World Cup Individual Advanced Stats

Discussion in 'The Beautiful Game' started by schwuppe, Dec 18, 2012.

  1. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    Yeah, his key pass totals add up to 21 indeed - I guess Opta would originally have had the Argentina data for him as 1 assist plus 2 chances created (meaning chances not converted into a goal).

    This was the minutes per chance created data for the midfield and forward categories, copied from the same place I quoted Gregoriak from before (from this thread), and he had in 1982 a chance created every 22.5 minutes, which equals the 20 chances created over the 5 games:
    Midfielders
    18.5 Strachan (Sco)
    19.3 Petrovic (Yug)
    22.5 Vercauteren (Bel)
    22.5 Hoddle (Eng)
    23.8 Giresse (Fra)
    23.8 Platini (Fra)

    Forwards
    23.3 Lopez Ufarte (Spa)
    24.5 Juanito (Spa)
    24.8 Caszely (Chi)
    25.0 Rocheteau (Fra)
    25.0 Fazekas (Hun)
    (So that didn't include assist it seems, which would disadvantage players such as Littbarski and Zico - they'd show up more with the key pass data in effect, although it seems still not enough that they'd have been listed in those top 5s I think since Zico played 89 minutes per game and Littbarski 84)
    Planet World Cup - 1982 - Statistics
     
  2. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    Looking more closely, I think that was wrong, because it seems an anomaly that Vercauteren's numbers for 1982 key passes and Opta Chances Created are different (by one, which is the number of assists he had) - maybe that was an error or maybe Sofascore determined a legitimate reason, Opta corrected their data, or Gregoriak might have had a typo (we can't check the Opta widget now of course; I'm not saying a typo seems the most likely to me but it can happen to us all sometimes obviously).

    Littbarski does have 19 key passes recorded on Sofascore (the same number as Gregoriak posted as his Opta Chances Created). Also Zico does have 17 too.

    The question remains though whether chances created/key passes do include assists or not....

    To confirm that: I checked Maradona for 1986 and he has 27 key passes over his 7 games, and Gregoriak listed him with 27 chances created too (the question is including the 5 assists or not, but maybe this video below can help because he's credited with 1 key pass and 1 assist too in the Final, so if there appears to be any second key pass, which I don't recall off the top of my head, maybe that means the assists aren't included in chances created numbers by Opta)
    Diego Maradona vs Germany [World Cup] - YouTube
    Laudrup has 11 key passes (and chances created as listed by Gregoriak) over his 4 games indeed in that World Cup too, and as a secondary check maybe this video can help because only 2 of them are credited to the Uruguay game and he has 1 assist credited on Sofascore too (the rebound assist as per Planet World Cup would surely not be a key pass anyway though with Sofascore/Opta so it depends whether there are one or two more key passes by him, which again I couldn't be quite sure of before watching again even though I already watched the video a number of times before, even if mostly a version that was previously on Youtube with English TV coverage - I guess the visual footage is 100% the same and full given the length of the video though; the actual video I saw before of Maradona vs Germany was also a different upload btw but again I assume this one is a full version and would show any other key pass)
    Michael Laudrup's Greatest World Cup Performance | Denmark vs Uruguay 1986 - YouTube
    While I'm at it, just because we were earlier discussing the most suitable positions for these players in that World Cup, it's maybe a good idea to link to the heatmap pages (although 1986 will need to be selected after using the link)
    Diego Armando Maradona No team videos, transfer history and stats - Sofascore
    Michael Laudrup No team videos, transfer history and stats - Sofascore
     
  3. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    I think based on those videos, assuming key passes wouldn't count if the recipient is given offside, the assists probably are included in the chances created numbers (which become key passes with Sofascore, despite what that tweet I found said), which was what I originally assumed lol!

    Laudrup has a key pass (actually a cross) at 9 and a half minutes on the video I think, but otherwise just a similar one where Elkjaer was offside in the first half, other than the assist. Maradona, other than the assist, just seems to have one pass that perhaps could have been a key pass near the start but there was also an offside.
     
  4. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    Yeah, Opta clearly have assists included in chances created anyway:
    Opta Event Definitions - Stats Perform
    It seems like there could even be doubt whether crosses are included as key passes (just because for assists the definition in a bit more expansive than for key passes), but I think they will be because chances created are defined as key passes plus assists, and obviously chances can be created with crosses.

    So Opta do have a key pass category, but for sure it seems (except for the Vercauteren anomaly) like for 1966-2014 World Cups at least the Sofascore key pass data is what used to be chances created on the OPTA widget (and so includes assists too) - there were no 'key passes' shown on that OPTA widget and maybe it's even been introduced as a category since then or something anyway....
     
  5. PuckVanHeel

    PuckVanHeel Member+

    Oct 4, 2011
    Club:
    Feyenoord
    #330 PuckVanHeel, Nov 22, 2022 at 5:06 PM
    Last edited: Nov 22, 2022 at 5:15 PM
    Based on the data, I guess the Orange dot below Hazard and next to Maradona is Robben but hard to tell.

    The 19th best World Cup player as an outlier there (below any other footballing country worth their salt, except the best ranked Croatian player), quelle surprise.

    Such is the ingrained and default inclination (of the type "Holland [insert another smaller country with a small language] cannot be good at anything"). When it comes down to popular culture (incl. sports), one can choose between the role of (belittled, pittoresque) sideshow, villain - or written out of the story altogether (Rory Smith, Honigstein et al...).

    The last paragraphs of the Athletic article are sad. Also sad is the current players not being good enough... (but I can take it if that means Europe survives the winter safely)
     
  6. Trachta10

    Trachta10 Member+

    Apr 25, 2016
    Club:
    CA Boca Juniors
    This is a graph I made some time ago,
    just a few players, maybe I can add more in the future with the new sofascore data

    is the % the player's chances created represent over the total of the team shots

    (player chances created)/(team shots-player shots)*100

    I think this adjust better the reality of players from different eras

    [​IMG]
     
    PDG1978 repped this.
  7. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    #332 PDG1978, Nov 23, 2022 at 2:11 AM
    Last edited: Nov 23, 2022 at 2:32 AM
    Interesting to see, and definitely worth a rep I think mate. I would say that (like with goal contribution numbers to an extent too) maybe it reflects how reliant a team was on a player to create chances, moreso than just how good the player was at creating them (so players on stacked teams on the creative side would be further down than they otherwise might, although on the flip side being on a stacked team could even help them and their possibility to be the one providing a final ball on occasions, so it's not straightforward I suppose).

    Also if a player misses game time (particularly whole games, like Platini in 1982 as the prime example that jumps out) then showing his involvement in all the team's chances created over the tournament is factual, but a bit misleading in terms of his involvement % when on the pitch. Having said that, with the Sofascore table comme posted about ball progression and increase in chance to score, while I think Cruyff's placing very well represents things (and for example Littbarski's is notable for sure), it would be fair to say that Maradona's would move a bit (more favourably) if just 1986 was shown, so to some extent that table reflects how many World Cup games have been played if I'm reading/interpreting it right anyway, and on that one therefore Maradona probably in effect moved to a less favourable position in his latter World Cup games I guess (while in yours above the combination of a team that did handle possession very well but still did lean on him to make the openings shows up very favourably I suppose).

    EDIT on this post too! - I guess it's debatable whether the player's own shots should be deducted too (there is reason to do it, but also reason not to maybe because it moves away from purely who was responsible for the most chances created % wise overall). I know there are similar decisions that have to be made re: the goal contributions/assist contributions on your thread though Trachta, and maybe neither way is really perfect. A player who takes lots of shots could show up better than he should for 'creativity' doing it the way you did (and a player who doesn't shoot much at all but creates a lot can be disadvantaged), but then a player who creates a lot but happens to have quite a few shots would be disadvantaged the other way, I understand (and other players who shoot less might not have been better creators....).
     
  8. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    #333 PDG1978, Nov 23, 2022 at 2:17 AM
    Last edited: Nov 23, 2022 at 2:22 AM
    Interesting to see Mbappe not getting the Sofascore MOTM last night....
    Sofascore on Twitter: "FT | France 4–1 Australia Les Bleus took care of business after HT, as they put the game to bed with two goals in the space of three minutes halfway into the second half. France win their third consecutive opening match at the #FIFAWorldCup! #FRAAUS #Qatar2022 https://t.co/mE29EA6oq0" / Twitter
    I guess the first half brought down his average (despite already a few nice flicks and quite a bit of 'promise'):
    Sofascore on Twitter: "HT | France 2–1 Australia After two goalless matches, we finally saw some goals! Socceroos stunned the reigning world champions at the start, but Les Bleus regained control with two goals in the span of 5 minutes to hold the HT lead. #FRAAUS #Qatar2022 #FIFAWorldCup https://t.co/Rb2JH9bw1k" / Twitter

    I think I had something like an 8.5 in mind for his performance if I did it the 'traditional' way. It'll be interesting to see the media grades (I haven't seen any yet) but I wouldn't be surprised if they were a bit closer to that, and often with him as MOTM I don't think....

    EDIT: I guess there will be a 'big chance missed' that will have counted against him though of course.
     
  9. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    There was another moment where Griezmann (I think) played Mbappe in and he did mis-control though I remember now, so maybe the Sofascore stats/ratings do help reflect that kind of thing. Griezmann seemed less involved and dangerous overall I'd say, but as the key pass numbers show he was coming up with quality final balls, while Rabiot scoring high isn't really surprising given his end product involvements (and generally he played pretty well I think, after an iffy start maybe IIRC).

    I think I'd still be inclined to score Mbappe's performance in the range of 8 to 8.5 anyway myself, but the Sofascore number in this case (as with Maradona vs Bulgaria in 1986 for example) might back-up the school of thought that says forget how flashy or impressive a player seems and concentrate on how decisive he is overall (but probably the system at least under-states the qualitative aspects a bit and the general influence a player can have when in vibrant form - it can only count whether a pass was successful, how many 'dribbles' there were etc, even if maybe the underlying formula has more aspects than the available stats indicate even: I'm not sure for example whether Mbappe would be credited with any 'dribbles' before his assist or not - I guess at least there would be some kind of duel won recorded in that play though).
     
  10. Trachta10

    Trachta10 Member+

    Apr 25, 2016
    Club:
    CA Boca Juniors
    Yes I agree

    About the graphs posted before from The Athletic
    The problem here is that when the sample is too small the players with the fewer number of games will have the highest averages.
    But when the sample is big the averages tend to be more accurate, and only there is correct to compare averages, for example when you compare the entire career of a player with another, when they have hundreds of games played.

    So basically, the fairest thing to do (when the sample is small) is to compare the averages when the number of games is very similar. That is, compare single world cup campaigns.


    I added more players, some numbers are slightly different, the shots blocked are counted, and only counted the team shots in the games the players played.

    [​IMG]



    The percentage multiplied by the number of games
    This is more a "rating", that benefits players that played more games.

    [​IMG]
     
    PDG1978 repped this.
  11. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    Interesting to see. Hassler didn't take many shots either, so clearly laid on a high percentage of German chances in 1994, like the Sofascore numbers hint at. Not to ask you for a version without players shots being deducted then: but I can see that for example the Italians Pirlo (especially) in 2006 and Giannini in 1990 would move even higher up that way (Xavi from 2010 too, but compared to Pirlo his team took more shots, despite the fact their overall goals tally might be quickly overtaken by Spain 2022 I guess now, so in effect also more that he didn't 'set up').

    I'm guessing Scifo in 1990 would have a lower involvement in setting up chances with final balls than in 1994 then, but Dragan Stojkovic from 1990 is one that comes to mind that perhaps could be worth adding. Also Littbarski (who like I say did show up well in that graph comme posted, but did also play a lot of World Cup games) for 1982 and 1990 I'd think. Laudrup took a low number of shots (and in 1986 less than half the total of his chances created) so with shots not deducted he'd move up (for 1986 above rather than below Zidane 2006 therefore - to be fair he played more as a forward than Zidane comparing those years; for 1998 above rather than below C.Ronaldo 2010 of course too). Off topic for this thread but re: Laudrup I guess I gave you enough to do the goal contribution entry on your thread for 'La Liga' only did I (although to be honest I'm not sure I can add more, so it depends maybe what you could find out about those games I couldn't see any footage for in terms of the goals and who assisted them)?

    Like I say, I'm not conclusively adamant it's better not to deduct the player's shots taken anyway - Giannini 1990 would climb above Hagi 1994 for example, but it's true Hagi was more inclined to shoot and perhaps was more involved in making runs into and around the penalty area too (although Giannini did score a goal doing that himself too of course, but yeah overall he played deeper and more of a traditional midfield role I suppose). With Zico 1982 (and Maradona 1982 actually) it's not so much about their shots taken as about the number of shots their teams took overall.
     
    Trachta10 repped this.
  12. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    I do think btw that the interesting thing about that graph was that it was a bit wider reaching, and not only based on final balls, but I also think it'd be interesting to see a version for single World Cups only too, indeed.
     
  13. Trachta10

    Trachta10 Member+

    Apr 25, 2016
    Club:
    CA Boca Juniors
    Yeah I will try to find those games of Laudrup, now that I have more time.

    Dragan Stojković and Pierre Littbarski added

    [​IMG]


    And not deducting the player's shots

    [​IMG]
     
    PDG1978 repped this.
  14. PDG1978

    PDG1978 Member+

    Mar 8, 2009
    Club:
    Nottingham Forest FC
    Ok thanks, nice work.
     

Share This Page