Morgan Schneiderlin, Fabian Delph, Gylfi Sigurdsson... Schneiderlin, Delph, Sigurdsson... over and over again we’ve seen this midfield and the results haven’t been great. Andre Gomes will hopefully be back healthy after this international break and the monotony will end, but there have been many people (with good reason) asking why on earth Marco Silva won’t do something different to break Everton out of the slump they’re currently mired in. I have my own ideas about that - Alex Iwobi as a 10 and dropping either Morgan or Gylfi comes to mind), but a very popular suggestion has been to play Tom Davies instead of Morgan Schneiderlin.
Marco Silva actually did exactly that last season, when a muscle injury delayed the start of Andre Gomes’ Everton career. Tom Davies was inserted into the lineup, first across from Morgan against Huddersfield Town (a draw), and then across from Idrissa Gueye against Arsenal (loss), Fulham (win), and Leicester City (win). The idea was that Davies’ work rate was preferable to Morgan’s deep lying passing. After the Leicester match, Gomes was healthy and Marco very happily moved the Portuguese into the lineup ahead of Davies.
So why did Davies work in this midfield? Well, first of all, Silva had Davies do less passing work than Gomes had to do in the same position. Here some examples from last season: Gomes averaged about eight more passes per 90 (and completes a higher percentage of his passes), more long balls (completes a much higher percentage), and more dribbles (completed about the same percentage). Gomes, who I don’t think anyone who call a star - though he is a very good player - was given much more responsibility in the attack and he completed it more efficiently. With that lighter work load Davies was able to compete within himself.
The other thing that allowed Davies to work in that system was being across from Idrissa Gueye. In our very small sample of games listed above we see a draw against eventually demoted Huddersfield when he was playing with Schneiderlin and wins against Leicester and now demoted Fulham when he’s across Gana. Gana’s ball winning ability - and the lighter workload - masked Davies’ key flaw as a player: turnovers. Despite touching the ball far fewer times than Gomes per 90 minutes last season Davies gave the ball away far more often. Here are some turnover numbers for Tom since making it to Everton’s first team: 2018-2019 EPL: 3.2 turnovers per 90, 2017-2018 EPL: 4.2 turnovers per 90, 2017-2018 Europa: 5.5 turnovers per 90, 2016-2017 EPL: 4.2 turnovers per 90.
Now, a few things about those numbers. Davies played further forward in 2016-2017, where Gylfi plays now, so that number may be slightly inflated (though the next season Gylfi only averaged 2.9 turnovers per 90). Also, Tom deserves credit for the improvement he showed on the turnover front last season, he handled a higher passing load per 90 and still got his turnovers down. Fair play to him. It should also be noted Gomes averaged fractionally more turnovers (classified as unsuccessful touches and times dispossessed) than Davies last season per 90, but most of that difference comes from Gomes having 0.3 more unsuccessful dribbles per 90, which is simply a product of volume as he and Davies completed about the same percentage of their dribbles but Gomes attempted more.
With Gana gone and his replacement, Jean-Philippe Gbamin injured, Marco Silva simply did not have someone healthy who he feels can mask Davies’ turnover weakness and he does not trust Davies to take care of the ball. Davies has played 20 league minutes this year and turned the ball over twice, not a representative sample of anything but you never know how that combines with what the manager is seeing in training. I personally think we could see more of Davies once Gbamin is healthy again, as Silva can in spots create a facsimile of what he ran early last season.
Davies is still only 21, he has the potential to grow as a player. I for one wish he had gone out on loan this season, especially if this is all Silva planned on using him. As it stands, Davies is just not that effective a player on a Premier League level. Analytics site smarterscout.com actually seeks to quantify these things. Here are the ratings, out of 100, that Davies receives in four key categories: Attacking Output: 19, Defending Quality: 59, Defending Quantity: 86, Ball Retention: 30. As you can see, Davies is a very willing defender, that quantity rating is a product of his work rate, but he struggles mightily on the ball.
Morgan Schneiderlin, for example, has a similar attacking output, slightly better defending quality, far less defending quantity, and more than twice the rating in ball retention. Fabian Delph has significantly better attacking output, is less of a defender, but has nearly three times the ball retention rating. Gomes doubles Davies’ attacking output, has similar defending quality (less quantity), and doubles his ball retention rating. All these numbers are based on a minutes played at defensive midfield under smarterscout’s definitions.
Smarterscout describes their ball retention model as follows:
The model calculates the likelihood of keeping possession when executing a specific action at a specific location on the pitch, then compares that likelihood to individual players’ success in keeping possession in the same situations. The differences between players’ success rates and the average success rates in their leagues are then averaged across all their actions and standardized as above.
This is the metric where Davies struggles the most, and what Silva or any other manager has to consider when playing Davies is how to mitigate this possession loss while he is on the field. Silva’s answer in this post-Gana Everton has been to simply not play him.
If Davies could just hang on to the ball, I think there’s a good player in there. He needs time to develop away from Goodison Park, but there’s a very understandable reason why he isn’t being selected right now. And with Silva’s seat getting hotter by the day, it’s unlikely he’s going to turn to Davies anytime soon.