A thread of Chelsea’s midfield issues – looking at how Chelsea have performed with different combinations and the characteristics of each partnership.

It goes without saying that many other factors – opponents, substitutions, teammates and others – affect performances besides just midfield composition. They have not been accounted for here.

We are going to assume the team’s performance is reflective of its midfield.

Here are the opponents faced by each midfield combination. Both sample size and opponent quality are important. Some combinations have played few games and only against inferior sides.

Anything fewer than 7 games will be considered “small” sample size.

A note: The combination “Jorginho DM” refers to Jorginho playing with two attacking midfielders. This is different from the Jorginho – Kovacic – Kante midfield Chelsea have used in other games.

While Barkley has left the club, Havertz is a suitable replacement.

First up, let’s look at the goal difference and expected goal difference per game with different midfield combinations.

Jorginho and Kante playing with two attacking midfielders has produced excellent results, Kante – Jorginho is promising too.

Looking at shot quantity, the Kante – Kovacic duo look promising while Kante - Jorginho look underwhelming both offensively and defensively.

Kovacic – Jorginho is not ideal but is passable.

While Chelsea do not create a lot of chances with Jorginho and Kovacic, the chances tend to be of reasonably high quality. Defensively they are not great.

Chelsea with Kante – Kovacic or Kante – Jorginho in midfield are above-average defensively but not great offensively.

In terms of passing the ball forward, playing Kante or Jorginho as the lone holding midfielder with more attacking teammates has worked well. Kante – Kovacic is very good too.

It is surprising to see Kovacic – Jorginho rank as low as they do.

Kante – Kovacic have helped Chelsea register some excellent progressive carrying numbers too to go with their passing. Interestingly, Chelsea struggle to progress the ball well when Kante plays with Jorginho.

The “free 8s” setup looks promising in different ways.

Playing Kante or Jorginho deeper allows Chelsea to spend more time in the final third. Also, it allows them to play more in the penalty box.

Surprisingly, Chelsea look very good in this regard with Jorginho – Kovacic – Kante playing together too despite the tough opponents.

This chart shows defensive activity, not quality. Playing Kante and Kovacic – two athletic players who tackle well – has resulted in a high number of tackles.

Playing Jorginho with Kante allows the latter to play his best role, explaining the high defensive numbers.

In order to explain a midfield’s efficiency, the opponent’s performance needs to be looked at as well.

Playing Jorginho as the DM with two AMs looks superb again, but you have to wonder how much of this is down to the opponent quality. I suspect the answer is “a lot”.

In terms of stopping opponent ball progression, the “free 8s” system with Kante and Jorginho looks very impressive.

Playing Gilmour deeper or Kovacic – Jorginho is not good however. If we allow opponents to progress the ball a lot, we will inevitably pay the price.

Kovacic – Jorginho help Chelsea manage possession well, preventing opponents from getting into the final third. However, they are not great at stopping teams from getting into the box once they are in the final third.

Kante – Jorginho is not good in either regard.

That’s the end of the charts. I’ll now try to rank the 7 options we’ve used so far (excluding the unseen combinations Gilmour – Kovacic and Gilmour – Jorginho) in descending order based on metrics, balance and quality.

7 is least impressive, 1 is most impressive.

7. Kovacic – Jorginho: A very superficial midfield. They seem to be doing a lot of things well they do not. They give the impression they are good but both statistically and on the eye test, they are not. Underwhelming in every aspect.

Should not be used except in emergencies.

6. Jorginho – Kovacic – Kante: Lacks balance and we will have to sacrifice an attacker. Should only be used in big games, if needed.

5. Gilmour DM: Has produced good results but the underlying metrics don’t look great. Sample size might be too small to judge.

4. Jorginho DM (2 AMs): The numbers are phenomenal in most aspects. However, the opponents were not great and the one time this midfield played against a challenging team (Sheffield Utd away), the team collapsed.

Can be used in home games against weaker teams.

3. Kante DM (2 AMs): A versatile midfield capable of playing against different opponents.

However, Lampard seems to have discarded this midfield and it cannot be viewed as a serious option unless both Jorginho and Kovacic get injured.

2. Kante – Kovacic: Accommodates our two best midfielders. However, they prefer charging forward at the same time to win the ball back - very risky defensively. Ball progression is good.

Can be used against teams with slow midfielders.

1. Kante – Jorginho: Arguably the most balanced option. Jorginho sits deep while Kante can win the ball back in advanced positions. However, Jorginho’s defensive weaknesses will inevitably be exploited by certain teams.

Can be used against teams that do not press high.

In games where opponents might press us with athletic midfielders, there is no formula. The midfield selection will depend on form and fitness. Keep in mind we do not know how Gilmour will perform.

A specialist defensive mid – be it Rice, Zakaria or Ndidi – will help us a lot.

If you’ve read this far, thank you very much. I hope this thread helps people understand the characteristics of our midfield combinations better.

A *lot* of time and effort went into making this - feedback, RTs and follows will be highly appreciated. :-)