Ozil vs De Bruyne vs Mikhi: Stat-Off

Discussion in 'Football Forum' started by Pogue Mahone, Oct 31, 2017.

  1. Oct 31, 2017
    #1

    Pogue Mahone Poster of the year 2008

    Joined:
    Feb 22, 2006
    Messages:
    96,854
    Location:
    "like a man in silk pyjamas shooting pigeons
    Turns out that, statistically, there's very little between them in the PL so far this season. Which, I'll have to admit, was a shock. De Bruyne and Mikhi both have more assists and shoot more accurately than Ozil. Ozil loses the most tackles per game. Otherwise, the differences are fairly marginal apart from "key passes" where Ozil makes not far off twice as many per game as either of the other two!

    So. What do we conclude? Are stats a nonsense?
  2. Oct 31, 2017
    #2

    charlie9882 New Member

    Joined:
    Oct 30, 2017
    Messages:
    115
    De Bruyne plays considerably deeper than both of them, even being deployed as a defensive midfielder this season. He's a far more complete player.
  3. Oct 31, 2017
    #3

    Varun Moderator Staff

    Joined:
    Mar 16, 2011
    Messages:
    43,262
    Location:
    Mumbai
    If you use them to judge their qualities as players, yes. De Bruyne is on another level to the other 2.
  4. Oct 31, 2017
    #4

    James Peril Banned

    Joined:
    Aug 10, 2016
    Messages:
    2,389
    What is a key pass, and would all of us agree to what the stats show is a key pass? And what is a chance? Özil has four in average every game? Arsenal don’t even create four chances every game! I hate stats
  5. Oct 31, 2017
    #5

    adexkola Arsenal supporter

    Joined:
    Mar 17, 2008
    Messages:
    32,326
    Location:
    Me, and you. Yo momma, and yo cousin too.
    Supports:
    orderly disembarking on planes
    Stats are not nonsense. If stats differ with your perception then one of 2 things is happening:

    1. The stats are not completely descriptive. Averages obscufate a lot of detail. The answer is to not dismiss stats, but demand more context.
    2. Your perception is shit. Either you don't watch full Arsenal/City matches the same way you do for United games, so every wart of Mhki is dissected, while it is assumed De Bruyne had a great game because of MOTD. Or you were drunk. Or more/less emotionally invested. Whatever. I'm not trusting your probably biased opinion over numbers that are collected without bias. Imagine if the world worked that way.

    Either way, interesting perspective on the "Mhki is shit" notion.
  6. Oct 31, 2017
    #6

    Pogue Mahone Poster of the year 2008

    Joined:
    Feb 22, 2006
    Messages:
    96,854
    Location:
    "like a man in silk pyjamas shooting pigeons
    Guilty as charged, your honour...
  7. Oct 31, 2017
    #7

    adexkola Arsenal supporter

    Joined:
    Mar 17, 2008
    Messages:
    32,326
    Location:
    Me, and you. Yo momma, and yo cousin too.
    Supports:
    orderly disembarking on planes
    :lol:

    I'm the same, mind.
  8. Oct 31, 2017
    #8

    ROFLUTION Full Member

    Joined:
    Jan 22, 2009
    Messages:
    2,955
    Location:
    Denmark
    Im in favour of stats, but this is probably a twisted example

    - De Bruyne plays deeper. Hailed as great.
    - Ozil had no end product in the first 5-6 Games. But had a craaazy amount of keypasses against Everton.
    - Mkhi was the opposite of Özil' s first 5-6 Games in end-product
  9. Oct 31, 2017
    #9

    André Dominguez Full Member

    Joined:
    Aug 7, 2017
    Messages:
    1,298
    Location:
    Lisbon
    Supports:
    Benfica, Académica
    Give me Ozil anytime. He might be a lazy wanker at times, but when he's focused, he's literally unstopable. Too bad for Arsenal fans those days are far away :D
  10. Oct 31, 2017
    #10

    Pogue Mahone Poster of the year 2008

    Joined:
    Feb 22, 2006
    Messages:
    96,854
    Location:
    "like a man in silk pyjamas shooting pigeons
    To be honest, I've started fast-forwarding City highlights on MOTD. Life's too short for that shit. Even being drunk doesn't make it any less painful.
  11. Oct 31, 2017
    #11

    Werd. New Member

    Joined:
    Jun 12, 2009
    Messages:
    1,238
    Nail. Hit. Head.
  12. Oct 31, 2017
    #12

