Author Topic: Statistical Comparison of LFC players to Rest of EPL - starting w/LB  (Read 33712 times)

Offline rickardinho1

  • Legacy Fan
  • ******
  • Posts: 17,138
  • The Earth is Flat
Re: An unbiased look at our full backs
« Reply #40 on: July 20, 2017, 12:50:09 am »
This is a phenomenal post, with some good data that is clearly presented and easy to digest. Thank you  :)

I know you've already put in A LOT of effort which I'm grateful for, but if you wouldn't mind I'd love to make a request,... to include some of the better left-backs from previous seasons in a chart like this?

The reason I ask is because i get the feeling that the batch of leftbacks in the PL in 2016/17 was comparatively very weak overall compared to previous seasons, especially compared to 2013/14 when Kolarov, Azpilizueta, Shaw (at Southampton), and Baines all had stand-out seasons. Going back even further you have Ashley Cole, Evra, or a prime-aged Clichy. If you want to include rightbacks you could even throw in Coleman, Bellerin, Glen Johnson, Debuchy, and/or Ivanovic for example.

As I said i don't mean to overburden you, but I feel like including some of those fullbacks (from their good seasons) would offer a clearer picture of where a PL leftback should be in terms of 1v1 stats, as you'd have more of a "standard" to compare to than "just" Clyne.

Offline rickardinho1

  • Legacy Fan
  • ******
  • Posts: 17,138
  • The Earth is Flat
Re: An unbiased look at our full backs
« Reply #41 on: July 20, 2017, 12:56:13 am »

One thing I can't help but think is that players who play in a pressing system will have more opportunities to make tackles, given that they are coached to squeeze the opposition in tandem with teammates, so Robertson could very well see a noticeable increase in the amount of tackles he makes when he plays for Klopp.

Yes, there's always a risk of saying "the player will improve playing for a better team", which isn't always the case (see Downing, Stewart), but in this particular statistical category I would think that it's almost inevitable.... indeed you'd think it would be harder to NOT increase the number of tackles you make when joining a gegenpressing team...

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #42 on: July 20, 2017, 01:08:11 am »


So I have put the circles on the same place in the image as points of reference. Green = good, Yellow = average, Orange = bad. Clyne 2016/17 was the best full back in the premier league too by some distance and was therefore what we considered "Elite".

Firstly, Jose Enrique. I knew he was good in 1-v-1īs but that is exceptional. He was almost 90% of his tackles (highest recorded to date), winning almost 3 tackles per game and conceding less than 1 foul+dribpast per 90 (best on record). Cissokho throws up some elite numbers too. Both their problem was always what happened when they had the ball though. Enrique also had additional problems in the fact he didnīt appear to want to be a footballer anymore and has a thing for the physio room.

Nathaniel Clyne was even better in 2015/16. So he actually regressed last season and was still by some distance the best in the league.

This is where things get interesting. Moreno 2014/15 appears as the back end of the average group. So in the 2015/16 season he regressed badly. Is this because he was more exposed in Klopps system, needed to make even more tackles and as a result ended up getting in trouble more often?

Johnson joins him there which is more or less what I expected.

Then there is Jon Flanagan who just redefined everything that is bad about defending 1-v-1īs. I know his numbers would be bad. I even made a point of highlighting yesterday how his body shape and positioning are both wrong for 1-v-1 defending and I barely understand this beyond a basic level so I am sure an expert could really go to town what he is doing wrong.

Firstly, The average tackles per game for a fullback in the premier league is 2,1. I mocked Moreno and his 4,3 tackles per game suggesting it was like he had a compulsive disorder to make tackles however Flanagan shits all over that number with the batshit insane amount of 5,64 tackles per game. Did someone leave his tackle button on auto and went to take a piss? His average jumps up to 9 (NINE!!!) per game (although small sample size of course) at burnley which explains why he was kept well away from the pitch while there. To put that into perspective, Flanagan made more tackles in his 359 minutes of football than both Cresswell & Olsson combined in their entire seasons. On top of that, he unsurprisingly sets a new record of dribbled past per game which is amazing to be honest because it seems like he just tackles anything that comes within 50 yards of him. I assume they sent a blocker in front of the person with the ball for Flanagan to destroy. Despite this though, Gibbs still somehow comes out of this looking worse. Flanagan only has the 2nd worst tackle accuracy (behind Gibbs, of course) and Gibbs makes more fouls with 1,41 per 90 compared to 1,18 for Flanagan (Pieters and Kolasinac lead the way here with 1,7).

What I draw from this is, we are seriously seriously harsh judges of a full back in terms of 1-v-1īs and probably set targets that literally no other club in the leagues full backs could ever match. Yet somehow, Flanno gets a bye from this overly critical judgement because he is local?

Robertson is coming here with the likes of Clyne & Enrique setting the bar for what is a good 1-v-1 defender whereas at any other club his benchmark would be someone like Monreal, Rose or Clichy. I suspect he will match or surpass all of those but very very few will get near to Clyne & Enrique.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #43 on: July 20, 2017, 01:16:54 am »
One thing I can't help but think is that players who play in a pressing system will have more opportunities to make tackles, given that they are coached to squeeze the opposition in tandem with teammates, so Robertson could very well see a noticeable increase in the amount of tackles he makes when he plays for Klopp.

Yes, there's always a risk of saying "the player will improve playing for a better team", which isn't always the case (see Downing, Stewart), but in this particular statistical category I would think that it's almost inevitable.... indeed you'd think it would be harder to NOT increase the number of tackles you make when joining a gegenpressing team...

I 100% agree with this in theory. However, there is also the fact that at Hull he would be defending 60% of the time and at Liverpool only 40% of the time on average. So now sure if that will balance each other out? Moreno saw an increase in the number of tackles since Klopp came in by about 33%.

It will be something to keep an eye on for sure. It also makes me wonder how many tackles Flanagan could make in a Gegenpressing team? I think 25 per game seems a realistic number. Maybe more if anybody can walk out again for the second half. Opposing teams will need to start equipping their wingers with riot shields.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #44 on: July 20, 2017, 01:20:06 am »
This is a phenomenal post, with some good data that is clearly presented and easy to digest. Thank you  :)

I know you've already put in A LOT of effort which I'm grateful for, but if you wouldn't mind I'd love to make a request,... to include some of the better left-backs from previous seasons in a chart like this?

The reason I ask is because i get the feeling that the batch of leftbacks in the PL in 2016/17 was comparatively very weak overall compared to previous seasons, especially compared to 2013/14 when Kolarov, Azpilizueta, Shaw (at Southampton), and Baines all had stand-out seasons. Going back even further you have Ashley Cole, Evra, or a prime-aged Clichy. If you want to include rightbacks you could even throw in Coleman, Bellerin, Glen Johnson, Debuchy, and/or Ivanovic for example.

