonso in front of the Liverpool back four in 2007. Alonso is, of course, more than just a defensive midfielder, but granted, he played deep enough to qualify.

Benitez was relentlessly criticized for being too defensive. Why play so many defensive players, even against weaker teams at Anfield?

And again the system - not its execution - was blamed for every bad result. Of course, double pivots are pretty much the norm these days. You certainly wouldn't hear such criticism of the system now.

Not so in 2008. Spain, for instance, who were also criticized for pairing Alonso and Busquets at the start of the World Cup in 2010, did not play a double pivot at the Euros that year, favouring Marcos Senna ahead of Alonso.

So, it didn't matter that Liverpool's double pivot allowed the full backs to contribute much more to the attacking play, or that, in 08/09, when this "defensive" set-up was at its best, Liverpool scored more goals than any team in the Premier League. Much easier to point the finger at the unfamiliar, the different, the double pivot as the cause of all Liverpool's woes in the latter Benitez years.

Which brings me to Brendan Rodgers and the 3-5-2. It's now been 5 league matches since a combination of injury (to Glen Johnson), a surfeit of centre backs and the need to fit in two in-form strikers, compelled the manager to change tack and jettison his preferred 4-2-3-1.

On the whole, 3 wins one draw and one defeat would suggest its not been a disaster. True, the second half troubles that characterized the first five games of the season, haven't been completely eliminated. But it's fair to say performance levels had improved significantly up to the West Brom game - the best of the season so far - and even in defeat at the Emirates, difficult as it was, there were signs of progress.

But it is that loss to Arsenal that has got many fans questioning the very notion of playing three at the back, and seemingly blaming this for any negatives we've seen over the past few weeks.

But is this just another case of blaming the unfamiliar? Has this different system with 3 centre backs really been any worse than the usual back four we are accustomed to?

Well, as far as performance and results go, it certainly hasn't, even if the return of Luis Suarez was always expected to improve the team regardless. Indeed, it would greatly surprise me if anyone suggests that we played better football in the first 5 matches, than we have since.

The one stick that has consistently been used to bash the back-three formation is the fact that the team hasn't kept a clean sheet since switching formation, conceding 7 times in the last 5 games. Of course, a back four had also let in three goals in the two games right before the switch, but its instructive to look beyond the bare stats when considering those numbers.

Of the 7 goals the back three have let in, three have come from long range shots; two from free kicks; one from a penalty and one (Santi Cazorla at Arsenal) from a cross into the box.

Now, while one could argue that the defence - or the midfield ahead of them - could have done a better job in preventing those long range shots, conceding shots from distance is hardly a consequence of playing three at the back. The same could happen with a back four. More often than not, you'd prefer your opponents to shoot from distance - rather than close up - especially with a shot stopper like Simon Mignolet between the posts. He did reach Ki's shot at Sunderland, but couldn't quite direct his deflection wide enough. As for the other two shots, from Yohan Cabaye and Aaron Ramsey, they were simply unstoppable.

Likewise, while it is disappointing to concede from free kicks - basically high balls into the box - when there are three centre backs on the field, it's not a consequence of playing a back three since pretty much the entire team drops into the box and gets organised for the set play. Dwight Gayle for Crystal Palace, and Paul Dummett at Newcastle profited from poor marking at free kicks, just as Dejan Lovren did for Southampton when we played a back four of centre backs in week 5.

The penalty against West Brom? Sure, it denied a clean sheet, but again, I don't see how that is a consequence of playing a back three. Back fours concede penalties too.

That leaves us with the one goal scored in the box from open play - Cazorla's volley at the Emirates last week - which would be the only one that can be put down to the decision to play a back three, because it clearly exploited a weakness in wide areas - worsened by Cissokho's poor showing - that is a clear shortcoming of playing a back three.

But every system has its weakness, and every system can be fallible when execution is below par or concentration drops - as it did when Swansea twice cut through the middle of a four man defence to score goals in the box from open play at the Liberty Stadium in week 4.

The back three aside, we've learnt a few things about Rodgers' 3-5-2: it's enabled Liverpool to play both Sturridge and Suarez in a position where they can wreak the most havoc; it's allowed Rodgers to keep three central midfielders in that key area in the middle of the park - either in a 1-2 or 2-1 pattern; and, for it to work well, both wing backs need to be both aggressive on the offensive and resilient in defence.

The latter point - proven by its absence at the Emirates - is the one facet of this system that we are yet to see. Jose Enrique seemed well suited for the role in the League Cup game at Old Trafford - when Glen Johnson was still out injured. When the England right back did return, against West Brom, Enrique was out injured, and both players missed the Emirates match. I think the outcome could well have been different had both played, and it would be nice to see just how things would work with both wing backs back to full fitness.

Rodgers may well return to a four man backline against Fulham at the weekend, but I doubt this is the last we’ve seen of the three-man defence.


No comments: