Lovren back in top ten overall, but is Per Mertesacker the most underrated footballer in England?

Players must have played at least 300 minutes to qualify.  Efficiency ratings are how likely a positive outcome will occur while that player is either in possession (attacking) or reacting to opposition possession (defensive).

He's baaaaaaaaaaccckkkk.  Yaya Toure retakes the top spot on attack, but only really because Nasri came up 24 minutes short.  Gerrard is Liverpool's highest rated player at 19th.

Attacking efficiency ranking

Y. Toure	0.829
C. Fàbregas	0.814
P. Mertesacker	0.806
M. Schneiderlin	0.792
E. Capoue	0.786
Juan Mata	0.783
F. Coloccini	0.783
S. Distin	0.774
M. Özil	0.770
G. Barry	0.768

Per Mertesacker has sky-rocketed to the top of the defence rankings following a few solid performances.  Lovren is 7th and rising.

Defensive efficiency ranking

P. Mertesacker	0.948
Curtis Davies	0.934
P. Jagielka	0.921
R. Ferdinand	0.920
S. Dann	0.916
S. Distin	0.914
D. Lovren	0.914
M. Demichelis	0.914
John Terry	0.912
F. Coloccini	0.904

Sir Per goes top as Lovren breaks into the top ten overall.

Overall efficiency ranking

P. Mertesacker	0.877
S. Distin	0.844
F. Coloccini	0.843
Y. Toure	0.833
J. O'Shea	0.832
E. Capoue	0.830
M. Schneiderlin	0.830
P. Jagielka	0.828
T. Blackett	0.817
D. Lovren	0.806

Super super Senderos is no longer in the top ten of players most valuable to their club efficiency-wise, but he has so much 'character'.  John O'Shea continues to be a head and shoulders above his Sunderland teammates.

Most valuable to their club efficiency-wise

J. O'Shea	0.197
D. Hammond	0.194
S. N'Zonzi	0.194
R. Ferdinand	0.180
F. Coloccini	0.178
David Jones	0.174
Andy King	0.172
Dean Marney	0.163
P. Mertesacker	0.159
M. Schneiderlin	0.153

Leaders, notes, and squibs:

- Chucky 'Disco Biscuits' Fabregas leads the league in both assists (6) and chances created (18)

- Aaron Ramsey is top with 558 touches, the second most is Capoue with 501

- Eden Hazard has 31 successful dribbles, more than twice as many as any other player, Lamela is second with 14, while Raheem Sterling is 3rd with 13

- Mile Jedinak leads the league with 30 tackles, while the irrepressible Morgan Schniederlin is second with 23

- Curtis Davies leads all players with 62 clearances

Sturridge dethrones Coutinho as the king of inefficiency: Liverpool stat leaders

Chances created

R. Sterling	4
S. Gerrard	3
Henderson	3
P. Coutinho	2
D. Sturridge	2

Passing

Emre Can	10	10	1.000
D. Lovren	119	129	0.922
M. Skrtel	118	130	0.908
A. Moreno	38	42	0.905
S. Gerrard	123	136	0.904

Shots

	sh	sot	shacc	sh%
D. Sturridge	8	3	0.375	0.125
R. Sterling	5	3	0.600	0.200
Henderson	2	0	0.000	0.000
G. Johnson	2	0	0.000	0.000
L. Leiva	1	1	1.000	0.000
R. Lambert	1	1	1.000	0.000

  

Aerial duels

M. Skrtel	9	11	0.818
D. Lovren	8	17	0.471
G. Johnson	3	4	0.750
S. Gerrard	2	3	0.667
J. Manquillo	2	3	0.667

 

Tackles

J. Manquillo	6
Henderson	6
D. Lovren	4
S. Gerrard	4
A. Moreno	4

  

Interceptions

D. Lovren	6
M. Skrtel	6
J. Manquillo	4
Henderson	2
S. Gerrard	2

  

Clearances

M. Skrtel	25
D. Lovren	17
J. Manquillo	5
S. Gerrard	4
L. Leiva	4

  

Attacking efficiency

S. Gerrard	0.797
D. Lovren	0.726
M. Skrtel	0.708
R. Sterling	0.681
J. Allen	0.672

  

Defensive efficiency

M. Skrtel	0.905
D. Lovren	0.896
S. Gerrard	0.842
L. Leiva	0.833
Emre Can	0.706

  

Overall efficiency

S. Gerrard	0.820
D. Lovren	0.811
L. Leiva	0.808
M. Skrtel	0.807
Henderson	0.674

  

Efficiency over team

S. Gerrard	0.135
D. Lovren	0.127
L. Leiva	0.124
M. Skrtel	0.122
Henderson	-0.010

  

So only four players boast a positive EOT, which means they make up for half of the sides' positive results in possession.  If that doesn't improve, Liverpool are in a world of trouble.

Least efficient players overall

D. Sturridge	0.494
P. Coutinho	0.571
G. Johnson	0.622
J. Manquillo	0.624
R. Lambert	0.625

Gerrard breaks into top ten of most efficient EPL players, Lovren falls and Yaya takes top

Top attacking efficient players in the Premiership 
(must have played at least 120 minutes thus far)

	ape
Y. Toure	0.878
S. Nasri	0.859
D. Fletcher	0.825
Cesc Fàbregas	0.822
F. Coloccini	0.817
N. Bentaleb	0.805
Eden Hazard	0.804
S. Gerrard	0.797
E. Adebayor	0.795
T. Blackett	0.795

Fletcher is a bit shocking, but he has posted a 87.3% pass rate and has only averaged 7 caused possession changes a match.  Blackett has managed an even better successful passing percentage at 88.4%.

If all players are included, despite the amount of minutes played, Sergio Aguero is king with a ridiculous 0.938 attacking efficiency.  That's what happens when you score 2 goals, post a 3/4 shot accuracy and are 10/10 on passing in 29 total minutes on the pitch.  Only a matter of time before they play the Argentine, Jovetic, and Dzeko up front at the same time and have the minnows of the Prem soiling their shorts.

Leicester's Andy King is a good example of 'levelling off''.  The Welsh midfielder dropped 22 places in rankings after facing Chelsea, the 2nd most efficient side in the EPL.  The Foxes next host Arsenal, which could go either way for King as he should have the amount of touches to show if he's as good as he was against Everton.

Top defensive efficient players
(must have played at least 120 minutes thus far)

	dpe
C. Chambers	0.914
R. Ferdinand	0.914
M. Skrtel	0.905
D. Lovren	0.896
F. Coloccini	0.894
Y. Toure	0.884
P. Jagielka	0.878
Fernando	0.877
V. Kompany	0.871
T. Blackett	0.865

Chambers continues to impress, but I did change up the formula to be more possession-friendly, meaning the 19-year-old right back benefits greatly from playing for Arsene Wenger.  That shouldn't help Rio though as QPR is the fifth-worst possession side.  Man United should have let him stayed another season.  Skrtel leap frogs Lovren and having both centre backs in the top for says a lot about Liverpool's somewhat deficient defence. Save Gerrard, only Jordan Henderson is  ranked in the top 75 at 74.

Top overall efficient players

(must have played at least 120 minutes thus far)

	ope
Y. Toure	0.881
F. Coloccini	0.856
C. Chambers	0.840
S. Nasri	0.835
T. Blackett	0.830
D. Fletcher	0.829
Fernando	0.825
G. Barry	0.821
S. Gerrard	0.820
N. Bentaleb	0.817

No Chelsea players? Fabregas is closest at 18th, but they are tied with City with a league best 8 players in the top 100.  It is a team sport after all.  Lovren dropped to 13th, one ahead of Skrtel.

Man United the third most efficient attacking side in the Premiership? Afraid so, but...