As I said i don't mean to overburden you, but I feel like including some of those fullbacks (from their good seasons) would offer a clearer picture of where a PL leftback should be in terms of 1v1 stats, as you'd have more of a "standard" to compare to than "just" Clyne.

I did all Liverpool full backs and it makes quite a difference both good and bad. Look back at end of page one.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline rickardinho1

  • Legacy Fan
  • ******
  • Posts: 17,138
  • The Earth is Flat
Re: An unbiased look at our full backs
« Reply #45 on: July 20, 2017, 01:41:16 am »
I did all Liverpool full backs and it makes quite a difference both good and bad. Look back at end of page one.
Thank you. It does make a huge difference to have some other good 1v1 fullbacks to compare to.

Has Robertson been fairly consistent in his "location" on this graph btw? It would be nice to see his other PL season mapped too. Come to think of it, it might be a useful indicator to see his 1v1 stats in the Championship too given that he played in one of the better sides in the league, and maybe that will give an indication of how much he will tackle for a side that dominates possesison more...

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #46 on: July 20, 2017, 02:10:07 am »
Is tackling is something you do in the middle third of the pitch? I mean, to take a random, arbitrary example of a back four player - Lucas made 1.5 tackles and 1.5 interceptions per game last season, playing mostly as a CB. Compare that to when he played more as a DM in 15/16, when he made 4 tackles and 1.9 interceptions per game. Similarly, you wouldn't see as many tackles made up in the final third of the pitch, and not just because it's mostly occupied by attackers.

My thesis is the number of tackles a player makes is mainly a function of:

1. The primary zone they operate in

a) Middle third players will make more tackles than defensive and attacking thirds
b) The amount of opponent's attacks that go through a player's part of the pitch
c) The amount of responsibility a player has for defending that part of the pitch

2. Their individual characteristics, comprising:

a) Positioning
b) Recovery pace
c) Intelligence (knowing when to stay and when to go)
d) Actual ability to make a tackle

I suspect a reason Moreno has been relatively less successful in his defensive responsibilities, and more successful in his offensive responsibilities, is Coutinho. Coutinho's been our main left-sided attacker for the last few seasons. He's certainly no slouch defensively, but you also won't see him sprinting back to make a tackle near our goal line to help his fullback. I suspect that's because he's neither quick enough, nor has sufficient stamina to do that very much. On the other hand, he's also been our best attacking outlet, which I think is why in 2015/16, there was a table that showed that one our biggest key pass generation combinations was Moreno to Coutinho. Coutinho cutting in with Moreno overlapping is also a natural attacking combination that takes advantage of both players' strengths. This goes to 1c) and also to a lesser extent 1b). Moreno probably had more responsibility covering his part of the pitch than say.. Clyne (Mané) or Clyne (Milner) in the last couple of seasons. Not ideal, when his sense of positioning wasn't good to begin with, and he tended to rely on recovery pace.

I don't have data to support any of this. Someone will probably produce a heat map that shows all the tackles he makes are central and outside our area ;D

Good post mate, itīs certainly something I will look out for during the season.

My gut tells me the players who make the most passes also make the most tackles = center mids and then full backs. My check on this will be James Milner. He played in the same team, for the same coach, in the same league in center mid one season then full back the next.

2015/16 Milner 2.5 tackles per game
2016/17 Milner 2.6 tackles per game

When you look at all players who played 10+ games though, I notice that almost all the top 20 are DMīs , then on page 3 you start seeing full backs and midfielders mixing in too.

I also notice german league full backs make a lot of tackles. Spanish league midfielders make a lot of tackles. Premier League a bit of a mix (lots of german and spanish influence in the tactics of the top sides?).

Some day I need to work out an average per position, per league to compare people to perhaps.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #47 on: July 20, 2017, 02:23:23 am »
Thank you. It does make a huge difference to have some other good 1v1 fullbacks to compare to.

Has Robertson been fairly consistent in his "location" on this graph btw? It would be nice to see his other PL season mapped too. Come to think of it, it might be a useful indicator to see his 1v1 stats in the Championship too given that he played in one of the better sides in the league, and maybe that will give an indication of how much he will tackle for a side that dominates possesison more...

Iīll have a look and see now.

Player Name   Season   League   Tackling Success Rate  Tackles per 90   Fouls per 90   Dribbled Past per 90   1-v-1 Score   
Robertson   2016/17   Premier League      67,92%   1,74   0,59   0,76   1,34
Robertson   2015/16   Championship      60,64%   2,20   0,45   0,87   1,31
Robertson   2014/15   Premier League      62,96%   4,77   0,65   1,77   2,42


Interesting. That kinda indicates that he is getting better every year, which makes sense. Encouragingly his number of tackles was previously way about average and then average and now slightly below average. That indicates he is learning to stay on his feet instead of diving in also. Thatīs very encouraging I think to see some history like that. Will be interested to see what happens with his numbers next season.
« Last Edit: July 20, 2017, 04:33:36 am by BabuYagu »
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline Flinstone

  • Whatever you like. Would like to slip it in without making it awkward... Chinese Information Minister
  • Legacy Fan
  • ******
  • Posts: 3,747
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #48 on: July 20, 2017, 03:03:51 am »
Fantastic stuff Babu,

I'm going to have a proper read as soon as I have some time but a quick question :

Y-Axis : Tackles won per 90 so higher the better
X-Axis : Fouls and Dribbles past per 90 so the lesser the better, ie closer to the origin

By this logic the group including Rose and Milner should be the best, why is it the group including Aanholt and Davies?

PS : Do you mind sharing the data? Cheers!
The West is finished, China is the future

Offline rickardinho1

  • Legacy Fan
  • ******
  • Posts: 17,138
  • The Earth is Flat
Re: An unbiased look at our full backs
« Reply #49 on: July 20, 2017, 03:04:17 am »
Iīll have a look and see now.

Player Name   Season   League   Tackling Success Rate  Tackles per 90   Fouls per 90   Dribbled Past per 90   1-v-1 Score   
Robertson   2016/17   Premier League      67,92%   1,74   0,59   0,76   1,34
Robertson   2015/16   Championship      60,64%   2,20   0,45   0,87   1,31
Robertson   2014/15   Premier League      62,96%   4,77   0,65   1,77   1,42


Interesting. That kinda indicates that he is getting better every year, which makes sense. Encouragingly his number of tackles was previously way about average and then average and now slightly below average. That indicates he is learning to stay on his feet instead of diving in also. Thatīs very encouraging I think to see some history like that. Will be interested to see what happens with his numbers next season.
That is very interesting, and as you say he does seem to be learning with each season. I'm looking forward to seeing if he's shown the same sort of improvement in other statistical areas too :)

Btw, I find it a little odd that his 1v1 score has only marginally improved, because looking at the raw stats it looks like he's improved by quite a lot more than just 1,42 to 1,34...

