Koeman needs a spark to unlock his old side

First, a note about last week: yes, it was frustrating as hell, but after the previous debacle against Chelsea it was good to see Everton tighten up the defense again. For me this season is as much about correcting the mistakes of the Roberto Martínez era and building a foundation for next year as it is about qualifying for Europe. Using that lens, you’d have to at least be somewhat pleased with the team’s organization and defensive form. Swansea did little going forward on Saturday and were probably lucky to get their goal. Meanwhile Everton created enough chances to win and on most other days would have. Silver linings and all that.

With that in mind (and to slowly segue to Everton’s upcoming match), here’s something I tweeted six and a half weeks ago:

I was looking at some expected goals numbers and noticed that it seemed even at that early stage that a top tier of the league was separating itself from the pack. Six matches later that hasn’t really changed, according to any model I can find out there. Both @analytic_footy and Paul Riley’s xG models have a clear top 8, and Steve Jackson posted a nice visual using his adjusted goal difference:

At this point, Koeman has Everton hitting that zone firmly in the Europa conversation but not really close to the top four.

Southampton overview

I bring this up now partially to note that the other team right around there is Southampton, Everton’s opponents this weekend. Despite again losing several key players and their manager over the summer, they have somehow managed to maintain their place as one of the better sides in the Premier League. They currently sit in 11th place but a number of metrics (including but not limited to the xG figures above) confirm that they are pretty unlucky to be there and should really more around sixth through eighth.

Saints have taken the 6th most shots and 7th most shots on target in the league this season but have only scored the 15th most goals. In other words, they are getting themselves into great positions but as Michael Caley might put it, the finishing pixie has deserted them. These things tend to work themselves out over the course of the season but in the meantime there have been some eyebrow-raising results to endure (losing to Hull, for example). The numbers and eye test suggest that Southampton, while not an elite attacking side, will likely start bulging the net with greater regularity as the league campaign rolls on.

Defensively they have been impressive to say the least. No one has conceded fewer shots on target, and xG likes them as well (third on Paul Riley’s model). While they rode their luck just a hair, it’s not easy to grab a clean sheet against Liverpool these days, and earlier in the season they held Arsenal and Manchester City to 1.0 and 1.1 xG, respectively. Furthermore, they are near the best in the league in shot quality against. Unfortunately they’ve been a bit snakebitten defensively just like they’ve been on the attacking side. In this case it’s that their opponents have been outperforming xG like crazy:

Essentially this means that despite not conceding many quality chances, their opponents have been converting low quality chances, which is something that is generally more down to crap luck than anything else (though it’s of course possible that the model is just missing something about their play).

In short the story so far for Southampton has been failing to get results to match their quality of play. The fact that they are level on points with Burnley is in the realm of the absurd (see below), but such is football. If I were a betting man, I wouldn’t bet on them to remain in the bottom half for long.

Tactics

After Koeman left St. Mary’s for Goodison Park, Southampton grabbed Claude Puel from Nice. Puel won Ligue 1 back in 1999-2000 with Monaco but beyond that his managerial career has been a bit of a mixed bag. He reached a Champions League semifinal with Lyon but domestically with them, Lille, and Nice, there were plenty of ups and downs. In recent times he garnered a reputation for playing attractive football with Nice using a 4-3-1-2/4-1-2-1-2 formation. He’s mostly continued that with Southampton thus far.

The core of Puel’s tactical setup at St. Mary’s is the midfield three—usually Steven Davis, Oriol Romeu, and either Jordy Clasie or new arrival Pierre-Emile Højbjerg. Romeu is the deepest of the three, while Davis is the most advanced and is tasked with linking play to the forwards.

Up top Puel has mostly maintained a front three of Dušan Tadić, Nathan Redmond, and Charlie Austin. Tadić has been productive pretty much since he stepped foot in the Premier League but for whatever reason seemed to fall in and out with Koeman. Puel has kept him in the side more consistently and has been rewarded. The most notable thing about Southampton’s front three is their lateral movement—each player is allowed to pop up in different areas across the pitch and are encouraged to run into the channels whenever there is space. This means that some games you might end up seeing Austin more wide and Redmond (formerly thought of mostly as a traditional winger) more central. With the front three constantly on the move, consistent width is to be provided by the fullbacks.

Putting it together, usually Southampton have a nicely spaced formation:

Translating that to the pitch, this is a typical setup of a Saints buildup:

The front three are occupying space in the center, scanning ahead to either make a run in behind or to hold up and offer a short pass to the ball-carrier. The fullback is providing width on the right, and either the left back or the furthest left of the midfield three is pulling out wide on the opposite side. Romeu is standing in the middle of the pitch, offering an option to circulate possession and making sure that there are no gaping holes in transition if Southampton lose the ball. While the front three do have freedom to roam, it’s impressive how often they maintain good spacing and don’t get in each other’s way (a notable exception was the match earlier this season against Manchester United).

This setup allows Southampton to vary their attack. Last year they became extremely reliant on crossing (6% of all passes were crosses) but they’re mixing it up more this year (4.9% through 12 games). Tadić is both an intelligent runner and passer, and Redmond has impressed in a more dynamic and central-ish role, involved in both buildup and the final ball.

Meanwhile, Austin remains a useful target in the middle, especially against better defensive teams who are less susceptible to through balls. His finishing hasn’t completely come together this season but he’s getting himself into good spots fairly consistently.

Matching up with Everton

Purely numbers-wise this is an even matchup. Somewhat worrisome from an Everton perspective is the midfield—contrast what I mentioned about Southampton’s solid 3 with Everton against Swansea, which featured a somewhat defensive and conservative Idrissa Gueye, a relatively nonexistent James McCarthy, and Ross Barkley acting as more of a second striker. The potential is here for Saints to dominate the middle of the pitch. This might be okay, and Everton could be content to focus their attack down the wings via Yannick Bolasie and whoever is on the other wing. Again though, one wonders where the creative spark will come from. Koeman still has some things to figure out in this regard.

On the evidence of last week, I wouldn’t be surprised to see Everton have some nice circulation and possession in the back and down the flanks, but it’s hard to imagine them regularly penetrating central areas. I’d be more than happy to be wrong though.