Not so long ago, Liverpool fans boasted of having “the best midfield in the world”, and it was no exaggeration. The world-class trio of Mascherano, Alonso and Gerrard were as good as any other in world football at the time, so it’s perhaps no surprise that those who have succeeded them have been made scapegoats for Liverpool’s recent decline. First Lucas was made the fall guy, then last season Henderson shared the burden of haters’ favourite with Stewart Downing. This year the role was filled by Joe Allen, seen by many as the poster boy for Rodgers’ new regime.
Things started well for Allen at Anfield. Liverpoolfc.com journalists named him their man of the match against West Brom, Man City and Arsenal, and for the first quarter of the season the young Welshman was one of the Reds’ best performers despite having to play as a defensive midfielder in place of the injured Lucas. It was just as the team as a whole started to perform better that Allen’s performances began to suffer, and we never really saw the best of him again.
For the first eight fixtures Allen’s Passing Accuracy Percentage was 92.875%, then in the away game against Everton at the end of October it dropped to 86%, where it stayed in the following match against Newcastle, and never really recovered thereon. As Andrew Beasley has already shown, Allen’s dip in form coincided with the shoulder injury he picked up sometime in late October, with both his passing and the physical side of his game and deteriorating. On top of that, Allen also became a father at the end of August, and as any parent knows, sleepless nights can leave you tired and performing under par. Being playfully labelled “the Welsh Xavi” by his manager probably didn’t help either, but was Allen really as bad as some fans think? And could he yet establish himself as a key player for Liverpool?
One of the common criticisms of Allen is that he passes backwards too often, but his percentage of backwards passes in 2012-2013 was only 11.93% – less than Jordan Henderson (19%), Steven Gerrard (15%) and Moussa Dembele (17%). Michael Carrick had the lowest percentage in the league for a midfielder on just 7.97%, and Mikel Arteta – who is widely considered an excellent passer of the ball – was near enough level with Allen on 11.64%.
In fact, of all the midfielders who played over 1700 minutes of football and 1000 passes, only Michael Carrick, Lucas Leiva, Craig Gardner, Ashley Westwood, Leon Osman, Claudio Yacob, Bradley Johnson, Gareth Barry, Yaya Toure, Youssouf Mulumbu, Darron Gibson, Yohan Cabaye and Mikel Arteta played fewer backwards passes. Lucas played the second least backwards passes of that lot, with just over 9% going towards his goal.
And Allen wasn’t playing too many passes sideways, either. 31% of Allen’s passes were played forwards, more than Mikel Arteta (29%), and only fractionally less than Yaya Toure and Claudio Yacob, both also with 31%. Not that passing sideways or even backwards is a particularly bad thing. Barcelona players do it all the time, as taking an ‘indirect’ route to goal is a key characteristic of the Tiki-Taka style. Demands for less sideways and backwards passes are essentially veiled calls for direct, typically ‘English’ football.
Another Allen myth says he can only pass the ball short. Again, this is a misconception. His long pass success rate last season was the second best in the league at 89.66%, so playing with Allen in midfield doesn’t stop us from breaking quickly when it’s on, or mixing the passing up from time to time.
Allen’s prefered game, however, is to receive and give the ball quickly, and to do that most effectively he needs teammates providing passing options close around him. And this perhaps gives us a clue as to why Allen was less effective as the season wore on as our play became more direct.
In Rodgers’ three-man midfield, the deepest player shields the defence, dropping in between split centre-backs and coming forward with the ball. The player furthest forward is the chief creator, linking the midfield and attack and threading passes through the opposition defence for attackers to run on to. The third midfielder – which was usually Gerrard for us last season – performs a sort of ‘shuttle’ function similar to a box-to-box midfielder or a ball-carrier in Basketball. His role is to bring the ball forward as the team transitions from a 4-3-3 (or 4-2-3-1) to a 3-4-3 by recycling possession via short passes and one-twos. To do this effectively, the player needs excellent ball control, decent dribbling skills, great awareness of where players are around him and most of all, the ability to keep possession of the ball and consistently find teammates with passes. Of course the more he can create the better (although with an AM and three forwards ahead of him there is no create pressure on him to provide chances), and defensively he needs to be adept at closing down and gaining possession through tackles and interceptions.
To asses Allen’s ability in this role, I’ve compared his stats for last season to other players who performed similar roles for their teams. The players I’ve looked at are primarily CMs as opposed to DMs or AMs, with one or two others thrown in for context.