    PedroMendez Acolyte

    Joined:
    Aug 9, 2013
    Messages:
    8,688
    Location:
    the other Santa Teresa
    Stats aggregate actions that shouldnt be seen as the same. KdB is at least two levels above the other two in every aspect. He is a lot more dangerous going forward despite playing deeper.
  13. Oct 31, 2017
    #13

    Eboue nasty little twerp with crazy bitter-man opinions

    Joined:
    Jun 6, 2011
    Messages:
    48,296
    Location:
    Kerala, India
    Supports:
    Arsenal
    this post is pointless. the definitions are easily found with a quick google. if you wanted to know you would have found out already. you just want to whine
  14. Oct 31, 2017
    #14

    SqualorVictoria New Member

    Joined:
    Dec 25, 2013
    Messages:
    345
    Supports:
    City
    "Key pass" is a bit of a shit stat as, according to Whoscored that is, they are simply the passes which led to a shot. But it's maybe worth noting that Mkhitaryan became an absolute beast under Tuchel, in a team that deployed the positional play as freakishly as Pep does. So it's not that much of a nonsense to suggest that maybe he would have a similar output to what De Bruyne is having currently.
  15. Oct 31, 2017
    #15

    Moonred Full Member

    Joined:
    Nov 15, 2012
    Messages:
    10,182
    Location:
    Virgo Supercluster
    I thought Mkhitaryan was incredibly productive at the beginning of the season which pretty much is carrying him till now compared to Kdb who started getting into his groove after a few matches. No idea about Ozil as I don’t watch Arsenal but it’s not surprising to see both Ozil and Mkhi still close to each other as these players are generally creative and being shit still translates to moments of opportunities.
  16. Oct 31, 2017
    #16

    Raees Legal Guardian of the Football forums

    Joined:
    May 16, 2009
    Messages:
    26,319
    Not surprised as Mkhi is a very outcomes based player. He's very direct and his flaws won't really get picked up on stats wise likewise the stuff that De Bruyne does in terms of linking up play or controlling tempo of a game doesn't get reflected as accurately statistically.

    Bit like comparing Mkhi to say Iniesta in his prime and saying he's a better player because statistically he's more effective.
  17. Oct 31, 2017
    #17

    deafepl Full Member

    Joined:
    May 19, 2017
    Messages:
    2,275
    The only difference that Mkhitaryan is incontinently and De Bryune is so continuous. Mkhitaryan assisted 5 goals in attacking midfielder just for the first 3 games and didn't have any assist after 3 games, I think maybe it had something to do with Pogba's absence?

    Manchester United with Pogba created average 2.45 per EPL game and without Pogba created average 1.3 per EPL games this season.

    tbh I'm more concerned about Mata than Mkhitaryan.
  18. Oct 31, 2017
    #18

    KirkDuyt Full Member

    Joined:
    Nov 25, 2015
    Messages:
    7,174
    Location:
    Dutchland
    Supports:
    Feyenoord
    Stats are not nonsense, they're factual data, so if anything, they're the opposite. What they don't do is tell the whole story. Not everything in football can be put into stats and stats don't take in to consideration all factors. Like for instance KDB being a ginger with a dumb face.
  19. Oct 31, 2017
    #19

    Massive Spanner Thinks Geoff Shreeves has one

    Joined:
    Jul 2, 2014
    Messages:
    9,797
    Location:
    Tool shed
    i've watched enough of Utd and City this season to accept that Mkhi has been mostly shit and KDB has been mostly very, very good.
  20. Oct 31, 2017
    #20

    tavtavtav New Member

    Joined:
    Jan 18, 2015
    Messages:
    90
    Also, maybe you could argue a fit Ozil playing further up the field does infact create loads of chances/a high percentage of his team's chances and Arsenal's forwards are incompetent.
  21. Oct 31, 2017
    #21

    Pogue Mahone Poster of the year 2008

    Joined:
    Feb 22, 2006
    Messages:
    96,854
    Location:
    "like a man in silk pyjamas shooting pigeons
    He's got a "similar output" right now now, playing for Mourinho.
  22. Oct 31, 2017
    #22