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #50 on: July 20, 2017, 04:35:30 am »
That is very interesting, and as you say he does seem to be learning with each season. I'm looking forward to seeing if he's shown the same sort of improvement in other statistical areas too :)

Btw, I find it a little odd that his 1v1 score has only marginally improved, because looking at the raw stats it looks like he's improved by quite a lot more than just 1,42 to 1,34...

1-v-1 score is a little misleading as itīs just fouls committed + dribbles past per 90. But I thought that got a bit wordy :D

Good spot btw, the last number is a typo, it was 2,42. One of the problems with manually copying a spreadsheet into a rawk table. I do it this way though so people can copy and paste it into spreadsheets themselves rather than taking screenshots.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #51 on: July 20, 2017, 05:01:30 am »
Fantastic stuff Babu,

I'm going to have a proper read as soon as I have some time but a quick question :

Y-Axis : Tackles won per 90 so higher the better
X-Axis : Fouls and Dribbles past per 90 so the lesser the better, ie closer to the origin

By this logic the group including Rose and Milner should be the best, why is it the group including Aanholt and Davies?

PS : Do you mind sharing the data? Cheers!

The X-axis is inverted actually so the larger numbers are closer to origin.

Tackles is actually an interesting thing in 1-v-1īs.  I was actually coached to play full back and told to pretty much not tackle as then you are committing yourself and doing the wingers job for him. Although that might just have been me that was told that being a big lumbering clutz of a center half who was moved to full back as I had pace and a good cross.

I use tackles as an indicator that perhaps people do not get many 1-v-1 situations (Olsson), dive in too much (Moreno), too Passive maybe (Cresswell). Because itīs a matric based on failure rather than success, it means that absence of proof isnīt proof of absence. A player not getting dribbled past or fouling much could be because they are great at staying on their feet and jockeying, they back off creating bigger problems than trying to engage the attacker, or they simply donīt get into many 1-v-1 situations per game. All three of those scenarios would look identical in terms of numbers and therefore this exercise was about identifying those with a clear weakness and then looking at those who appear strong.

The likes of Enrique, Cissokho and Clyne put in a lot of tackles yet rarely foul or get dribbled by. You could probably mark these down as people good in 1-v-1 situations (although basing opinions solely on stats is just as wrong as watching a player a few times or some highlight reels and thinking you have a good idea what they are like). Everyone else I would have a good look at to see if what the numbers tell me is accurate or misleading. Someone could be winning lots tackles in situations they outnumber an attacker but on the few occassions isolated they foul their man or he goes by them every time. This would show as someone with good tackle numbers and decent at everything else. Of course this is unlikely but not impossible.

Having seen every Liverpool minute of Cissokho, Enrique and Clyne I knew they would come out well. I was surprised quite how well though, but then I think that is just my own caution in rating our own players too highly. Robertson Iīve seen play for Scotland a few times and hull a few times in the Championship and then several games the tail end of last season when Silva came in and they became interesting to watch. Heīs good in 1-v-1īs too. I also thought Moreno was wreckless but his pace bailed him out, Gibbs is shite, Rose was similar to Moreno and, Milner was average. Flanagan was really bad. The stats showed me what I thought they would, although with a few surprises like Milner being more towards bad than average. Gibbs and Moreno being very unique in their issues with 1-v-1s.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline Flinstone

  • Whatever you like. Would like to slip it in without making it awkward... Chinese Information Minister
  • Legacy Fan
  • ******
  • Posts: 3,747
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #52 on: July 20, 2017, 06:23:47 am »
The X-axis is inverted actually so the larger numbers are closer to origin.



Shit, I should''ve seen that. In my defense I am at work lol.

Is that Tableau you're using? Out of curiosity have you thought of superimposing an attacking metric like say crosses and see if it makes players like Danny Rose come out near the top?
The West is finished, China is the future

Offline redmark

  • RAWK Scribe
  • Legacy Fan
  • ******
  • Posts: 21,395
Re: An unbiased look at our full backs
« Reply #53 on: July 20, 2017, 09:02:12 am »


Nathaniel Clyne was even better in 2015/16. So he actually regressed last season and was still by some distance the best in the league.

Did he? I'd have thought this shows that his 2016/17 outcomes were actually slightly better than in 2015/16, despite - or more likely at least partly because of - being significantly less aggressive (reduced tackling numbers).

I suppose you could argue that the marginal improvement in outcomes doesn't quite make up for the more significant drop in tackles; but as you've previously allueded to, the only real 'point' is to improve outcomes.
« Last Edit: July 20, 2017, 09:03:52 am by redmark »
Stop whining : https://spiritofshankly.com/ : https://thefsa.org.uk/join/ : https://reclaimourgame.com/
The focus now should not be on who the owners are, but limits on what owners can do without formal supporter agreement. At all clubs.

Offline Flinstone

  • Whatever you like. Would like to slip it in without making it awkward... Chinese Information Minister
  • Legacy Fan
  • ******
  • Posts: 3,747
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #54 on: July 20, 2017, 09:53:04 am »
Shit, I should''ve seen that. In my defense I am at work lol.

Is that Tableau you're using? Out of curiosity have you thought of superimposing an attacking metric like say crosses and see if it makes players like Danny Rose come out near the top?

Just read everything now. Ignore me!
The West is finished, China is the future

Offline penga

  • What you get if you cross Pingu with Jenga.....
  • Legacy Fan
  • ******
  • Posts: 2,662
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #55 on: July 20, 2017, 10:42:40 am »
The likes of Enrique, Cissokho and Clyne put in a lot of tackles yet rarely foul or get dribbled by. You could probably mark these down as people good in 1-v-1 situations (although basing opinions solely on stats is just as wrong as watching a player a few times or some highlight reels and thinking you have a good idea what they are like). Everyone else I would have a good look at to see if what the numbers tell me is accurate or misleading. Someone could be winning lots tackles in situations they outnumber an attacker but on the few occassions isolated they foul their man or he goes by them every time. This would show as someone with good tackle numbers and decent at everything else. Of course this is unlikely but not impossible.

Having seen every Liverpool minute of Cissokho, Enrique and Clyne I knew they would come out well. I was surprised quite how well though, but then I think that is just my own caution in rating our own players too highly. Robertson Iīve seen play for Scotland a few times and hull a few times in the Championship and then several games the tail end of last season when Silva came in and they became interesting to watch. Heīs good in 1-v-1īs too. I also thought Moreno was wreckless but his pace bailed him out, Gibbs is shite, Rose was similar to Moreno and, Milner was average. Flanagan was really bad. The stats showed me what I thought they would, although with a few surprises like Milner being more towards bad than average. Gibbs and Moreno being very unique in their issues with 1-v-1s.
Some great analysis here. Not surprised at all by Enrique, but was quite surprised Cissokho had such good stats defensively. Also with the little I've seen of Kolasinac he sort of reminds me of Enrique but currently making a lot more fouls and dribbled past more - I'm expecting that to improve. Surprised at van Aanholt as well, doesn't look defensively great the few times I watched him and part of a very weak defence - also has a reputation of not being great there.