Incredibly, despite having the worst tackle success percentage, Allen won possession more often than any of the other midfielders listed in the table above. This was in part down to his excellent interception rate, with one made every 38.38 minutes compared to an average of one every 47.75 among this group of players. Other than that though, Allen’s defensive stats are distinctly average. However, being ‘average’ compared to the league’s ‘elite’ midfielders when you carried an injury for most of the season isn’t all that bad.
Allen was better than the average here for Minutes Per Possession Won (10), Minutes Per Ground Duels (10.1) and Minutes Per Interception (38.38), suggesting he gets a lot of defensive work done, and his Minutes Per Tackle (36), Ground Duels Won Percentage (51%) and Minutes Per Loss Of Possession (101) rates were at least in the same ballpark as the average figures.
Where Allen really needs to improve defensively is on Defensive Errors and his Tackle Success Percentage, but as Andrew Beasley discovered, his Tackle Success Percentage prior to getting injured was 67% – still below the average at 75.37%, but not the lowest score (Cabaye, 65%). Beasley said of his findings:
“The interesting thing to note here is that the frequency with which Allen attempted both tackles and especially aerial duels has decreased, whilst he is attempting ground duels more often. In other words, he has been less involved in the more physical types of challenge but happier to take players on with the ball at his feet on the deck. As would you be if your shoulder was in constant pain.”
Despite a dip in his statistics after October, the only players to out-pass Allen across the season overall were Mikel Arteta and Moussa Dembele. His Open Play Pass Completion (90%), Attacking Zone Pass Completion (85%) and Final Third Pass Completion (81%) were all well above the average (86.92%, 80.15% and 70.31% respectively) and his Minutes Per Successful Dribble (174.45) and Minutes Per Chance Created (91) were also above the averages of 223.51 and 96.69.
Clearly Allen would be a better player if he scored and assisted more goals, but if Jordan Henderson is anything to go by, it’s possible Rodgers can help him improve on these aspects of his game. In his Premier League season with Swansea, Allen made two assists and scored a goal every 723 minutes, so we know he can do better than he did for us last season. It is also worth bearing in mind that Allen played as a DM for most of the first half of the season until Lucas returned. If he were to play the ‘shuttle’ role with a more solid DM behind and/or beside him he would have more opportunity to get involved in chance creation and assisting and scoring goals.
Attributes that can’t yet be adequately conveyed via statistics are the quality of a player’s first touch, his mobility and his movement, but you don’t have to watch Joe Allen for very long to realise that these are some of his greatest strengths. Calling him the “Welsh Xavi” was no doubt a factor in some fans seeing Rodgers as a Brent-like comedy figure who ought to talk less, but while he’s clearly nowhere near the Spaniard’s level (who is?), there is some merit to the comparison.
On his day, and when fit and on form, Allen exudes class, especially in the way he turns away from challenges and plays quick, short passes to teammates. He’s the kind of player who is often available for a return ball, too, and these characteristics will be vital if Rodgers tweaks the system next season in the way I suspect he will. After Carragher replaced Skrtel we played with a deep defensive line, which meant the midfield were stretched over a large area between the defence and attack. Should we bring in some quick defenders like Papadopoulos and Ilori to join the already signed Toure, we could then play a higher line, squeezing the gap between centre-backs and our most advanced player, and bringing all our players closer together in a more compact formation. That will play right into the hands of a player like Allen, who although adept at playing long passes as we’ve seen, excels in a short passing game.
At the start of the season it sometimes looked as if Allen was the only player who fully understood how Rodgers wanted the team to play, and the other midfielders seemed almost out of sync with him. After adjusting the style Rodgers struck on something of a balance with Lucas and Gerrard occupying the two deepest midfield positions, making Allen the odd man out, but with the addition of Coutinho (or Mkhitaryan) at the front of the trio, and the possibility of a more physical DM to rival Lucas, Allen could be the perfect player to bind the midfield together.
Of course, that raises the question of where Gerard fits in, but the captain is getting no younger and can’t go on playing every game forever. But at only 23, Joe Allen has plenty of time to win Liverpool fans over, just like Lucas and Henderson before him.
[box_light]All of the stats from this article have been taken from the Opta Stats Centre at EPLIndex.com – Subscribe Now (Includes author privileges!) Check out our new Top Stats feature on the Stats Centre which allows you to compare all players in the league & read about new additions to the stats centre.[/box_light]
Dec 19, 2014 0
Dec 14, 2014 0