    Pogue Mahone Poster of the year 2008

    Joined:
    Feb 22, 2006
    Messages:
    96,854
    Location:
    "like a man in silk pyjamas shooting pigeons
    Is he though? I know that Iniesta has famously poor goals/game stats but isn't his record for assists and key passes very good?
  23. Oct 31, 2017
    #23

    iluvoursolskjær Full Member

    Joined:
    Dec 31, 2015
    Messages:
    1,793
    Location:
    Searching for life's white text
    Supports:
    The return of Comeback Kings
    Stats don’t lie, but when you’re looking at averages it can make the edges a little fuzzy. Miki started superbly, output-wise at least, whilst we were battering teams but has since played to our detriment. Haven’t watched either of the other two enough for me to offer any real meaningful commentary, but that De Bruyne is seemingly - by the stats - on par with the best 10s in output whilst playing deeper tells us a little about the level he’s at atm (just look at the greats a la Iniesta where the stats ovvi fail to articulate his effect on the team). It obviously helps that he’s got a ridiculous number of good options available to him, while Ozil probably suffers from the opposite being true in his case.

    De Bruyne has been consistently the better player. That isn’t even up for debate.
  24. Oct 31, 2017
    #24

    Cristiano Lell BANNED LOLZ

    Joined:
    Nov 20, 2014
    Messages:
    2,201
    Location:
    Berlin
    Supports:
    Bayern München
    I'm not sure. I think Iniesta might be one of those player whose actual greatness is not captured by stats.
    I had this argument with an stats-obsessed person about Iniesta's performance vs us in the CL semi 2015, and he kept claiming Iniesta did nothing, which he could 'back up' by stats, while I watched the game and had the impression Iniesta put in an absolute world beater performance carrying the ball through midfield pressure.

    On topic, on the one hand stats do not tell the whole story, and get more relative and debatable the more broad they get(already your number of assists will not only depend on your ability, but on the chance conversion of your teammates, etc, and then key passes or duels will be very hard to judge or even discern).

    On the other hand, I don't see what's so surprising about the fact all of De Bruyne, Özil and Mkhi would have similar stats - after all, they're all without a doubt absolute top players who are among the best attacking midfielders in the world.
  25. Oct 31, 2017
    #25

    MThomas Banned

    Joined:
    Apr 14, 2004
    Messages:
    16,728
    Location:
    Figo: In Spain we called Benitez 'píldora para dor
    :lol::lol:
  26. Oct 31, 2017
    #26

    Cristiano Lell BANNED LOLZ

    Joined:
    Nov 20, 2014
    Messages:
    2,201
    Location:
    Berlin
    Supports:
    Bayern München
    Explain to me what's so funny about it.
  27. Oct 31, 2017
    #27

    oneniltothearsenal Arse Lover Scout

    Joined:
    Dec 17, 2013
    Messages:
    6,576
    Supports:
    Brazil, Arsenal,LA Aztecs
    There is definitely something to this. Football is harder to reflect in stats than other sports or games. Things like off the ball movement, decision making and positioning simply can't be captured in a number. Things like the heat maps and passing charts can start to provide more context but there is still a degree of incompleteness that I think defies reduction to numbers.
  28. Oct 31, 2017
    #28

    JPRouve can't stop thinking about balls - NOT deflategate

    Joined:
    Jan 31, 2014
    Messages:
    33,462
    Location:
    France
    A quick look tells me that they are not good at the same things and/or not used in the same roles. The stats aren't that close.
  29. Oct 31, 2017
    #29

    ThatsGreat New Member

    Joined:
    Aug 1, 2016
    Messages:
    329
    Supports:
    Arsenal
    Ozil creates chances for the likes of Welbeck and Ramsey. Take the stoke game that Arsenal lost, Ozil created three clear chances, none of which were finished. The key pass stat catches this contribution, whereas MOTD will show you only the goals and assists. Even in the Watford game which he didn't even start, he created a chance for Iwobi to miss immediately after he came on.
  30. Oct 31, 2017
    #30

    VeevaVee despite the protests, wears Ugg boots

    Joined:
    Jan 3, 2009
    Messages:
    27,333
    Stats are ok to a point. But stats said we had a good defence under LVG when really we just passed it around the half way line so nothing happened.

    In this case, the stats aren't taking into consideration the other stuff that's happening and they're contributing to. I'd be intrigued to see times lost possession though. Is that successful take ons? If so Mhki is on 48% while the other two are on 75 and 89%. Which would account for him running into opposition.
  31. Oct 31, 2017
    #31

    Chipper Full Member

    Joined:
    Oct 25, 2017
    Messages:
    917
    Stats are interesting but never account for the specific role that a manager is asking a player to perform.