Interesting when I look at the Spurs fullbacks though. From watching them play Rose and Walker look generally quite solid - take into account last season where they only conceded 26 goals which is almost half the amount of the season Moreno played and it's hard to see a correlation that they are actually similar in reality (as opposed to just statistically). What I'm trying to say is they really don't look like a problem for Spurs defence and they are 2 out of the regular back 4 (occasionally 5 if 3atb formation is played) whereas Moreno clearly does at times. Also you might remember Mane for example tearing Davies apart time and time again when we played them in the 2nd half of the season but struggled a lot more against Rose in the 1st half of the season. Another question is why such a great coach as Guardiola wanted Walker for 45m if as the stats suggest he might not be elite in defence and also can be frustrating going forward.
« Last Edit: July 20, 2017, 11:37:20 am by penga »

Offline penga

  • What you get if you cross Pingu with Jenga.....
  • Legacy Fan
  • ******
  • Posts: 2,662
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #56 on: July 20, 2017, 11:20:49 am »
Oh and for people who like visuals, this will help give you an idea where this thread is ultimately going and what inspired me to start on this.



These numbers are useful but donīt mean much in terms of game situations. Although itīs nice to see that Robertson has no real weaknesses in this graphic I made and isnīt in the bottom 2 in any category. But perhaps we can create our own visual like this but that has more meaning?
One thing I'm encouraged by is Robertson's take on numbers the success rate is quite high and also the comparative volume to other fullbacks which isn't shown in that graphic. But the aerial part which I've previously mentioned is a concern and you've not shown the volume.



As you can see 0.33 aerials won per game is equivalent to just 10 in the whole season which strongly suggests, like Clyne he often chooses not to compete for them (unless he really thinks he can win it or is forced to in a dangerous situation) and is generally passive, waiting for the 2nd ball but at least wins at a much better rate than Clyne. In stark contrast Elmo playing as a full back in the same system shows how much more superior a fullback can be in this area winning 6 times the amount at a higher rate. Noticeably when we compare Clyne and Milner who also play in the same system and by watching them play, Milly is more willing to throw himself in and compete for the air ball (especially long balls) and though slightly down in % win rate against Robertson I would say he much stronger in the air.

Also just filled in a few more stats in there that I'm not really gonna bother trying to interpret too much there - what I will say though is his key passing or chances created could improve when you look at Elmo playing in the same team and getting more from the other side. Suggests he wasn't really creating too much danger with his deliveries into the box last season. However you look at previous seasons of Robertson including in the championship and there are some better numbers. It is logical he will improve in the area playing for more attack minded team with better players, however I don't think he has the quality to get towards the top end of the league and what Moreno was producing in 2015/16.


Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #57 on: July 20, 2017, 12:37:35 pm »
Shit, I should''ve seen that. In my defense I am at work lol.

Is that Tableau you're using? Out of curiosity have you thought of superimposing an attacking metric like say crosses and see if it makes players like Danny Rose come out near the top?

Yes. Just started using it yesterday so still learning. The experts in this are people like JCB who encouraged me to use it. Iīm going to get through as many categories as possible with the data I have. Some things I have no access to, like the passing metrics as those are made using game event data. This works out things like how much a player advances play with their passing, passes completed into the box, etc

Iīll then rank them in each category and, at the end, try to find a way to show them in terms of all round full back. One problem I would have with what you suggest is that anything that makes Danny Rose, one of the worst in dealing wiht 1-v-1īs, come out on top is wrong. Itīs the same thing that happens with Moreno where the volumn of good blots out the bad and then itīs easy to get the impression that these people are somehow better. I alternatively believe you are only as good as your weakest link. If your full back gets destroyed in 1-v-1īs every game, his crossing being good doesnīt change that. You will need to replace him. This is where we are at with Moreno. His numbers are world class. Unfortunately his problems canīt just be blotted over because he does a lot of things very well. We hope he learns because then we have a potentially world class full back. Yet here we are 4 years on and he seems to be on his way out the door without ever really improving in the one area needed perhaps.

Did he? I'd have thought this shows that his 2016/17 outcomes were actually slightly better than in 2015/16, despite - or more likely at least partly because of - being significantly less aggressive (reduced tackling numbers).

I suppose you could argue that the marginal improvement in outcomes doesn't quite make up for the more significant drop in tackles; but as you've previously allueded to, the only real 'point' is to improve outcomes.

Indeed and this is where interpreting data becomes more important than the data itself. This is how the statement "Stats said Stewart Downing was a great winger" are allowed to be true. Stats never said that. Some idiot did. Stats said he created a lot of chances. But that in isolation means nothing. Did variance play a big part in his numbers? What sort of chances? For who did he create them? Could we replicate that here with the way we play? Is it inflated by set pieces? Would he take those with Gerrard etc here? Can he make space for himself and others?

You go back and look at Downings underlying numbers and he did very little different at Liverpool to Villa when you just look at it from open play. His expected assists and expected goals each season was about the same. The difference was simply variance it seems. Like Ramsey, he had a very purple patch that inflated the numbers one season making people believe he was a lot better than he was. Unfortunately too much importance was placed on one number rather than the bigger picture would was silently screaming "Wait, this is Stewart fucking Downing. Look at these numbers over here. Listen to this alarm ringing too. Donīt do this!" But then analytics of then werenīt what they are today. They could have saved us some expensive, stupid, mistakes if they were. I still think we would have access to the numbers to get that decision right though, we just seemed to place a high importance on chances created as we signed the top 3 in the league who werenīt already at a top 4 club.

As for why this is relevant for Clyne - I am trying to take more of a bigger picture look at this. Let me show you side by side.

Player Name   Season   League   Tackling Success Rate  Tackles Won per 90   Fouls per 90   Dribbled Past per 90   1-v-1 Score   
Clyne   2016/17   Premier League      76,25%   1,65   0,38   0,62   1,00
Clyne   2015/16   Premier League      84,00%   2,55   0,58   0,48   1,06


For me here, I am trying to avoid a Stewart Downing moment in placing too much importance on one number. In 2015/16 Clyne's tackle success was insane. He also made a lot more tackles too so he wasnīt just diving in. He actually lost more tackles in 2016/17 despite making far fewer. He also got dribbled past more in 2016/17 (which I probably value highest of all the numbers if I am honest). Fouls go up slightly more than dribbled past goes down though, so he ends up very slightly worse in terms of his 1-v-1 score but overall I think he 2015/16 numbers look better. But thatīs just my interpretation of them. Itīs not right or wrong. The end result is 2015/16 Clyne failed 0,06 times more per game to adequately deal with 1-v-1īs so maybe that is right? I guess thatīs one for Klopp to decide. Does he like more aggressive Clyne or not? :D

I like this actually it shows a correlation for one player that when he is more aggressive and tackling more, his fouls go up but he gets dribbled past less. That makes sense. Although is you start tackling more to the point you are just aimlessly diving in a lot like Flanno & Moreno then dribbled past will start going up too as full backs will scout you and know they donīt need to do anything. Just show you a taste of the ball then let you over commit and skip past you while laughing maniacally.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline redmark

  • RAWK Scribe
  • Legacy Fan
  • ******
  • Posts: 21,395
Re: An unbiased look at our full backs
« Reply #58 on: July 20, 2017, 12:46:45 pm »

Indeed and this is where interpreting data becomes more important than the data itself. This is how the statement "Stats said Stewart Downing was a great winger" are allowed to be true. Stats never said that. Some idiot did. Stats said he created a lot of chances. But that in isolation means nothing. Did variance play a big part in his numbers? What sort of chances? For who did he create them? Could we replicate that here with the way we play? Is it inflated by set pieces? Would he take those with Gerrard etc here? Can he make space for himself and others?