Most efficient attacking sides in the EPL:

	ape
Man City	0.700
Tottenham	0.698
Man U	0.690
Arsenal	0.674
Chelsea	0.669

City and Tottenham are not surprising here, but ManU at third? Really?  Well two things play into this, first the quality of their opponents.  Home against Swansea and at Sunderland are not very tough 'tests', the latter is eighth best in defensive efficiency so far, but they got a boost from playing West Brom.  The Cardiff club is second to last in defensive efficiency and will probably have to rely more and more on scoring goals to keep up with opponents.

Second, United are actually decent at passing and lead the league in completion percentage at 87.3%. They play at Burnley next week, so expect them to continue their competency, but once the meat of their schedule begins, their numbers should 'level out'.

Liverpool stand 6th, by the way.

Most efficient defensive sides in the EPL:

	dpe
Stoke	0.366
Chelsea	0.350
Arsenal	0.318
West Ham	0.310
Hull City	0.278

Stoke shouldn't be much of a surprise here considering their defence-oriented style of play and the fact that their first two opponents this season are currently the 17th (Hull) and 19th (Aston Villa) most efficient attacks.  Let's see how they fair at City this weekend before we pass judgement.

Liverpool currently sit 11th.

Most efficient overall:

	per
Chelsea	1.019
Arsenal	0.992
Stoke	0.974
Man U	0.943
Tottenham	0.936
Man City	0.933
Liverpool	0.897

Before anyone loses the plot, this measures 'efficiency', there is no 'extra' value given to goals nor to defensive errors that lead to goals.  It will even out as the season continues and opponents become more common.  The fact that Man U isn't at or near the top of the table after two matches against less-than-mediocre opposition says plenty.

Team stat stand-outs

- West Ham are the king of the crosses with a by far league best 19 successful at an also impressive 39.6% rate.  West Brom has completed the second most with 12 and Sunderland is the second most efficient at 29.0%.

- Stoke have been dominant in the air.  The Potters have won the most duels (59) at the highest success rate (65.6%).

- West Brom has been 'dispossessed' 44 times, 15 more than the next nearest club, which is surprisingly Man City.

- Aston Villa has one shot on goal in 15 attempts, it also resulted in their only goal by Andreas Weimann.

- Chelsea lead the league in goals scored, chances created, shots-on-goal, and successful dribbles.

- Stoke and Hull lead the league with 94 clearances a piece, 20 more than any other club.

Steven Gerrard, King of the APEs

Attacking Possession Efficiency that is.  I haven't been posting much lately because I've been working on a lot of stats projects.  One of these is APE.  APE measures how efficient a player is in possession.  It takes in account all positive results from possession and gives relative value to these results and then divides this number by total individual possessions.  A goal is given four times the value of a completed pass and shot on goal, while an assist is given double value to a chance created.  Here is Liverpool's top ten APE from last season (the team APE was 0.43563):

Gerrard	0.46478
Henderson	0.46115
Skrtel	0.45314
Coutinho	0.45302
Suarez	0.45268
Johnson	0.44870
Lucas	0.44444
Sterling	0.43388
Sakho	0.42635
Sturridge	0.42570

Skrtel was more efficient than Suarez? Yes, because he did more with less possession and turned over the ball much less, which is essentially what this stat measures.

I also created a Positive Possession Result Quality, which gives value to the quality of positive result compared to overall 'stripped-down' results. The top ten PPRQ for Liverpool last year (the team average was 0.52788):

Suarez	0.62686
Sturridge	0.60947
Sterling	0.56348
Coutinho	0.54459
Gerrard	0.53748
Henderson	0.52640
Allen	0.51978
Johnson	0.51628
Flanagan	0.51585
Skrtel	0.51147

This basically means that Suarez produced a higher quality result (goal or assist) 62.686% of the time, that shouldn't be much a surprise.  

Individual defensive stats, for the most part, are irrelevant

Recently when I was researching centre back prospects, I started to notice a trend.  It appeared that individualized defensive stats such as tackles, interceptions, clearances, and blocked shots had little relevance in regards to team defensive performance.  Though not completely irrelevant and in the very least a sign of hard work and involvement in a match, it just does not reflect on the scoresheet.

