Why the pundits are wrong about Ange Postecoglou: playing out from the back

Editor’s note: the majority of this piece was written last week, so while I touch on some moments from Spurs’ last two matches, the majority of the analysis only includes data from the first 20 matches of Spurs’ Premier League campaign… if you can call it that.

It’s high time (damn, should have made that joke last time) for a third installment highlighting the inane commentary we hear every week from the commentary box. It’s been a long time between drinks (my apologies for that), and there’s been a lot of water under the bridge since my last piece: Ange Postecoglou potentially earned himself a temporary reprieve from criticism with Spurs’ EFL Cup win over Liverpool; but then losses against Arsenal and Everton following a turgid win against Tamworth in the FA Cup mean there has been no respite from the familiar refrain of low-effort blather emanating from your TV and computer screens.

So far, we’ve talked about misplaced criticism of inverted fullbacks, as well as taken a data-driven deep dive into Spurs’ high line. But why stop there? So much #narrative, so little time – so today, it’s time to talk about one of every old-school, “propa footaller” pundit’s bugbears: playing out from the back. Is it really as risky as it seems?

Spurs are far too risky with the ball, and play it out from the back far too much

We’ve all heard this one. Any time any player turns over the ball in a deep area of the pitch (and I’m not just referring to Spurs here) commentators ridicule the attempts of sides to build from deep, retain possession, and pull opposition defenses out of position. These same commentators also conveniently forget the fact that professional footballers in today’s era are ridiculously fit athletes, allowing for complex and high energy pressing systems that minimize time on the ball for any player. Gone are the days are dropping into the “two banks of four” and waiting for the opposition to turn over the ball.

It is a criticism that has been levied at Ange-ball on a regular basis; especially when Fraser Forster basically kicked the ball into his own net twice in a League Cup match against Manchester United. And after moments like those, it’s a critique that is easy to make; but is it a valid stick with which to bash Ange-ball?

First things first, let’s be very clear: there is very little correlation between playing out from the back and dangerous giveaways. Much like in my previous piece on the high-line, there are limited publicly-available data that would allow somebody to quantify how much (or how successful) a team is playing out from the back, but we can make a few inferences from said available data. To that end, I’ve charted up buildup percentage, the percentage of passes that occur outside the attacking third for a team, which is a decent measure of whether sides prefer to build via passing from the back, or to go more direct, the results of which I think line up with the eye test. I’ve plotted this against dangerous turnovers, all losses of possession close to a team’s own goal:¹

Manchester City and Arsenal, two sides that like to play out from the back and build from deep, are the safest sides in possession; conversely, Leicester City and Crystal Palace, two sides that tend to be more direct in their approach routinely turn over possession close to their own goal. That’s not to say playing out from the back works for all sides: Russell Martin was bullish in his approach with Southampton, where he clearly didn’t have the players to execute his vision of flowing, passing football – and that resulted in his sacking and will likely result in Southampton’s eventual relegation.

With Spurs’ poor form, we’re getting into the territory where the same questions could be asked of Ange Postecoglou’s side – but as you can see from the above, Spurs aren’t really moving the needle in that respect, nor does playing out from the back (in theory) affect defensive records. Spurs play out at a similar rate to a cluster of other sides, a number of which have decent defensive numbers, and don’t tend to give the ball away in dangerous areas a significant amount. Tottenham are not elite in this respect, but comfortably top half.

Let’s have a look, then, at how Spurs’ squad has handled this way of playing, which is where things get interesting. Looking at data² on Spurs’ center backs, we see some very notable… differences in skill:

It is immediately obvious just how dependent Spurs are on their starting center backs, not just defensively but also in retaining possession and starting attacking moves. When it comes to ability on the ball, however, there are two players that have clearly struggled: Archie Gray and Radu Dragusin. Both have poor ball progression numbers; but while you can’t really criticize Gray as an 18-year-old playing out of position for playing it safe, it’s hard to say the same about Radu Dragusin who while struggling to progress the ball has also made four errors. When opposition players feel free to admit in post-game interviews that they targeted a specific player for their shortcomings on the ball, you have a problem with your squad. It starts to become apparent that the problem isn’t the tactics, per se; it’s the players available (much like with Southampton).