    Me, OP or any other fan might consider certain areas to be the best ones to compare whereas a manager might be looking at other things. I just added the "total score" stat for example, and Mikhi seems to be way behind. I don't know what that is for certain. I'm assuming it's a summary of all the stats available to choose from so it looks like there are certain areas where he's definitely trailing. The role the manager is telling the player to perfrom also affects the stats of course.

    They also lack detail:
    - A pass behind a player that the receiver needs to turn backwards to retrieve
    - A pass blasted at a player that he does brilliantly to control
    - A perfectly weighted ball that finds a player's run, or even tells him where to run
    = All completed passes and none of it accounts for the pressure they were under/time they had on the ball when making the pass, whether it was a passed on the volley or on the ground, whether it was a first time pass or if they controlled it first, whether or not the passer was on the move or stationery, whether or not the passer had to find space himself to receive the ball before making the pass etc.
    Last edited: Oct 31, 2017
  32. Oct 31, 2017
    #32

    MThomas Banned

    Joined:
    Apr 14, 2004
    Messages:
    16,728
    Location:
    Figo: In Spain we called Benitez 'píldora para dor
    The notion that Mkhi is among the best attacking midfielders in the world

    It's priceless
  33. Oct 31, 2017
    #33

    Cristiano Lell BANNED LOLZ

    Joined:
    Nov 20, 2014
    Messages:
    2,201
    Location:
    Berlin
    Supports:
    Bayern München
    :rolleyes:
  34. Oct 31, 2017
    #34

    adexkola Arsenal supporter

    Joined:
    Mar 17, 2008
    Messages:
    32,326
    Location:
    Me, and you. Yo momma, and yo cousin too.
    Supports:
    orderly disembarking on planes
    Off the ball movement can absolutely be captured by numbers. How does a player without the ball impact the game? Does he draw defenders to him? Complex questions can be broken down into very specific questions which are fodder for quantitative methods. Same with positioning. Decision making is even more complex... because what is the right decision to make at a point in time, and who decides that, we armchair fans? Is the right decision always less risky, and where does that leave more ambitious moves on the field a Mhkitaryan would make?
  35. Oct 31, 2017
    #35

    oneniltothearsenal Arse Lover Scout

    Joined:
    Dec 17, 2013
    Messages:
    6,576
    Supports:
    Brazil, Arsenal,LA Aztecs

    How would you quantify off the ball movement into a statistic?
  36. Oct 31, 2017
    #36

    M113FF New Member

    Joined:
    Oct 11, 2017
    Messages:
    45
    Supports:
    City
    From whoscored:

    WhoScored.com Ratings are considered to be the most accurate, respected and well-known performance indicators in the world of football. Our ratings are currently used among media giants, bookmakers and football clubs.
    WhoScored.com Ratings are based on a unique, comprehensive statistical algorithm, calculated live during the game. There are over 200 raw statistics included in the calculation of a player'’s/team’'s rating, weighted according to their influence within the game. Every event of importance is taken into account, with a positive or negative effect on ratings weighted in relation to its area on the pitch and its outcome.

    Current top 10 based on their criteria:

    Kane
    Aguero
    De Bruyne
    Silva
    Fernandinho
    Sterling
    Lukaku
    Fellaini
    Sane
    Davies

    Ozil meanwhile is about 60th. Says it all really...


    Pogba was right up there as well before the injury
  37. Oct 31, 2017
    #37

    LennonNL nipple flasher and door destroyer

    Joined:
    May 8, 2013
    Messages:
    8,074
    Location:
    M20
    By watching a player move off the ball in order to create space for another player, draw a defender with him or to receive a pass himself. These statisticians meticulously watch over games and I’m sure they’d come up with a way to quantify off the ball movement, if not with those 3 criteria.
  38. Oct 31, 2017
    #38

    roonster09 Correctly predicted France to win World Cup 2018

    Joined:
    May 10, 2009
    Messages:
    12,942
    You check their top 10-15 players for the last 3-4 years and you will see lot of strange names.
  39. Oct 31, 2017
    #39

    ADJUDICATOR FULL MEMBER

    Joined:
    Aug 13, 2013
    Messages:
    4,659
    Supports:
    THE BRAVE AND THE BOLD
    You could have 'how many times player x is the recipient of a clear cut chance', as a starting point.
  40. Oct 31, 2017
    #40

    MThomas Banned

    Joined:
    Apr 14, 2004
    Messages:
    16,728
    Location:
    Figo: In Spain we called Benitez 'píldora para dor

    Oh come on...

    It's not a good starting point either, good movement would also be creating space for others. It's close to impossible to measure this in numbers.