Tom Carroll: An early assessment of Poch’s midfield experiment

A curious phenomenon has emerged in the past three Spurs games: one set of fans seem to be watching two entirely different performances.

I commented upon this in my recent piece on Christian Eriksen following the draw at Everton, but it continued in our disappointing double-header against Leicester.

On Wednesday night, particularly in the first half, the divide became even more stark if my Twitter timeline is a decent guide. While those in the stadium thought Spurs were excellent, those watching at home thought the team wasn’t creating nearly enough chances.

At the heart of this divide is one player: Tom Carroll.

This was nicely demonstrated by the Reddit poll after the Leicester game.


Was Carroll really much worse than Lamela, who had an average game and missed two good chances? It feels very harsh. Nonetheless, it is clear that his performances are causing considerable frustration among some parts of the fanbase.

I’d include myself in this camp. Of course, I want to see Carroll succeed and continue the “one of our own” philosophy that is bringing the club and fans together. But I am concerned that the recent inclusion of Carroll, over other potential Mousa Dembele replacements, is making Spurs a less effective team. There comes a point where you have to ask whether it is worthwhile persisting, not least because this ultra tight season potentially offers so much.

My theory is that, with Carroll, we play too “safe” — we have lots of possession, but there is limited movement from midfield into advanced areas. This means we aren’t creating many chances, let alone good ones. The football reminds me of the early AVB era, when we were competitive and organised, but too often the passing went nowhere. We’d appeared to crack that earlier this season when Dele Alli started in central midfield alongside Eric Dier, interchanging with Dembele in a more advanced role according to the match situation. It is frustrating seeing us slip back into bad habits.

So I thought I would try to back up this argument with some evidence, using various online tools. I’m going to ignore the Europa League, because no-one cares.

Possession vs productivity

The basics: Carroll has played 520 minutes in the league and two domestic cup competitions. He has started six times, scored once and has zero assists.

I wanted above to all see if Spurs had more possession when Carroll was playing, and whether we were creating fewer chances.

Using data from the club website, without Carroll in the starting line-up, through 17 matches Spurs average 54.8 percent possession, and 7.0 shots on target per game.

In six matches with Carroll starting, Spurs average 61.6 percent of possession, and 5.3 shots on target.

I’m not reading too much into the shots on target as it is a clumsy tool, I’d much rather know the expected goals per game with and without Carroll. Greatly appreciate anyone who can provide me with this information.

But the difference in possession seems quite large. Yes, it is a small sample size, and we’ve played a low-possession team in Leicester twice, but Carroll also started against Arsenal and Everton, two teams you would think see a lot of the ball. It feels a decent guide at least.

Passing troubles

Watching Spurs, you can see the ball going through Carroll a lot. At times, we’ve seen that lovely left foot attempt to slide in a through ball, but more often he is merely recycling possession.

Via @footballfactman, this is Carroll’s key pass map for his league appearances. In 362 minutes, the number of important contributions seems modest.


For comparison, this is the same map for Dembele and Ryan Mason, both of whom have played more minutes.


As you can see, they would both appear to be making key passes slightly further forward than Carroll. (I’d add that Dembele is a unique player who defies this sort of analysis — he is all about ball carrying and retention at key moments when he is playing well.)

What both Mason and Dembele do, and Alli even more so, is get forward from the central midfield role. With Eric Dier on patrol, they have license to do this.

carroll_passingI pulled together Carroll’s passing maps from Watford, Everton and Leicester (on the left) in the league (via Squawka). I thought these told a story.

Against Leicester (the bottom one, sorry they are so small but data viz is not my strength), you see a couple of attempts at passing into the area from an advanced position. But against Everton and Watford, Carroll made no forward passes from within 25 yards. In all three matches, there are no passes from what you would consider “behind” the defensive line, the sort of passes when you’ve made a break forward and stretched the defense.

To me, this may indicate what could be called “static” play. For what it is worth, Harry Kane hasn’t scored from open play in any of the six games that Carroll has started in domestic competition — he had one good chance against Leicester on Wednesday, but I can’t recall him having many chances in the other ones.

Is this Carroll’s fault? Of course it isn’t entirely his fault — the likes of Lamela and Eriksen should also be creating chances. But it is reasonable to suggest that Carroll’s inclusion is part of the creativity problem for Spurs of late. He may be being “scapegoated”, but not entirely unfairly.

I understand why, when you are at the game, it must feel like Spurs are playing well. Possession makes you think you are dominating, and Spurs try to play with tempo — they do zip the ball about.

But possession does not necessarily translate into effectiveness — just ask Manchester United fans how Louis van Gaal’s possession-based approach is going. They’ve not scored a first-half goal at home in 10 games. Likewise, it was noted on Monday Night Football a while back that Arsenal’s record in “big” matches was much improved this season, and one noticeable change was less emphasis on possession.

I suspect, judging from the Leicester match on Wednesday, that Pochettino has noticed this. In the second half, Spurs were much better — there were more attempts to get behind the solid Leicester defence, and Spurs created some good chances. It was ultimately an unlucky defeat, but it was hard to have much sympathy after a sterile first half.

I suggested previously that Carroll was essentially getting the Townsend treatment of last season — a run of games to prove himself. Poch is a stubborn man, so don’t expect him to deviate from this plan. Townsend started seven in his mini run, so Carroll may have another couple of starts coming up — Sunderland at home would be exactly the sort of game where a manager may “experiment”.

But I would suggest the evidence is building that Spurs would be better off ending the Carroll experiment and returning to either Dembele or Alli alongside Dier in the central role. Eriksen, meanwhile, is evidently frustrated with his peripheral role — we’ve had very few rumblings of this type from within the camp under Poch.

I await the starting line-up on Saturday with interest.

Thanks for reading. Please follow me on Twitter for more articles and general Spurs chat.

1 thought on “Tom Carroll: An early assessment of Poch’s midfield experiment

  1. Pingback: How Spurs can take it to the next level: A blueprint for the summer of 2016 | The Spurs Report

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s