The logical question then follows: why hasn’t Ange Postecoglou made adjustments to his way of playing if he doesn’t have the squad to execute his vision?

The thing is… he has!

Let’s do a bit of a breakdown on Spurs’ passing by fixture. By looking at the proportion of long passes against total pass attempts,³ we can see how willing Spurs were to kick it long to relieve pressure. We can also take a look at how many errors were made in each match² at the same time:

Across the first 10 fixtures, Spurs only attempted to play more than 6% of their passes long on two occasions: against Brentford, and a high number against Crystal Palace, a match in which Spurs struggled against a very targeted press masterminded by Oliver Glasner. In contrast, Spurs went long with over 6% of their passes seven times out of the next 10 matches, as injuries set in and Ange began to make adjustments. It’s not charted above, but Spurs went long 10% of the time against Arsenal as well, while still making 3 errors. There’s clearly not much in the way of correlation between playing out from the back and errors resulting in shots; in fact, you could almost say the opposite, with the three matches in which Spurs made the most errors also matches in which Spurs looked to play more directly (four, if you include the most recent North London Derby).

We can look at this from another perspective as well. I manually assessed match data to quantify turnovers by location³, to then analyze how often Spurs were losing possession in their defensive third and plotted this against the above pass data for each fixture (numbered by matchweek):

There’s not much to say here, except for the fact that Spurs have been willing to be more direct at times and there’s no correlation between that supposed conservative approach with the ball and a reduction of giveaways close to one’s own goal. There are certainly more data points in that top right quadrant (playing a more direct style, but turning over possession closer to goal at a higher rate) than there is in the top left (playing shorter, but losing possession in the final third). Again, it all points towards the players over the tactics.

The reality is: there is only so much you can do to account for individual errors. Football is a weak link sport, where a single player (or a single mistake) can decide matches and completely skew the way your team performs. Go back and rewatch Yves Bissouma’s giveaway that led to Arsenal’s second goal in the North London Derby. Is it an ideology, or tactical direction that resulted in him dribbling into two Arsenal midfielders? I would hazard to say no, especially given Djed Spence’s movement into space for the return pass following him playing the ball to Bissouma, which was much more in keeping with what we know of how Ange Postecoglou likes his sides to progress the ball.

And that brings us back to the aforementioned errors by Fraser Forster in the League Cup. If we look at data¹ on our goalkeepers this season, much like with our center backs we see a wide range of proficiency in terms of on the ball skill:

You can see there is a gulf between Vicario and Spurs’ other keepers when it comes to playing it short (or long, in this case), but that hasn’t prevented Fraser Forster being Spurs’ most error-prone keeper, with three errors in seven appearances in the league (so that’s not even including THAT Manchester United performance). The slight outlier there is new signing Antonin Kinsky, who is clearly comfortable with the ball at his feet but still launching it on quite a regular basis; likely thanks to his superior passing ability resulting in him finding a teammate with more efficacy than Spurs’ other keepers (46.7% of the time, compared to Forster at 35.5%).

There is a bit of a chicken-egg situation here. Are the players launching the ball long because of a shift in tactics, or because of their lack of ability on the ball? I would say it is a bit of both, with the likes of Forster and Austin less confident in their ability to find teammates at close quarters but with all of the above data as well as the consistency of the shift in launch numbers post-Vicario’s injury also painting a picture of tactical shifts to accommodate this.

Looking at the data, the assertions from the pundits become somewhat amusing; not because Spurs are particularly secure at the back, but because Ange Postecoglou has shifted somewhat away from that which pundits have been critical, and it’s actually caused MORE problems for Tottenham at the back. Well, that, and that old familiar refrain: injuries. This makes the pickup of Kinsky early in the January window a real coup for Spurs, though his… “confidence” on the ball could potentially cause its own issues.

Doesn’t it become profoundly frustrating when you hear complaints coming from the talking heads that don’t align with anything you see on your screen? Check back in in the next couple of weeks, where I wrap up the series in something hopefully resembling a neat little bow.

1. Data from MarkStats
2. Data from Fbref
3. Data from Whoscored