I like this actually it shows a correlation for one player that when he is more aggressive and tackling more, his fouls go up but he gets dribbled past less. That makes sense. Although is you start tackling more to the point you are just aimlessly diving in a lot like Flanno & Moreno then dribbled past will start going up too as full backs will scout you and know they donīt need to do anything. Just show you a taste of the ball then let you over commit and skip past you while laughing maniacally.

This is all fantastic stuff and I'm only getting to post the odd thought while at work. Would love to get hold of the data and play around with it, weighting certain 'traits' over others, etc.

I wonder how much 2015/16 Clyne led to 2016/17 Clyne?

Let's say clubs now do put a fair bit of effort into data analysis as well as scouting - and last season, as a result of that analysis opposing managers said to their players - "Don't bother taking on Clyne. Switch to the right and attack Milner". Even if just a few clubs did it, or even if it only impacted a couple of attacking decisions during games, that could lead to the sort of differences we see between 2015/16 and 2016/17 Clyne.

Do we have stats for the zonal distribution of weight of attacks we faced?
Stop whining : https://spiritofshankly.com/ : https://thefsa.org.uk/join/ : https://reclaimourgame.com/
The focus now should not be on who the owners are, but limits on what owners can do without formal supporter agreement. At all clubs.

Offline Rush 82

  • Seth Iffricans don't take the dog out for a walk - they take the line out!
  • Lead Matchday Commentator
  • Legacy Fan
  • ******
  • Posts: 22,214
  • From Cape Town to Anfield
Re: An unbiased look at our full backs
« Reply #59 on: July 20, 2017, 12:47:04 pm »
.


A genuinely valuable contribution - kudos to you sir.


Any chance that you're thinking of doing this for center backs? (leaving price out of it the equation, I'd love to see what the stats say about VVD's relative position in the Prem)
« Last Edit: July 20, 2017, 12:48:50 pm by Rush 82 »

Offline Klippity Klopp

  • LFC. For life! Yes! Not just for Kloppmas.....nooo.....
  • Legacy Fan
  • ******
  • Posts: 3,438
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #60 on: July 20, 2017, 12:50:20 pm »
Also you might remember Mane for example tearing Davies apart time and time again when we played them in the 2nd half of the season but struggled a lot more against Rose in the 1st half of the season.

I thought Mane destroyed Spurs in both games.

Rose had a nightmare trying to deal with him and in the end would resort to fouling him, pretending he was injured & trying to get Mane sent off.  As that was the only way he seemed to be able to deal with him - Mane off the field.

Can't wait for him to be back. 

Very interesting thread, great work Babu. 

I don't have much to add, other than I am not really surprised by the results really.  I always thought Enrique & Cissokho were very good defensively.  Clyne I don't think there is much doubt has been very strong in that area and I thought he was arguably our player of the season two years ago.  Moreno & Flanagan, rash - well we knew that too.  I guess Milner looks to be a steady eddie.  Not particularly great at any category, but also not a liability.  If Robertson can enhance that position, then that would be great. 

I feel defensively a full back has usually done his job when you don't notice him too much.  Guess that goes in line with the argument that tackling is a last resort.  Saw Klopp saying full backs shouldn't really be isolated and be in 1 vs 1 situations too often, which might be what he values the least, given he likes Milner and that's his weakest area?

Offline Flinstone

  • Whatever you like. Would like to slip it in without making it awkward... Chinese Information Minister
  • Legacy Fan
  • ******
  • Posts: 3,747
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #61 on: July 20, 2017, 01:10:38 pm »

One problem I would have with what you suggest is that anything that makes Danny Rose, one of the worst in dealing wiht 1-v-1īs, come out on top is wrong. Itīs the same thing that happens with Moreno where the volumn of good blots out the bad and then itīs easy to get the impression that these people are somehow better. I alternatively believe you are only as good as your weakest link. If your full back gets destroyed in 1-v-1īs every game, his crossing being good doesnīt change that. You will need to replace him. This is where we are at with Moreno. His numbers are world class. Unfortunately his problems canīt just be blotted over because he does a lot of things very well. We hope he learns because then we have a potentially world class full back. Yet here we are 4 years on and he seems to be on his way out the door without ever really improving in the one area needed perhaps.


That's totally fair. My personal opinion is that if there was any full-back Klopp could pick from the premiership it would be Rose.

I was just interested in seeing whether those clusters start to look a bit different if we start bringing attacking numbers into the picture (Whilst keeping some defending metrics). If we get the numbers of Marcelo and say Alaba and the reasoning is showing that those two are in the top cluster then we would know the build is robust enough to draw conclusions.

Are you able to share the raw data or did you get it with an NDA?
The West is finished, China is the future

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #62 on: July 20, 2017, 01:32:43 pm »
Some great analysis here. Not surprised at all by Enrique, but was quite surprised Cissokho had such good stats defensively. Also with the little I've seen of Kolasinac he sort of reminds me of Enrique but currently making a lot more fouls and dribbled past more - I'm expecting that to improve. Surprised at van Aanholt as well, doesn't look defensively great the few times I watched him and part of a very weak defence - also has a reputation of not being great there.
I was one of those championing for Cissokho over Flanagan that season. Our defence was what hurt us that season and having the more steady option there seemed sensible to me. Plus if all else is equal, give me a left footer there for balance over a fight footer who narrows play.

As for van Aanholt, I checked him out last night and noticed Kent stood him up twice and twice failed to get past him. There were other time he cut in on his left and Aanholt tracked him and then Kent got a shot off and almost scored. But thatīs a different thing to 1-v-1 scoring and actually what I plan to look at next. Thatīs the first time Iīve paid attention to his 1-v-1 abilities and he did okay v Kent who, Iīll be honest, looks a shit to deal with in 1-v-1s. His feet are fast, he is two footed, he has fantastic balance and agility. He will fuck up some full backs for us this year for sure. Maybe just not van Aanholt. I do also have a feeling for the side he is in, van Aanholt is quite passive. His numbers should be higher than Clyne, not lower. So that is something I am looking for. Is he just standing off peopel and letting them get shots, crosses and dangerous passes around him easily? Kent did that after failing to beat him twice. So the gut says yes, but I need to look at it properly later today.

