It’s the defense: By the Numbers


I keep turning this post over in my head. What numbers matter?

I remember when Arsenal got beat 5-1 at Anfield in February 2014. That felt like a low point but one which we turned around: beating them 2-1 in the FA Cup one week later on the way to our first piece of silverware in 9 years.

The next year Arsenal played an unusual defense-first approach to Liverpool and beat them 4-1. And this really felt like we had turned a corner, like we had learned how to play in the big games for the first time in years.

Then Liverpool fired Rodgers and hired Klopp and since then, Arsenal have played 5, lost 3, drawn 2. In the last three games, all losses, Arsenal have conceded 11 goals to their 4.

This is the problem. These are the numbers. Wenger’s weakness as a manager has always been defense. His best defensive performances were in 98/99 when his team conceded 17 goals and 03/04 when his team conceded 26 goals. Both of those teams conceded the fewest goals of any team in the League.

It’s important to understand defense in the context of each League season. The average goals allowed by the best defensive team in the League (just counting since 1997) is 27. Wenger’s average has been 36 but.. from 1997-2008, Wenger’s teams averaged 33.45 goals conceded and from 2009-2017 that average jumped to 40.44.

Wenger’s nadir in terms of defense was 2011/12 when Arsenal conceded 49 goals. That was the infamous trolley dash season where Wenger tried to hold on to Cesc Fabregas and Samir Nasri but gave up after Arsenal had their asses handed to them by… Liverpool in a 2-0 loss on opening day at home. Nasri wanted out and had a stinker. One of the few stats that I have memorized: he was dispossessed 9 times on that day.

Buying Mertesacker and Arteta helped and over the next four seasons, Wenger seemed to have gotten control of the defense – in 2012/13 Arsenal conceded 37, then a bad year with 41, then 36 and two years ago posted an impressive defensive record, allowing just 36 goals, 1 more than the best defense in the League. That was the Coquelin season and not coincidentally the last season that Arsenal have made a real title challenge.

Here is the screen shot from my database with all of the above data:

That 2015/16 title challenge also had underlying stats that looked great for Arsenal. Again, a screenshot from one of my databases:

Arsenal only allowed 35 big chances that season and just 97 SiP (shots in prime, shots between the penalty spot and the goal). And when we did allow shots Cech saved them at an unbelievable rate. Where Arsenal failed that season was simply in conversion: we couldn’t put away the chances we were creating.

But the next season, in an attempt to generate more offense, Wenger dropped Coquelin and Arsenal went right back to conceding chances:

Arsenal went from conceding 35 big chances to 59 (from 0.92 per game to 1.55 – and Big Chances are scored at a 50% rate, so this is a key stat) and from 97 SiPs allowed to 124. Özil and Alexis also stopped creating as much, the BC created and SiP created numbers went down but that was OK because Arsenal’s conversion jumped from a bog average 11% to 14%. What lost Arsenal the top four finish last season was defense. Arsenal conceded 44 goals, 18 goals more than the best defense in the League.

And now already in this season Arsenal have allowed 7 Big Chances and if I sort the entire table by defense only, Arsenal are 5th worst in the League:

Of course it’s three games in and there’s a lot of room to turn things around but it looks grim, folks.

I need to say two major things here. First, I’m not advocating for the use of Coquelin. I have a lot of love for Coquelin and I think many Arsenal supporters underrate him but a club like Arsenal shouldn’t be relying on a player of his limited abilities. In the match against Liverpool, Bellerin made an error on the corner kick but it was also Coquelin’s responsibility to cover there and he was way too deep in the opposition box which left Bellerin all alone and gambling. Coquelin was also caught ball-watching on their 4th, leaving huge spaces open for Emre Can to roam into. Coquelin may be a short term answer but over the long run Arsenal need a player like Kante or Leipzig’s Keita.

However, it’s also not a coincidence that Wenger’s 07/08 team only conceded 31 goals – which is the third best of any Arsene Wenger side – and that side also featured a terrier-like midfielder in Mathieu Flamini.

But second, the problem here is larger than simply personnel. I have no doubt that Aaron Ramsey and Granit Xhaka are capable of playing as a two man shield in front of the back three. They would need to be told that’s their job and they would need to be trained on how to play that way but I’ve seen Aaron play a more defensive role for Wales and he’s not a stupid boy. He can play in a variety of positions. With his lungs and movement, he could be a great asset in a two man midfield.

And Xhaka’s obvious flaws are both his physical limitations in terms of speed but also his decision-making. Again, this is something that a good manager would either get out of his game, find a way to compensate with a good pairing, or would get rid of the player.

The problem is that Arsenal don’t seem to train on defense. There’s no defensive cohesion to this team and players (like Ox and Ramsey) are caught cherry-picking up the pitch or allowing their marker to run right by them. Instead of a team approach to defense, Wenger relies on special performances from individuals. And when those individuals fail to perform he seems to have little time for them: Mustafi is being sold, Mertesacker was dropped 105 games ago, Vermaelen, Szczesny, Gibbs, Holding, Chambers, even Sol Campbell was unceremoniously dumped after a bad game, on and on – Wenger consistently blames the defensive personnel and not the system which exposes them. Modern football managers will shred your team unless they are organized.


Sources: my databases


Leave a Reply

82 Comments on "It’s the defense: By the Numbers"

Sort by:   newest | oldest | most voted