Everton’s home form in 2017 had been spotless coming into the weekend, but the combination of mounting injury concerns and the arrival of a Chelsea side with one hand on the Premier League trophy proved to be too much for the Toffees to overcome. Goodison Park has been a fortress this season, but The Grand Old Lady only has so much power.
Given the discrepancy in talent between the two clubs, injuries to Seamus Coleman and Morgan Schneiderlin, and Chelsea inevitable greater motivation coming into the match, that Everton managed to keep the match scoreless for 65 minutes was impressive, even if the subsequent collapse was predictable and disappointing. Those opening 65 minutes weren’t pretty, but they spoke to a reasonable tactical plan from Everton’s Dutch manager, who used an unconventional approach to keep Chelsea at bay.
On paper, his lineup for the match looked something like this.
With Schneiderlin missing, it seemed that Idrissa Gana Gueye was set to play as Everton’s deepest-lying midfield player, with Tom Davies playing a box-to-box role and Ross Barkley as the team’s No. 10. In reality though, the central midfield setup was more complex than that, which defined the way the entire team played.
Taking a page out of Jose Mourinho’s book (yuck), Koeman gave Gueye the duty of man-marking Belgian superstar Eden Hazard, as Mourinho had Ander Herrera do in Manchester United’s 2-0 victory against Chelsea last month.
I was skeptical that the Senegalese midfielder would be able to keep up for 90 minutes, but was pleasantly surprised at how effective the tactic was. Check out two key maps that tell the story of Hazard’s day, first his passes made, then his passes received (both courtesy of FourFourTwo.com).
Gueye didn’t completely eliminate Hazard, but he got about as close as you can hope against a player of his quality. The couple of chances Hazard did get came largely as a result of his pure pace, not something you can really blame Gana for failing to match.
Of course, Chelsea doesn’t lead the Premier League solely because of the efforts of one player — the Toffees still had to manage other dangerous attacking players, primarily Diego Costa and Pedro. In defense, Everton altered its formation to look something like this.
The back four remained largely unaffected, but Gueye’s man-marking changed the setup of Everton’s front five.
In defense, Dominic Calvert-Lewin dropped very deep, essentially replacing Gueye in the midfield three. With Gueye already marking Chelsea’s attacker on the right wing, Barkley and Davies focused primarily on harassing N’Golo Kante and Nemanja Matic in the midfield, while Calvert-Lewin helped to marshal Pedro and Victor Moses on the Everton left.
For 60 minutes, this defensive plan worked pretty well. Chelsea created a few decent chances, but those came more as a result of the individual shortcomings of Phil Jagielka and Mason Holgate, rather than any tactical misdeeds.
Everton’s average position map (courtesy of EvertonFC.com) perfectly summarizes this setup.
Calvert-Lewin was basically on top of Baines in defense down the left, while Barkley and Davies roamed the center of midfield. Gueye went wherever Hazard went, putting his average position almost on top of Jagielka.
You may have noticed that Romelu Lukaku and Enner Valencia’s average positions were very similar as well, but that’s a bit misleading. Often when you see attackers with such similar average positions, it’s an indication that a spacing issue occurred, but that wasn’t really the case with these two.
Look at their heatmaps, which give a better indication of where the pair spent the match.
Lukaku
Valencia
Lukaku’s biggest contributions came just to the left of center, while Valencia’s came just to the right of center, but both players got on the ball in all three channels. Their overlapping average positions reflect their positional fluidity up top, rather than a positional logjam.
In short, both Lukaku and Valencia moved freely at the top of the formation, with Valencia sometimes playing as a right wing and sometimes as a central striker, while Lukaku was a central striker, right wing, or left wing depending on the moment.
With Calvert-Lewin often coming up from such a deep position, this fluidity was necessary to ensure that defenders and midfielders had targets up top when the Toffees looked to move from defense into attack.
The entire setup looked something like the following.
As you might expect against a side playing three center-backs and two ball-winning central midfielders, the Toffees didn’t have much success attacking through the center of the pitch. Ultimately, their attacking plan probably didn’t revolve too much around playing through the middle, and their attacking-third passmap makes pretty clear why.
Instead, Everton correctly looked to isolate and get in behind the Chelsea wing-backs, Marcos Alonso and Victor Moses. This is most effective when done in transition, and you can see pretty clearly from the passmap that the Toffees weren’t effective enough at getting the ball wide to Valencia and Calvert-Lewin quickly to spring counter attacks.
Notice where most of the completed passes to the wings come from — 15 to 20 yards into the attacking half. This is an indication that Everton usually got the ball into wide positions when Chelsea was well-settled in its defensive shell — and asking Valencia or Calvert-Lewin to win 1-v-2s against a quality side simply isn’t going to work.
The result was that the team sent in a few tame crosses, but little else.
Still, Koeman’s setup put his team in a position where just one lucky bounce, important pass, or impressive run on a counter attack could have put his team ahead, without sacrificing defensive solidity. With Coleman and Schneiderlin out, that was the best he could hope for.
When Pedro finally broke through (on a goal I think of as 90% skill, 5% missing Schneiderlin, 5% slow reaction from Jagielka), Koeman had little choice but to push his left winger into a more attacking role and unshackle Davies and Barkley in hopes of finding a goal.
The results were predictable — Chelsea scored twice more as Everton never really got into an attacking groove, before or after the introduction of Kevin Mirallas and Arouna Kone (yuck). There isn’t much to say about that, frankly; the Premier League champions are better than a team that uses Arouna Kone when it needs a goal.
But there were some encouraging performances, particularly from Gueye and Calvert-Lewin — and of course, Koeman, whose plan kept his team in a match that it probably had no right to win.