1. Log in now to remove adverts - no adverts at all to registered members!

More defensive stats

Discussion in 'Liverpool' started by Jonesey, Apr 12, 2014.

  1. Jonesey

    Jonesey Well-Known Member

    Joined:
    Mar 2, 2011
    Messages:
    2,247
    Likes Received:
    93
    I've been looking at our CB pairings for the 33 games played, this is what I've come up with:
    [TABLE="width: 286"]
    [TR]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [/TR]
    [TR]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [/TR]
    [/TABLE]
    [TABLE="width: 372"]
    [TR]
    [TD]Pairing[/TD]
    [TD]No Games[/TD]
    [TD]Goals[/TD]
    [TD]Per Game[/TD]
    [TD]Clean Sheets[/TD]
    [/TR]
    [TR]
    [TD]Agger/Sakho[/TD]
    [TD]1[/TD]
    [TD]1[/TD]
    [TD]1.00[/TD]
    [TD]0[/TD]
    [/TR]
    [TR]
    [TD]Agger/Skrtel[/TD]
    [TD]12[/TD]
    [TD]13[/TD]
    [TD]1.08[/TD]
    [TD]6[/TD]
    [/TR]
    [TR]
    [TD]Agger/Toure[/TD]
    [TD]2[/TD]
    [TD]0[/TD]
    [TD]0.00[/TD]
    [TD]2[/TD]
    [/TR]
    [TR]
    [TD]Sakho/Skrtel[/TD]
    [TD]8[/TD]
    [TD]9[/TD]
    [TD]1.13[/TD]
    [TD]1[/TD]
    [/TR]
    [TR]
    [TD]Sakho/Toure[/TD]
    [TD]0[/TD]
    [TD]0[/TD]
    [TD]0.00[/TD]
    [TD]0[/TD]
    [/TR]
    [TR]
    [TD]Skrtel/Toure[/TD]
    [TD]10[/TD]
    [TD]17[/TD]
    [TD]1.70[/TD]
    [TD]1[/TD]
    [/TR]
    [TR]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [TD][/TD]
    [/TR]
    [TR]
    [TD][/TD]
    [TD]33[/TD]
    [TD]40[/TD]
    [TD]1.21[/TD]
    [TD]10[/TD]
    [/TR]
    [/TABLE]


    Thoughts?

    It certainly adds weight to the idea that Agger/Skrtel are by far our best CB pairings, though you cannot of course blame EVERY goal conceded on the CB's - Toure's against WBA notwithstanding <doh>
     
    #1
  2. jenners04

    jenners04 I must not post porn!

    Joined:
    Feb 10, 2011
    Messages:
    15,143
    Likes Received:
    4,582
    did you do this before you got a job :bandit:

    only messing mate, respect for people who can be bothered to to do this, irony being statistician was recommended for me for a job to do lol

    good stats, i am not sure about agger and skrtel any more, i know the stats would argue against that but i would like to see Skrtel and Sakho as the future choice personally, will take time but i think it will work eventually, ie start using it more in pre season so they get to understand each other better.

    suppose it depends on the games played the quality of the opo as well, as looking at that Agger and Toure jump out as havent conceded but if those games are against relegation then its hardly worth pointing out, if the 2 was say against a top 6 team on both occasions might be worth looking at.
     
    #2
  3. jaffaSlot

    jaffaSlot Well-Known Member

    Joined:
    Jun 4, 2011
    Messages:
    16,229
    Likes Received:
    7,284
    Sakho was playing regularly when we played city and the chavs away. Also, recently Rodgers has adopted the diamond shape to sure us up a bit.
     
    #3
  4. Skylarker

    Skylarker PL High Commissioner

    Joined:
    Apr 15, 2011
    Messages:
    49,428
    Likes Received:
    30,917
    Agger and Skrtel know each others game so well, they'd probably be the safest bet at the moment.


    I just can't help thinking Sakho has more to offer than Agger and deserves to be our new main CB. I think this will probably happen next season, but I expect Agger and Skrtel to continue this season (if Agger remains fit).
     
    #4
  5. Jonesey

    Jonesey Well-Known Member

    Joined:
    Mar 2, 2011
    Messages:
    2,247
    Likes Received:
    93
    Call me sad, I just love playing with spreadsheets!!
     