Interesting when I look at the Spurs fullbacks though. From watching them play Rose and Walker look generally quite solid - take into account last season where they only conceded 26 goals which is almost half the amount of the season Moreno played and it's hard to see a correlation that they are actually similar in reality (as opposed to just statistically). What I'm trying to say is they really don't look like a problem for Spurs defence and they are 2 out of the regular back 4 (occasionally 5 if 3atb formation is played) whereas Moreno clearly does at times. Also you might remember Mane for example tearing Davies apart time and time again when we played them in the 2nd half of the season but struggled a lot more against Rose in the 1st half of the season. Another question is why such a great coach as Guardiola wanted Walker for 45m if as the stats suggest he might not be elite in defence and also can be frustrating going forward.

This is where expected goals comes into play. Spurs gave up more shots on target than Liverpool last season. The expected goals value of those shots was slightly higher than Liverpool. Yet Liverpool conceded more goals. From those numbers nothing says to me Spurs Defence > Liverpool defence. For me itīs either equal or Liverpool a scratch better.

However, if instead you used results based logic (which I hate), and say Spurs gave up 26 goals, Liverpool 42. Then those numbers say to me Spurs Defence >>> Liverpool defence. That canīt be right though. How can both those numbers be true?

My Answer? Lloris >>>>>>>> Mignolet (+ Karius for 10 games). I think our goalkeeping is swinging what would be title winning shots against numbers into mid table goals against. If you look at shot locations, we give up goals from distance at above the going rate. Why? We give up goals from angles WAY above the going rate. Why? A guy whose stuff I follows believe itīs a weakness of Migs for a while now and he makes a compelling argument for it. If you imagine your positioning if poor for angled shots, you can see how just being 1/2 yard off would result in people being able to get the ball into the far post much easier. It might be something we donīt notice but the stats say itīs happening. Every single year. Regardless of the players in defence, the system, the manager, every year we concede too many from distance and from angles. Like, 100% too much.

Here is the most interesting thing. When compared to averages, Karius crushes it from angles and from long range. He outperforms the averages by almost 50%. He outperforms expected goals consistently (last year was he first he didnīt but then young keeper, new league, teething problems?). His distribution numbers were among the best around too. And heīs young. So it seems like someone at the club looked at all our specific issues and went out and found the best young keeper in terms of outperforming xg, distribution and dealing with shots from angles and long range. Seems he is shite at crosses but then I think most goalkeepers bar the truly world class will have something they struggle with. Plus he can improve. Migs has a lot here with them (ignore that flap last night, he has). I just went back and watched Karius goals against last season in the league and none of them were from the angles that Migs struggled with. One from range was a Payet freekick that we wouldnīt have stopped if Migs and Ward was in goal with him. In fact watching all the Karius goals, he would struggle to stop any of the shots. His problem was more poor goalkeeping that resulted in goals rather than his ability to position himself to reach shots that he should reach.

As for the question about full backs at Spurs though. Imagine for a second you have full backs who struggle on 1-v-1īs, what would be a hack for solving that problem other than just replacing them? 3 at the back. But they only played 3 at the back for 1/3 of their games, what about the rest? First, get center halfs that are good at covering full backs and ideally played full back themselves (both Vertonghen & Alderweireld played full back at Ajax). The second thing you can do is play with 2 DMīs similar to how Benitez did with Masch and Xabi. Did you ever notice how much time they spent helping out the full backs and covering for them?

So this is exactly what Spurs did last season. Their go to formations were either 4-2-3-1 or 3-4-2-1 every game. Dier and Wanyama were the DMīs who saw the most minutes too. Midfielders who are far more D than M I think. Spurs then went with VERY attacking full backs with insane recovery pace knowing that in their DMs and CBs they had players who could cover them while they were up the pitch and also support them well in 1-v-1 situations.

Your Mane question? I think Mane is most effective against full backs he can destroy with pace and power. Rose he cannot. Davies he can. But you put Rose (or Walker) up against someone like Coutinho and itīs a bloodbath.

Guardiola doesnīt sign good defensive full backs. His #1 priority is always recovery pace. After that he looks at them with the ball. Then he looks at defence. In many ways Walker offers Guardiola what he wants (but not everything he needs). Ted Knutsen said last season that Man City wouldnīt win anything with a bunch of old full backs. He was right too. Recovery pace at full back is so important to Guardiola and how he plays (us too, btw, which is why Milner at 31 will be a problem sooner rather than later). Walkerīs pace and ability to hit hard right footed crosses first time mean he will naturally stretch opponents tactically as you cannot give him 15 yards on the right flank to hurt you with. Walker isnīt great through. Heīs no Dani Alves in terms of technique and dribbling. Heīs just average/above average at lots of things and his pace takes that up a level above his underlying skill set. Milner with Walkerīs pace would likely be a beast. Moreno with Milnerīs pace wouldnīt even be a professional footballer perhaps.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #63 on: July 20, 2017, 01:45:03 pm »
This is all fantastic stuff and I'm only getting to post the odd thought while at work. Would love to get hold of the data and play around with it, weighting certain 'traits' over others, etc.

I wonder how much 2015/16 Clyne led to 2016/17 Clyne?

Let's say clubs now do put a fair bit of effort into data analysis as well as scouting - and last season, as a result of that analysis opposing managers said to their players - "Don't bother taking on Clyne. Switch to the right and attack Milner". Even if just a few clubs did it, or even if it only impacted a couple of attacking decisions during games, that could lead to the sort of differences we see between 2015/16 and 2016/17 Clyne.

Do we have stats for the zonal distribution of weight of attacks we faced?


I fucking WISH I had that :D I suppose you could go on somewhere like whoscored.com or anywhere that has heatmaps and look at the volume of attacking play we face game by game on each flank. But yes, if I was a coach I would fuck off the left flank and just destroy down the right against Milner/Moreno/whoever. I think when we had Enrique the opposite happened too. I think rather than taking on Enrique itīs better to give him the ball. He will fuck up enough to create those chances for you if you do :D

Our current head of Data Analysis is Ian Graham who holds several degrees in physics I believe and has been guest speaker at Barcelona and other top clubs about the subject of analytics in football. I also know until recently he had his own blog which wasnīt much viewed and has subsequently been taken down. I know this because I read an old article last season which referenced his "pioneering work" and the name looked familiar to me, but I couldnīt track his work down. I think part of his contract with Liverpool involves anything he produces as intellectual property and now nothing of his is available online. So for sure... we take this seriously and have one of the very very best in the business overseeing it.

