Premier League
November 21, 2016

How much have Chelsea actually improved recently?

Have Chelsea improved dramatically, or is the momentum just flowing right now?


Chelsea narrowly beat relegation threatened Middlesbrough over the weekend to put them top of the league. That result was Chelsea’s sixth consecutive victory while not conceding in the process. These results have coincided with the switch of formation to the 3-4-3. A similar formation that Conte has used to great success in the past.

With those numbers, it’s easy to see why the media have hailed Conte as a ‘genius’ for changing Chelsea’s fortunes. But have Chelsea actually improved that much? And are they as good as the media make them out to be? In my opinion, no, and here’s why.

Before the switch, Chelsea lost to both Liverpool and Arsenal while Conte was completely outcoached by both Wenger and Klopp. Some of the reasons for the change in formation, are the system needing Hazard to do defensive work, Kanté not having a man playing next to him and Diego Costa being isolated up front. In the new system, Hazard has almost no defensive responsibilities, Kanté has a man playing next to him, and the two attacking midfielder in Hazard and Pedro play more attacking to stop Costa from being isolated like he was in the early stages of the season.

Chelsea were improved at Hull but were still getting used to the new system. Against Leicester however, Chelsea were very good and that form continued onto the Manchester United game wherethe Blues won four-nil.

Away to Southampton, another good performance but a showing that was overhyped by the media. Although they won two-nil, that match could’ve quite easily gone the other way had Costa not have scored from over twenty yards out. The Everton performance was also overhyped by the media; Chelsea played brilliantly, but with the chances they created, they probably only deserved to win three-nil at the most.

Another key factor in Chelsea’s change is fortunes, is that they are constantly scoring early goals which put the opposition on the back foot so soon into the match. Against both United and Everton, they were 2-0 up inside twenty minutes without really deserving a two goal lead. A similar thing happened against Leicester and Southampton with Chelsea being 1-0 up inside ten minutes.

With all these early goals going in, it fills the players with confidence to go and score again which happened against Everton when Chelsea scored under two minutes after scoring their first. After the game at Hull, Chelsea were conceding an average of 1.37 goals per pame. Doesn’t look great when you first look at it, but when you look deeper into the analytical data, they were actually defending quite well.

Although the stat above shows they have the eleventh worst defence in the League, their Expected Goals (xG) against were the lowest in the League (0.86). xG measures the quality of a shot based on distance to goal, the angle, if it was a header or not and numerous other factors. It is an analytical model that has a brilliant track record for predicting future results and is a far better indicator of how good a side is playing rather than more well-known terms, such as points, goals, shots or shots on target.

In Chelsea’s first four games (excluding the Liverpool and Arsenal games) they averaged the exact same amount of xGF (xG for) then they did in their most recent six matches (2.01). A similar thing can be said defensively where if you exclude the Swansea game where a couple defensive errors cost Chelsea maximum points, they only conceded slightly more xG then they have recently. (0.47 to 0.50)

Another thing is that they are massively outscoring their xGF. Chelsea have had 12.08 xGF in the last six matches while they have actually scored seventeen goals. They will more than likely not score nearly one and a-half times their xG by the end of the season.

To support my theory that Chelsea weren’t actually that bad at the start of the season, they had conceded eight goals in their first five matches, but they only had 5.65 xGA which probably points towards them being unlucky, rather than bad. In their last six matches however, they have conceded zero actual goals but have had 2.83 xGA recently. This also points towards fortune more than skill and proves the old cliché that football evens itself out over time, true.

With all this in mind, Chelsea have definitely improved of late but nowhere near as much as the media make it out to be. Luck probably wasn’t on their side early on but it definitely has been recently. In my opinion, the 3-4-3 is the best formation for Chelsea. Performances have been improved but they also had some early season results/score lines that probably didn’t show how good they actually were.

About this author

Premier League