In my last article, I covered the data analysis for the match against Southampton on a big-picture basis; looking at the tactical battle and the main match results of that tactical battle. If you haven't checked out that article and would like to, please click the link below:
Today I will be concentrating on the analysis for our midfield as a unit as well as the individual performances of Britton, Fer, Carroll, and Routledge. With Routledge, keep in mind that we started in the diamond with him as the attacking midfielder but ended up changing to a 4-3-3 with him playing as a wide midfielder.
From my perspective, the weakest section of our midfield was at the top and the bottom of the diamond; Routs and Leon. Neither exerted the influence that we needed them to in order to really give Southampton a go of it. While Leon's passing accuracy was spot on at 96%, he didn't see nearly enough of the ball. Leon had 30 significant touches in his 73 minutes on the pitch, which averages out to 1 significant touch about every two and a half minutes of match time. Over the course of a standard 90 minute match, Leon would have only had about 37 significant touches if his average continued. Now let's compare that to the back-end of last year when we had our stretch of 5 games unbeaten.
In the 2-0 win against Stoke, Leon had 44 touches in 84 minutes, which would have resulted in about 47 touches over 90 minutes. In the 1-1 draw against Manchester United, he had 37 touches in 64 minutes, which equates to 52 touches per 90 minutes. In the 1-0 win against Everton, a match where we only maintained 38% possession, Britton had 34 touches in 72 minutes, or about 43 touches per 90 minutes. In the 2-0 win against Sunderland, Britton notched 52 touches in 76 minutes for about 62 touches per 90 minutes. Lastly, in the final match of the season Britton had 52 touches in 65 minutes or 72 touches per 90 minutes. As you can see, Leon had quite the off day based on his high standards and simply could not make any sort of lasting impression from his ability to control and distribute the ball. It seemed to me as if he wasn't his normal industrious self; he seemed to make himself available as a passing option for a teammate much less often than normal. These are different matches with different possession values, however. So let's scale these values relative to possession similarly to our Passing Frequency metric (see the previous article for the rationale) so we can obtain a more similar comparison.
Southampton: 36.99 touches/90 mins; 40% possession
Possession Influence = 36.99 / 0.4 = 92.47
Stoke: 47.14 touches/90 mins; 45% possession
Possession Influence = 47.14 / 0.45 = 104.76
Man United: 52.03 touches/90 mins; 45% possession
Possession Influence = 52.03 / 0.45 = 115.63
Everton: 42.50 touches/90 mins; 38% possession
Possession Influence = 42.50 / 0.38 = 111.8
Sunderland: 61.58 touches/90 mins; 47% possession
Possession Influence = 61.58 / 0.47 = 131.02
West Brom: 72.00 touches/90 mins; 65% possession
Possession Influence = 72.00 / 0.65 = 110.77
With these values, we can compare the type of influence Leon has had on different matches. As you can see, the Southampton match saw a major drop-off for Leon. In fact, Leon's Possession Influence metric for Southampton was 11.76% lower than the next lowest value (the Stoke match) and 29.42% lower than the highest value (the Sunderland match). We can clearly see here that Leon did not have the sort of influence and impact on the game in terms of activity than we're used to seeing from him.
He also didn't have the defensive contribution that we normally expect from him. In fact, according to WhoScored, he didn't register a single tackle or interception. That doesn't mean he wasn't doing anything defensively; his positional awareness is excellent and helped keep Southampton at bay. However, it does show that Britton never directly contributed to our regaining possession. He didn't timely intercept a pass or stick out a boot at just the right moment as he normally does. The "regista" role that Leon occupies is supposed to act as the main link between defence and the more advanced midfielders while shielding the back four and smartly winning possession. Unfortunately this did not happen against Southampton.
Routledge had similar issues. The number 10 role is supposed to act as a link between midfield and attack while creating opportunities for teammates (or himself). Routledge managed even less touches than Leon did, which should probably be expected considering just how under the cosh we were. He only managed 28 touches in 85 minutes (about a measly 30 per 90 minutes). Unlike Leon though, his passing was not very accurate as he managed a 66.7% passing accuracy. Unfortunately, that's been par for the course for Routledge recently though as he averaged 64.8% passing accuracy last year. He did have that excellent cross that Tammy should have at least put on target, but didn't create any other chances as we never really managed periods of sustained pressure in Southampton's final third. Like Leon, Routs did not manage a single successful tackle or interception. We had both of our midfield "linkers" showing very little influence on the game and were incapable of contributing anything to a very lacklustre pressing effort. That's never a recipe for success, particularly in a system like the diamond which relies so heavily on the midfield presence.
The sides of the diamond, the midfield "shuttlers", actually fared slightly better. Carroll and Fer were both able to exert slightly more influence on the match. Fer had 56 significant touches while Carroll notched 51 while maintaining 87.5% and 84.2% passing accuracy, respectively. They also contributed much more effectively to our efforts at winning possession. Fer actually led the team in successful tackles with 5, while he had 0 interceptions. In fact, only Naughton won more possession for us than Fer (on 6 occasions, 1 tackle 5 interceptions). Carroll, on the other hand, had 2 successful tackles and 2 interceptions.
Why then did the linkers in the diamond struggle so mightily? I believe part of the reason is due to the intensity of Southampton's press. As I mentioned in my tactics analysis article, Southampton's press often caused us to either clear the ball up the pitch in a hurry or to make a few quick short passes and then attempt a hopeful longball. Unfortunately Southampton won most of the 2nd balls which meant we would see cycles of sustained Southampton pressure, a clearance or longball, Southampton quickly winning back possession, and then another sustained period of Southampton pressure. I mentioned in my previous article that we completed the 4th highest number of clearances and attempted the 5th highest number of longballs in the league against Southampton. In fact Fernandez completed the 2nd most clearances in the entire league while Mawson completed the 9th highest amount.
This meant that our hoofing effectively bypassed our linkers by trying to move directly from defence to attack. Due to this we were unable to exert any amount of control on the game. Still, I don't think Routs and Leon are blameless from this match. Both exerted significantly less influence on the match than Carroll and Fer did in comparison and both (or whoever occupies those positions) will need to do better going forward if we're to have a functional midfield.
It'll be interesting to see what Clement decides to do against Manchester United. I could see Mesa getting his competitive debut after Leon failed to impress against Southampton. I certainly don't think Mesa could have made our midfield worse. If Clement expects to be under pressure early and often like against Southampton again, perhaps he chooses to play wingers from the off to provide extra cover out wide and to stretch United. Whatever happens, we will need a much better performance from the centre of the pitch if we're to get anything against Manchester United. Let's hope Clement and the boys are up for it. I'll be back with analysis for the Swansea - United match.