A lot of articles have already been written on whether Di Matteo should stick with his favoured 4-2-3-1 formation or revert to the 4-3-3 used regularly by Chelsea over the last few seasons. Rather than offering you a straightforward one-or-the-other decision, I will attempt to provide the analysis that will allow you to make your own decision as to which is the more sensible option. Don't worry though, I will plump for one of the two anyway.
I will say first of all that I am a huge fan of the 4-2-3-1 on paper. I use it when playing FIFA, Football Manager, and if I was Chelsea manager I would ideally be aiming to buy and sell players in order to make this formation a no brainer.
However, I am not convinced that we can get the best out of all our key players by using this system. Why? Two reasons: the pivot and the right wing.
The pivot is the biggest issue with the 4-2-3-1, simply because Chelsea's central midfielders are not best suited to it. In my opinion, you need two different types of players in the pivot: one deep-lying playmaker (a la Pirlo, Alonso, Sahin) and one defensive midfielder who will win/receive the ball and look to give it his pivot partner (a la Fellaini, M'Vila, Vidal). Now, we have Mikel for the defensive option, but he has always seemed more comfortable in a midfield three. As for the deep-lying playmaker, Lampard has been shoehorned in there and has looked far from his best, and Oscar would need time to adjust to such a role.
The right wing is less of an issue, but still an area where I believe Chelsea look unconvincing. Ramires is inconsistent in that position, Moses is arguably not of a high enough quality to be a starter, and neither Mata or Hazard have looked comfortable on that side of the pitch.
So, how could 4-3-3 change these problems? Well it would certainly solve our pivot puzzle. Mikel, with Romeu as backup, would provide the defensive option. Both have looked more comfortable in that role than they have as part of a pivot. As for the two more attack-minded midfielders, Ramires and Oscar/Lampard would find their roles in a midfield three much easier and more natural than in a pivot or in the band of three in the 4-2-3-1. So, a perfect midfield three of Mikel, Ramires and Oscar... What's not to like?
Well, the problem lies further up the pitch in the attacking three. Mata and Hazard, rather than playing their favoured attacking midfield (central or left) roles, would be forced to play on either side of Torres in the front three. As neither has looked comfortable on the right of the band of three, pushing one to the right of a front three would perhaps be counter-productive. In other words, the right wing problem from the 4-2-3-1 remains in the 4-3-3, with the only consolation being that Moses may prefer being part of a front three. Add to this that a certain £50 million striker has looked far better suited in front of the band of three where he gets more service from the creative players (god knows how disappointing he might be if we do revert to the 4-3-3) and there are significant doubts over the front three.
Essentially, Di Matteo will be forced to choose between perfectly accommodating his midfielders or doing the same for his attackers (including Mata and Hazard). In the long run, I still believe 4-2-3-1 is the right option, provided we either nurture Oscar into a deep-lying playmaker or look into purchasing someone who can play there instead. The fact that I am moaning about Ramires and Moses being 'unconvincing' right wing options is testament to the strength of Chelsea's squad (other than the striker position, obviously). Adapting Hazard to a role on the right of the band of three is an option, allowing Oscar to play in the middle and Mata on the left. This would definitely demand the purchase of a new deep-lying playmaker.
Let me know which formation you would go for and why in the comments section. Thanks for reading!