Statsbomb talk about that a lot too, those "edges" you get from using analytics. For example, if you know that van Aanholt is hard to beat as he doesnīt get very tight but gives you too much space to shoot/pass/cross - do you even try to take him on then or just destroy him with the space he gives you? I noticed last night Kent twice tried to take him on, and failed. Third time he went the opposite way, tracked by van Aanholt, out and across the box and then fired one at goal that almost crept in the near post. Lesson learned Mr Kent - donīt take him on, just make a little space and heīll let you cross or shoot at will.

EDIT:- Found this article on Opta site about Ian Graham. He was working on analytics for 7 years before Liverpool recruited him in a "newly created role". So perhaps our analytical work started with his appointment in 2012?
« Last Edit: July 20, 2017, 01:47:47 pm by BabuYagu »
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline Suareznumber7

  • Gullible. Lost in the modern world, thinks all tweets are true.
  • Legacy Fan
  • ******
  • Posts: 10,958
Re: An unbiased look at our full backs
« Reply #64 on: July 20, 2017, 01:46:34 pm »
Perhaps the best thread on the site at the moment.  Great job and really interesting.

Quick question, where do you get all this data from? 

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #65 on: July 20, 2017, 01:50:25 pm »
A genuinely valuable contribution - kudos to you sir.


Any chance that you're thinking of doing this for center backs? (leaving price out of it the equation, I'd love to see what the stats say about VVD's relative position in the Prem)

Iīll give it a go. Ideally I want to do it for every position so that, at the end, I have something to plug numbers into to make like-for-like comparisions. So the next time someone suggests we sign Keiran Gibbs I can put that shit down in seconds rather than needing to go away and take a proper look at it by which time the conversation has moved on.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #66 on: July 20, 2017, 01:53:38 pm »
Perhaps the best thread on the site at the moment.  Great job and really interesting.

Quick question, where do you get all this data from?

Iīm working on something with JCB that we will share with you all, some opta data we got hold of for last season. Unfortunately Iīm not sure we can get before 2016/17 or going forward either.

However most if not all that I am using here is publically available by just collating numbers from all those sites like whoscored.com, squawka, etc

When everything is organised and ready, I hope to start a LFC Analytics thread with JCBīs help and weīll share all the data and tools we have (and hopefully between us all we can find more) and then see what we can all come up with in there using it.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #67 on: July 20, 2017, 01:56:06 pm »
That's totally fair. My personal opinion is that if there was any full-back Klopp could pick from the premiership it would be Rose.

I was just interested in seeing whether those clusters start to look a bit different if we start bringing attacking numbers into the picture (Whilst keeping some defending metrics). If we get the numbers of Marcelo and say Alaba and the reasoning is showing that those two are in the top cluster then we would know the build is robust enough to draw conclusions.

Are you able to share the raw data or did you get it with an NDA?

Answered above this one. Yes, working on it.

As for Rose, Klopps full backs always seemed steady in defence before anything else. Even the ones who were converted wingers like Rose. Itīs an interesting question though, guess we will see what Rose looks like when we are done. I do think, like Guardiola, recovery pace is everything for Klopp. If you donīt have it, you cannot play full back for him. Everything else seems negotiable.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline Legoland

  • Plazzie Brick Lid :)
  • Kopite
  • *****
  • Posts: 584
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #68 on: July 20, 2017, 02:02:36 pm »
A genuinely valuable contribution - kudos to you sir.


Any chance that you're thinking of doing this for center backs? (leaving price out of it the equation, I'd love to see what the stats say about VVD's relative position in the Prem)

It's funny how you quoted a full stop and said genuinely valuable contribution  ;D ;D

Offline Legoland

  • Plazzie Brick Lid :)
  • Kopite
  • *****
  • Posts: 584
  • We all Live in a Red and White Kop
Re: An unbiased look at our full backs
« Reply #69 on: July 20, 2017, 02:04:09 pm »
So our options are:

RB - Clyne and TAA
LB - Milner and Robert

Jon Flanagan to be third choice for both? or possibly compete for the LB spot?

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #70 on: July 20, 2017, 02:06:27 pm »
/snip

I donīt want to get into aerials or the other things yet because the discussion is still focused on 1-v-1s. What I would say is that in a low block, the full backs are nearly always close to the center backs who will take charge of crosses coming in rather than leave it to a full back. Also we play higher up the pitch and sides try to play over us into the space behind the full backs. Some of those balls are telegraphed resulting in headers. That wonīt happen to Hull either. Then you have role at set pieces. Being one of the shortest in the Hull team his job my be to prevent short corners or take a post, whereas Elmohamedy as a 6 footer will have a job requiring aerial duels more. Likewise he may be in the box for corners while Robertson is taking them or covering the counter attacks. Maybe the CB on the left side is more aerially dominant to help out Robertson while the right cb knows elmohamedy can hold his own? There could be a tactical answer for low numbers but I guess we are back where I was when talking about Goulam, none of this is to make decisions from. Just theories. Ones that need to be tested. Iīve seen Robinson a lot and never seen him avoid aerial challenges that he should be going for. But then was I really paying attention?

Basically.... I think the system & tactics comes into play a lot more here so itīs hard to compare. When I look at the numbers and share it with you all, then we can properly debate it with 20+ comparative full backs to work with.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline Rush 82

  • Seth Iffricans don't take the dog out for a walk - they take the line out!
  • Lead Matchday Commentator
  • Legacy Fan
  • ******
  • Posts: 22,214
  • From Cape Town to Anfield
Re: An unbiased look at our full backs
« Reply #71 on: July 20, 2017, 02:21:54 pm »
It's funny how you quoted a full stop and said genuinely valuable contribution  ;D ;D
:lmao - yeah, snipped the quote so that it's easier to follow on a smartphone

Offline Rush 82

  • Seth Iffricans don't take the dog out for a walk - they take the line out!
  • Lead Matchday Commentator
  • Legacy Fan
  • ******
  • Posts: 22,214
  • From Cape Town to Anfield
Re: An unbiased look at our full backs
« Reply #72 on: July 20, 2017, 02:22:55 pm »
So our options are:

RB - Clyne and TAA
LB - Milner and Robert

Jon Flanagan to be third choice for both? or possibly compete for the LB spot?
Based on the graphs, Flanagan should be sold, preferably to a team who don't know about this thread

Offline redmark

  • RAWK Scribe
  • Legacy Fan
  • ******
  • Posts: 21,395
Re: An unbiased look at our full backs
« Reply #73 on: July 20, 2017, 02:35:33 pm »
Based on the graphs, Flanagan should be sold, preferably to a team who don't know about this thread

Or watched the friendly last night. Particularly after having read this discussion prior to that point, it was a painful watch.
Stop whining : https://spiritofshankly.com/ : https://thefsa.org.uk/join/ : https://reclaimourgame.com/
The focus now should not be on who the owners are, but limits on what owners can do without formal supporter agreement. At all clubs.