    #5
  6. Skylarker

    Skylarker PL High Commissioner

    Joined:
    Apr 15, 2011
    Messages:
    49,428
    Likes Received:
    30,917
    Did you include the game where all 4 CB's played - and none of them had a ****ing clue what they were doing? <laugh>
     
    #6

  7. jenners04

    jenners04 I must not post porn!

    Joined:
    Feb 10, 2011
    Messages:
    15,143
    Likes Received:
    4,582
    i sort of here you, i used to love making templates for spreadsheets and doing the equations but then i would get bored for some reason when i had done the template, then need a dif challenge.

    you ever use macros so you merge dif spreadsheets into one?
     
    #7
  8. Skylarker

    Skylarker PL High Commissioner

    Joined:
    Apr 15, 2011
    Messages:
    49,428
    Likes Received:
    30,917
    Now we're talking <laugh>


    On the topic - Best program for messing about with stats has to be SPSS. It messes with your head too though, so use with caution.
     
    #8
  9. Foredeckdave

    Foredeckdave Music Thread Manager

    Joined:
    May 30, 2011
    Messages:
    19,804
    Likes Received:
    132
    Nah! The best thing is to know enough about stats and programmes to be able to ask intelligent questions bout the methodology and their interpretation of the results and leave the donkey work to the nerds ;)
     
    #9
  10. Skylarker

    Skylarker PL High Commissioner

    Joined:
    Apr 15, 2011
    Messages:
    49,428
    Likes Received:
    30,917

    SPSS experts are real nerds. I got to about page 1000000 of the manual and gave up - luckily I learnt the basics though!

    They're a sneaky bunch though as they can always get their data to suggest what they want it to :)
     
    #10
  11. Foredeckdave

    Foredeckdave Music Thread Manager

    Joined:
    May 30, 2011
    Messages:
    19,804
    Likes Received:
    132

    Now then Sky, let's face some very serious truths about the majority of statistical evidence. Via careful manipulation of the universe, sample frame, etc. etc. and then some weird and wonderful mathematical devices they can always provide evidence that proves whatever it was you wished to prove in the first place!! Shamans the lot of them <laugh>
     
    #11
  12. moreinjuredthanowen

    moreinjuredthanowen Mr Brightside

    Joined:
    Jun 9, 2011
    Messages:
    122,734
    Likes Received:
    29,599
    It's all about context

    1 no partnership can be judged on a couple of games

    2. It depends on opposition

    3. It takes time and agger and Skertl have years of playing together

    Frankly it proves toure can't play lcb .... After that I think Skertl sahko partnership needs time to see where it goes
     
    #12
  13. Agger & Skrtel have the most cleans but that makes the 13 the conceded in the remaining 6 games they played together rather worrying. Basically tells me they are hit or miss.
     
    #13
  14. Jonesey

    Jonesey Well-Known Member

    Joined:
    Mar 2, 2011
    Messages:
    2,247
    Likes Received:
    93
    Yeah - I'm pretty well versed in Visual Basic - I've used it to create loads of "solutions" for various organisations, all with user friendly "front end" forms.

    Some of my projects have ended up with upwards of 10,000 lines of code
     
    #14
  15. InBiscanWeTrust

    InBiscanWeTrust Rome, London, Paris, Rome, Istanbul, Madrid Forum Moderator

    Joined:
    May 22, 2011
    Messages:
    72,282
    Likes Received:
    27,243
    It's a good basic thing to look at but you'd need loads more info in order to make a judgement.

    Who the opposition was against each pairing, what the midfield were like (if you have Gerrard, Henderson, Allen in front you'd maybe expect more protection than Gerrard Henderson Coutinho for example). If any individual errors were caused.

    What is interesting is that Sakho/Skrtel have played 8 games and conceded 9 goals with 1 clean sheet. Means in the 7 games they've conceded they've only let in 1 goal a game in 6 and 2 in 1. At the moment, that is giving us a huge chance of winning games a we'd back ourselves to score at least 2 goals in a game.

    compare that to Agger/Skrtel who although have 6 clean sheets, mean in the 6 games they've played and conceded, they've let in 13 goals! Over 2 goals a game.
     
    #15

Share This Page