Is possession worth having?
At the start of the season there were two measures by which Villa were dead last in the league and by a long way - possession and shots attempted.
Well, one of those measures has improved. It was 9 games and the end of October before the Lions had more than 50% of possession, against QPR, and it would take until late November before they did it again. However since then Villa have had 50% or more possession in six out of seven games.
But is it worth having?
Here are the Win/Loss statistics for Villa's matches so far:
Games where Villa had possession under 50% - W3, D4, L5
Games where Villa had possession equal to or over 50% - W2, D2, L3
Possession doesn't seem to have any clear correlation to winning or losing. Admittedly Villa have only really dominated possession (i.e. approaching 60%) in two of those games, against QPR and Leicester but even then they lost one and won one. It's pretty clear that possession isn't translating into points on the pitch - but why?
Possession and offense - Low scoring vs REALLY low scoring
For all the improvement in possession, Villa are still dead last in terms of shots on goal with an average of 9.8 shots attempted per game.
Now there are shooting differences between Villa's ´good´ games and their ´bad´ possession games. Where Villa had possession under 50%, they created an average of 6 chances (defined by Squawka as passes that lead to shots on goals - thus excluding shots from interceptions, etc). Where they had possession equal or over 50% they created an average of 9.4 chances a game.
Now clearly on the counter-attack Villa have had some shots that haven't come about from passes as defined by Squawka. So perhaps the Villans score more goals when they don´t concentrate on passing?
Well actually, in terms of goals scored, Villa have scored 4 goals in their 7 games with more than 50% possession. They've scored 5 goals in 12 games with less than 50% possession. That's a major difference - but it's been obscured in the points total by the fact that both numbers are abysmally low.
Possession clearly isn't going to solve Villa's scoring issues - it might lead to a small improvement, but there are other tactical changes to be made if the Claret and Blues are going to get off the bottom of the scoring charts. But it certainly doesn't hurt.
Possession and defence - reducing goalscoring chances
But the true value of possession is to be found in defence.
Villa have allowed an average of 15.75 shots in games where they've had less than 50% of possession. They've allowed an average of 14.14 shots without it.
That difference may seem marginal but there's clearly some kind of positional effect going on as well - Villa have allowed 16 goals in 12 games with less than 50% of possession. They've let in 6 goals in 7 games with more than 50% of possession. In the games they have more possession they're clearly also reducing the quality of goalscoring chances for their opponents.
Villa are being almost entirely buoyed at the moment by this impressive defensive showing and that has been correlated with improved possession - but it's not enough to push them further up the table.
Conclusion - possession is a small part of the puzzle
To a certain extent this kind of possession analysis breaks down due to one simple fact - Villa have more possession where they have played poorer sides. The games where they've let in lots of goals were against better sides where Villa's lack of possession was due to a gap in quality, not a tactical choice. Therefore analysing possession can lead to few meaningful tactical points, apart from ‘avoid playing good sides'.
However that's not quite true. Villa's improved possession, according to Lambert himself has been based on a conscious choice and earlier in the season, Villa were playing a possession-light counter attacking game even against some of the sides that we expect to be around them in the table such as Stoke, Hull, Newcastle and West Ham (before the recent impressive streaks of the latter two). Only against Hull was Villa's performance particularly impressive, the only game in which Paul Lambert successfully used a high press.
As such Villa's improvement in possession has been the main tactical shift this season, and is perhaps the biggest tactical shift we've seen in the last three seasons. And it successfully stopped the rot. Now Lambert needs to show he's got other tricks to push us up the league.
Adam's Formation of the (next) Week
And that therefore brings us to my suggestion for the upcoming match against Crystal Palace. It will be interesting to see in what state the Eagles are after their managerial drama, but it shouldn't necessarily affect Villa's approach to the game.
This is one of those weeks where personnel dictates. Without Delph or Westwood there's a huge issue in linking Villa's defence and attack. With the exception of Joe Cole, no player has managed to link Villa in the centre of midfield, except perhaps Christian Benteke,
As such I'd be keen to see Cole back in the side, but at the same time it's obvious we need more of a threat on the wings. Grealish and Bacuna are the obvious answers, with Sanchez and Cleverley providing a defensive base to shuttle behind them. Up front, Benteke is automatic and I still feel that Weimann deserves the nod over Gabby, being better at holding up the ball.
The last issue is whether Villa should go for 4 at the back, or three at the back. Personally I think that 3 at the back has been working so far and we could do with the extra man in midfield to really pile on the pressure while at home.
So here's my formation of the week...
As ever, let me know what you think in the comments!