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Full back Analysis part 2 - Recovery Runs
« Reply #74 on: July 20, 2017, 02:35:45 pm »
So as I said in the opening post, 1-v-1 situations is just 1 part of being a full back. Depending on your system it can be vitally important (Liverpool, hence why I did it first) or not very important at all (Chelsea). So we shouldnīt place too much value on just that one category, nor take any of it as fact either as itīs just a theory we are working on that we need to prove, as interesting as it may be.

Next up is 2. Recovery Runs.
This may even be more important than 1-v-1īs to be honest with you. Every full back Guardiola, Pochettino or Klopp signs will have this in spades. When your starting position is so high, especially now at Liverpool with our CBīs and only 1 deep midfielder rather than the usual two to cover full backs, then your full backs need to get back the second attacks break down (unless the move breaks down near them and they need to press while the #6 will likely cover in behind).

Unfortunately I have no data at all to help you here. But in analysing full backs part of the job was always to look as much, if not more, than crunch the numbers. The key things you are looking for here is turn speed, acceleration from a standing start, straight line speed, and most important of all, the determination to make those recovery runs. You can be the fastest player in the world but if you just jog back when your team loses the ball itīs useless (and you are a dick). Believe it or not, studies show that the far far majority of players get forward much faster than getting back (so they are all dicks?). However there are some players that tick all the boxes in terms of physical ability to do the task and have the desire to get back to prevent goals. Moreno is the best example that we all know. Insane pace in every way measurable. Agility too. Iīve also never seen a player who is so determined to get back as him. Itīs one of the reasons I root for him. He really really does try. Of course most of the time he will be running back to fix his own mistakes ;) :D Another who makes those runs is Salah. Youīll see.

So this is something you will all just have to work out amongst yourselves and you will probably need to watch full matches to get a feel for it as clips is just people showing you what they want you to see. It could be the one recovery run a player makes in his career you are watching. Also you are not looking at JUST straight line speed. I saw a clip to show Skrtel isnīt slow which was basically skrtel, from a running start, running fast past a bunch of players and getting faster too. That just tells me that once Skrtel gets going, heīs not slow. But how quick can he change direction? Can he accelerate from a standing start? How determined is he to run backwards? Actually itīs Skrtel, we know he has no trouble running backwards ;) But anyway, you get the point hopefully.

So in summary this will be the category where we will each need to go out and use Mark I eyeball. I donīt have access to any data to help with this and, even if I did, I donīt like the information we usually get showing someones top speed as that regularly indicates things like Ilori being faster than Ronaldo, Heskey being faster than Owen. It doesnīt really take into account anticipation, reaction times, turning speed, acceleration, agility etc. Plus, some players have what they need to make recovery runs at pace and simply chose not to *cough, Keiran Gibbs, cough*. So that *giving a fuck* quality is vitally important.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #75 on: July 20, 2017, 02:37:54 pm »
Or watched the friendly last night. Particularly after having read this discussion prior to that point, it was a painful watch.

If you want to check something else we learned from this thread, look at Milner diving in a few times just after coming on. First time he gets beaten pretty easily. Second time he takes someone out. Third time he wins the ball at full stretch. All that was within a small 5-10 minute window at the start of the second half but did illustrate that he does tend to just dive in. Sometimes it works, twice as often it does not.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #76 on: July 20, 2017, 02:38:21 pm »
Based on the graphs, Flanagan should be sold, preferably to a team who don't know about this thread

:D
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
3. Preventing shots/crosses/passes into the box.
« Reply #77 on: July 20, 2017, 02:53:12 pm »
Next up is 3. Preventing shots/crosses/passes into the box.
To steal a line from Grobellaar, we have more trouble with crosses than Dracula! A lot of criticism falls on our keepers for this but I think as fans we have a habit at looking at the last mistake, labelling it THE mistake, and moving on. That probably comes from sub-par analysis by pundits on shows like MotD for years. Iīve always believed that preventing crosses completely is the safest way to go. If a ball is coming into your 6 yard box you already have a problem. Liverpool are one of the best in the league in preventing passes into their box. As a result we are also the best in the league in terms of shots against in the box. Two problems we have that seem to cause us trouble is that our goalkeepers are not stopping shots from angles (or long range) at an acceptable rate, and far too many of the chances created against us are from crosses. If we work so hard in preventing people passing the ball into our box, surely our full backs should be working hard to prevent crosses too. Simply not letting the ball get into the most dangerous area of the pitch and then rolling the dice on trying to win it seems a high risk strategy. So letīs find someone who says "fuck that, Iīm preventing this shit from happening".

Going to go and compile all the data on this and then start looking at full backs who appear good at this. I expect we might see some inverse correlation here between being difficult to dribble past (not getting tight) and being easy to play around (standing off too much). If I am right, those people I highlighted as passive will do badly here (e.g. Cresswell, Olsson, van Aanholt) particularly with crosses. Those who I highlighted as too aggressive (e.g. Flanagan, Milner, Moreno, Chilwell, Pieters) should come out of this pretty well.

Give me a bit of time and Iīll see where we are at with this and if my predictions come out as I expect (hope) they will.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd

Offline redmark

  • RAWK Scribe
  • Legacy Fan
  • ******
  • Posts: 21,395
Re: An unbiased look at our full backs
« Reply #78 on: July 20, 2017, 02:58:56 pm »
If you want to check something else we learned from this thread, look at Milner diving in a few times just after coming on. First time he gets beaten pretty easily. Second time he takes someone out. Third time he wins the ball at full stretch. All that was within a small 5-10 minute window at the start of the second half but did illustrate that he does tend to just dive in. Sometimes it works, twice as often it does not.

There was a couple of minutes spell also where Moreno got ran at twice - first time he ran alongside, but clearly looking for an opportunity to get his foot in rather than jockey: he tripped over his own feet and ended up on his arse. Second time, he went to ground and whacked it into touch.
Stop whining : https://spiritofshankly.com/ : https://thefsa.org.uk/join/ : https://reclaimourgame.com/
The focus now should not be on who the owners are, but limits on what owners can do without formal supporter agreement. At all clubs.

Offline BabuYagu

  • It's Portuguese for 'BabyYoghurt'. The John Motson of RAWK. Or Barry Davies. Or Charley Boorman, even. Expertly silent fist-pumper. Needs to pay more attention. Repeatly analing goalkeepers.
  • Legacy Fan
  • ******
  • Posts: 17,350
  • wakelet.com/@BabuYagu
    • Wakelet of the Articles I have written
Re: An unbiased look at our full backs
« Reply #79 on: July 20, 2017, 03:09:37 pm »
There was a couple of minutes spell also where Moreno got ran at twice - first time he ran alongside, but clearly looking for an opportunity to get his foot in rather than jockey: he tripped over his own feet and ended up on his arse. Second time, he went to ground and whacked it into touch.
Yes I remember those also now. I watched the game at 3am so the memory was quite hazy. Mostly I was looking out for van Aanholt and Milner as they are the two who surprised me in being better or worse than expected.
My first article on Anfield Index on Shaqiri. Enjoy. bit.ly/2mAq3Qd