MANCHESTER, ENGLAND - MARCH 21: Samir Nasri of Manchester City shoots past Petr Cech of Chelsea to score their second goal during the Barclays Premier League match between Manchester City and Chelsea at the Etihad Stadium on March 21, 2012 in Manchester, England. (Photo by Alex Livesey/Getty Images)
Chelsea suffered their first loss of the Roberto di Matteo era when they were downed 2-1 by Manchester City at the Etihad Stadium. This is no embarrassment - no team has come away from their ground with so much as a point in the league (although Napoli did manage to get a draw there in Europe), and before the match a look at the numbers would have given Chelsea a less than 23% chance of not losing there.
That said, Chelsea were 1-0 up with 15 minutes to go, and turned three points into zero in pretty painful fashion. The team has now dropped more more points from winning positions than any other team in the league save Queens Park Rangers, losing like that hurts, no doubt, especially when Arsenal and Tottenham Hotspur gained ground on the Blues in the process. What could have been a season-defining win turned into a horrible, horrible loss. So what happened?
Team Shape And Selection
Roberto di Matteo apparently set his side up in a loose 4-2-3-1 with Frank Lampard and John Obi Mikel in the pivot, Raul Meireles in the hole, Juan Mata wide right, Ramires wide left and Fernando Torres central. I understand the reasoning behind most of the selections. Lampard was superb against Napoli. Mikel's been excellent recently. The Meireles-Torres axis seemed to be pure gold a few days ago. Micah Richard's move to the centre meant that Mata's lack of defensive presence on the left was at least partially mitigated. And, to top it all off, a match that you expect nothing from is a reasonable time to experiment.
In practice, however, the shape wasn't particularly well defined. That might be at least partially by design, but Mata's habit is to cut inside to take up the position nominally occupied by Meireles, Meireles is an attack-minded central midfielder rather than a true trequartista, and Lampard's discipline in the pivot is at least slightly questionable. During the match, it was difficult to tell what the attacking formation was actually supposed to be - the 4-2-3-1 blurred into something approximating a narrow, right-sided 4-3-3, and that lack of coherency probably contributed heavily to Chelsea's inability to do much from open play.
The lack of width wasn't a major problem. Manchester City's fullbacks aren't much of a threat when Richards is shackled to the centre of defence and Adam Johnson, City's only true winger, didn't even make the bench. So congesting the centre was fine, and Mata wide left wasn't an issue. The problem was that the midfield was too heavy on the risk-taking/attack side of things, which isn't a wise way to play at the Etihad. We saw this when Frank Lampard gave the ball to Mario Balotelli in the first half and in Chelsea's poor passing game in the middle third, and I maintain that a diamond with Michael Essien and Ramires in the centre would have been a much more stable base from which to play.
The Fernando Torres vs. Didier Drogba question is a fascinating one. Torres played well against Leicester City on Sunday, while Drogba was superb against Napoli in the Champions League last week. Obviously, those two teams are on completely different levels, so the purely rational analysis has Drogba starting here. Emotionally it's an entirely different question - the need for Torres de-pumpkin himself is acute, and I'm sure di Matteo took the decision to start the Spaniard over Drogba was based with psychological factors in mind.
Realistically, Torres didn't play particularly badly (although after a strong start, he faded pretty significantly) and Drogba was quite poor when he came on as a substitute, but I'm still not convinced that starting Torres was the correct move considering how unlikely it was that City would give him any space to play in. Striking the correct balance between attempting to restore Torres's confidence and fielding the best team possible is a difficult challenge for di Matteo, and I'm glad I don't have to deal with it.
- It was clear that something wasn't right with Branislav Ivanovic from the beginning. The right back, who was superb against Napoli, was a slight doubt before the game with muscle tightness, and he showed that he shouldn't have been on the pitch at all with a woeful (and obviously injury-impacted) display. That di Matteo fielded a hobbled Ivanovic over the presumably-fit Jose Bosingwa is an indication of just what the interim manager thinks of the Portuguese right back. Eventually, however, his hand was forced, and di Matteo had to burn his first substitute to bring Ivanovic off.
Bosingwa's issues are well chronicled, and they're significant enough that there were some serious suggestions to play Michael Essien or Ramires at right back rather than give the job to the actual specialist. It's impossible to know how either would have played there, and I'd suspect that neither would have done a particularly good job there, but we know Bosingwa's essentially useless, and it's time for Chelsea to start looking at other options. Something for the training ground, perhaps.
- Chelsea's goal came off a set piece, which marks the sixth time in the last three games the Blues have scored after being pushed up for a corner or a free kick. David Luiz did excellently to brush Gareth Barry off the ball and Gary Cahill was on hand to squeeze a deflected effort past Joe Hart. It was against the run of play, for sure, but it's important to note that City weren't looking particularly likely to score either. After the first 15 minutes, the defence had been playing superbly.
- Just before the opener, di Matteo made his second switch, bringing Essien in for a badly misfiring Meireles. This was pretty obviously the right thing to do - pushing Lampard higher up and dropping Essien into the pivot makes Chelsea more defensively sound and less prone to losing possession in the middle, but, just like the Oriol Romeu substitution at home against Manchester United, it backfired badly when Essien played poorly and gave away the equaliser via the penalty spot. That doesn't make it a poor decision, of course - sometimes good ideas just don't work out.
- The final Chelsea sub was Drogba for Torres, presumably in an attempt to improve Chelsea's defending on crosses and general ability to hold up play. Despite Torres' hard work in defending, this was also probably the correct call - Drogba's a significantly better defender than Torres is and his style allows Chelsea to relieve pressure with long balls forward far more easily than they can with Torres on the pitch. Again, it didn't work out, and Drogba was poor, but I don't think there was anything wrong with the substitution itself.
- City made several changes throughout the game, all of them interesting to some extent. Balotelli was pulled off for Gareth Barry at half time in a move that only makes sense if you see it as Roberto Mancini punishing his striker for a lackadaisical display, but the other two subs were directly designed to break down Chelsea's back line. Nigel de Jong's withdrawal in favour of Carlos Tevez gave them another quick, tricky attacker who would eventually win the match for City, and switching on Edin Dzeko for David Silva meant that City could then compete with Gary Cahill and David Luiz in the air, where they'd been getting mauled earlier.
I'm not sure I want to give too much credit to Mancini for those changes in turning the match around, because it simply looked like the fight and discipline went out of an otherwise quite good Chelsea side as soon as Sergio Aguero scored that penalty, but I'm fascinated by that final goal - City had set up to attack Chelsea via the air, but were still good enough to play through us on the ground thanks to Tevez. The Blues adjusted to the threat of Dzeko but left themselves open to being torn apart on the ground, and the line was a complete mess by the time Tevez and Samir Nasri were done with it.
It's tempting to see the match as three points dropped, but that's not really how we should be looking at it. Before the game, Chelsea were severe underdogs, and running the numbers gave an expected points total of 0.4. This was the most difficult match left on the domestic slate, and the fact that Chelsea came close to winning is fairly impressive.
That doesn't mean that falling further behind our rivals for fourth place is any easier, however. Chelsea now need to beat Tottenham Hotspur at Stamford Bridge, or this season is toast. Do it. DO IT.