Ranking team defense ultimately comes down to goals allowed, four of the Prem clubs that finished the season in the top five last year, also finished in the top four in goals allowed.  Liverpool was the only top five club not to, they were 8th.  So the key is to figure out what stats are most important in goal prevention, but it's not that easy.  There should be a stats called 'turnover produced' and/or 'chance prevented', but what exactly would that entail and how would it be measured?

There are team stats that already exist that should carry more weight: shots conceded, possession, and passing are the most important in preventing chances for the opposition. These are better described as 'anti-attack' stats instead of defensive ones, but they still strive to accomplish the same objective. There are two interesting samples to discuss from last season's Prem, Man City, who won the league and was 2nd in least goals allowed, and Crystal Palace, who were far and away the best defensive-minded individuals thrown on the pitch falsely called 'a team'.

First, I set up a table taking the rank of each Prem club in each respective stat.  Those stats are goals allowed, non-error goals allowed, shots conceded, tackles, tackle %, clearances, interceptions, blocked shots, defensive errors, critical defensive errors, possession, and passing %.  I then broke that table down into two, one with only non-individual team stats and the other with just individual team stats.

Overall Table

Non-Individual Team Table

Individual Combined Table

Man City finished 7th on the overall table, 1st on non-individual team stats, and dead last (20th) on the individual combined table.  If that's not convincing enough for you, Crystal Palace tied with Chelsea for 1st on the overall table, was 16th on the non-individual table, and finished 1st by a mile on the individual combined table.  In fact, Palace had an avg rank of 2.86 on the ict, a full four points ahead of second place Hull and 11.71 average places higher than Man City.

Crystal Palace finished 11th in the table, scoring only 33 goals (2nd worst overall) and ranked dead last in both passing and possession.  Being that Tony Pulis is still their manager and they haven't brought in any significant attacking transfers, I don't see that changing much next season.  Meanwhile Man City lead the league in goals scored and was 2nd in both passing and possession.  Maybe the old adage 'the best defense is a good offense' rings true.

Sterling or Morata: Who would you take?

StatsBomb, one of my favorite blogs, recently wrote an article claiming that Alvaro Morata is the 'best young prospect in Europe.'  I have to respectfully disagree though, Raheem Sterling is much better and younger.  Here are stats from last season for Sterling and the last two combined for Morata:

	Morata	Sterling
age	21	19
app (sub)	8(27)	24(9)
mins	994	2220
goals	10	9
assists	1	5
chances	18	51
npg90	1	0.36
gcp90	1	0.57
ccp90	1.64	2.06
pass %	66.5	81.7
bp %	48.1	46.8
sacc %	58.8	48.9
shot %	15.7	20
take on %	42.5	64.3
TTI90	18.4	12.1

They play different positions, Morata up front and Sterling on the wing, so you would expect Morata to have a higher non-penalty goal and goal created per 90.  Other than goal scoring and shot accuracy, Sterling is better in every other category.  

He creates more chances per 90, has a significantly better passing rate with a lower back pass percentage, is much more efficient at taking on defenders, and turns over the ball considerably less.  Shot selection is a bit mixed though as Morata has more on target, but Sterling puts a higher percentage in the back of the net.

One stat that I didn't include was aerial duels, for a few reasons.  First, Alvaro Morata is 6'3" and Sterling 5'6".  Second, positioning, Morata plays up front in the middle and obviously gets more crosses fed to him.  Third, these numbers are reflected in shot and passing stats.  Also, it would be one thing if Morata had scored a ton of header goals, but he has one in the last two seasons, and no one I know keeps track of key headers, chances created via header, header pass percentage, nor header shot percentage.  I have a feeling if someone did track the latter, neither Morata nor Sterling would have an impressive rate.

The bottom line is Sterling is the better chance creator.  He can create his own shot for himself and others, where Morata seems to rely on players like Sterling for service.  Morata is a one-toucher, a good one, but still if he doesn't shoot or pass with his first touch, there's a high probability that he's going to lose possession. Of course, if you already have a Sterling, adding a Morata wouldn't be a bad idea.