Attention: Please take a moment to consider our terms and conditions before posting.

Championship XG

PolzeathNick
PolzeathNick Posts: 332
in General Charlton
I follow the XG data on Experimental361 to give accurate indicators on where we are.  Makes interesting reading after the first 3 games.  Leeds are comfortably the best side so far...their expected Goal Difference is +3.4, with the next best Fulham on +2.2.  Quite easy to make the case that Leeds are value at 9/4 to win the division.

We currently rank 8th.  With XG for of 4.1, and XG against of 3.1, and a net of +1.  So we are outperforming on goals scored, but conceded one more than we should have too.

Early days obviously, and this doesn't adjust for quality of opposition, but the signs are encouraging, especially given two of our matches have been away from home.

It will be interesting to see how we go against Forest (XG for 3.9, XG against 2.9) who have a similar expected goal difference to us and went well against Leeds.

Wigan, Luton and Birmingham are adrift at the bottom with XG difference of -2.5,-2.5 and -2.9 respectively.


«134567…11

Comments

  • shine166
    shine166 Posts: 13,918
    Useless statistics that seem like they are from a bookies.
  • se9addick
    se9addick Posts: 32,035
    I’ve never really thought of XG as a statistic, it’s really hypothetical and (as you say) doesn’t take into account a range of factors that it probably should.

    I think it’s interesting, but I wouldn’t place to much weight on it as an assessment of how we are doing. 
  • Dazzler21
    Dazzler21 Posts: 51,344
    What the flip are XG?
  • PolzeathNick
    PolzeathNick Posts: 332
    Over the season their accuracy tends to get borne out.  You do get exceptions though...Reading were a massive outlier a few seasons back when they got beat in the Playoff final by Huddersfield, and they should've been flirting with relegation!
  • PolzeathNick
    PolzeathNick Posts: 332
    Dazzler21 said:
    What the flip are XG?
    Expected Goals.  Effectively what the theoretical fair score should've been in a match.
  • Briston_Addick
    Briston_Addick Posts: 11,677
    PolzeathNick said:
    Dazzler21 said:
    What the flip are XG?
    Expected Goals.  Effectively what the theoretical fair score should've been in a match.
    If @happyvalley's in charge of the stats the XG will always be a comfortable Charlton Athletic 4 Oppo 0.
  • stackitsteve
    stackitsteve Posts: 12,102
    Last season...

    Premier League top 3: 1 Man City 2 Liverpool 3 Chelsea
    end of season xG stats: 1 Man City 2 Liverpool 3 Chelsea

    Championship: 1 Norwich 2 Sheffield United 3 Leeds
    xG: 1 Leeds 2 Sheffield United 3 Norwich

    League One: 1 Luton 2 Barnsley 3 Charlton
    xG: 1 Luton 2 Barnsley 3 Pompey

    There is certainly more to it than useless statistics. 



    Also the following via @markohaire on twitter

    If you also look at 'xG from open play' and 'shots in the box'...
    The top-three teams in the league filled 30 of the available 36 data slots – 5 more were filled by Pompey and Mansfield, who finished 4th.
    Colchester - finished 8th - were the only team to finish below 4th in the league and take a top-three data position.
  • cafcsinger
    cafcsinger Posts: 5,548
    Playing in the rain we have a 100% win record, in the sun its only 50%. Therefore we want it to rain on Wednesday with the expected weather forecast a light chance of showers.
  • Gravesend_Addick
    Gravesend_Addick Posts: 7,299
    I've heard a lot about XG, but I don't really understand it.  Where can I find XG stats for tonight's premier league game & what does it all mean?
  • se9addick
    se9addick Posts: 32,035
    stackitsteve said:
    Last season...

    Premier League top 3: 1 Man City 2 Liverpool 3 Chelsea
    end of season xG stats: 1 Man City 2 Liverpool 3 Chelsea

    Championship: 1 Norwich 2 Sheffield United 3 Leeds
    xG: 1 Leeds 2 Sheffield United 3 Norwich

    League One: 1 Luton 2 Barnsley 3 Charlton
    xG: 1 Luton 2 Barnsley 3 Pompey

    There is certainly more to it than useless statistics. 



    Also the following via @markohaire on twitter

    If you also look at 'xG from open play' and 'shots in the box'...
    The top-three teams in the league filled 30 of the available 36 data slots – 5 more were filled by Pompey and Mansfield, who finished 4th.
    Colchester - finished 8th - were the only team to finish below 4th in the league and take a top-three data position.
    I don’t think “statistic” is the correct term at all though. In fact I think some of the resistance to taking XG seriously is because it is prevented as a stat in the same way that goals scored, assists etc are
  • Sponsored links:



  • golfaddick
    golfaddick Posts: 33,628
    I hate this "expected goals"  rubbish they show on MOTD  Please can someone explain it to me as it makes no sense.
  • Callumcafc
    Callumcafc Posts: 63,764
    edited August 2019
    https://youtu.be/zSaeaFcm1SY
  • Callumcafc
    Callumcafc Posts: 63,764
    Personally I love it because I’m a proper numbers nerd. If you really wanna get into it you can look at stuff like expected assists too.
  • el-pietro
    el-pietro Posts: 611
    edited August 2019
    XG is just putting a number to the sort of thing football fans have talked about for years. We've all come away from a game saying, how did we lose that, they only had one chance, we had loads?!
    XG does its best to answer that, it is measuring the quality of chances created. Just about everyone here has talked about sitters and worldies. Sitters have high XG numbers, close to 1. Worldies and shots from distance tend to have lower XG, closer to 0. If you have a bunch of sitters in a game then XG will be high. If you give up a lot of sitters in a game your opponents XG will be high.
    So after a game you can look at the XG and say we created a lot of good chances and XG says we shoudl have scored about 3 goals, and our opponents shoudl have scored about 1. The actual score might have been 1-1 but you can feel good that your team is creating quality chances and over time that should balance out, and there will be games where you get a little bit lucky, but over time XG will be more and more accurate as the sample size improves.
    If you are consistently under performing your XG then you probably need to replace your strikers/goal keeper.

  • DoctorCharlton
    DoctorCharlton Posts: 2,463
    Had one click on the website and came straight off. Too many numbers here there and everywhere for an old goat like me.

    I'll take Lancashire Lads player marks thread over that all day.
  • PaddyP17
    PaddyP17 Posts: 13,035
    I have an inherent dislike of xG, because useless pundits tend to wheel the concept out to make themselves sound smarter than they actually are, but there's no denying it's a useful way of plotting how good/certainly how clinical a team is.

    However, it seems as if a little too much of it is "after-the-fact" data that seems obsolete except for talking in hindsight. Meh. Whatever. All that really matters, of course, is the actual Gs.
  • Covered End
    Covered End Posts: 52,006
    PaddyP17 said:
    I have an inherent dislike of xG, because useless pundits tend to wheel the concept out to make themselves sound smarter than they actually are, but there's no denying it's a useful way of plotting how good/certainly how clinical a team is.

    However, it seems as if a little too much of it is "after-the-fact" data that seems obsolete except for talking in hindsight. Meh. Whatever. All that really matters, of course, is the actual Gs.
    Exactly.
    If in our next 10 matches we get into a load of excellent scoring positions and get shots in on goal. whilst at the other end we stop the opposition from getting a sniff at goal, they will have us top of the table.
    Mind you if that happens and we are top, it won't be hard to predict that we should be top. Wibble.
  • cantersaddick
    cantersaddick Posts: 16,915
    I love it but them I'm an analyst by profession so bit of a nerd when it comes to sports analysis. 

    Don't get me started on fielding metrics in cricket.
  • Callumcafc
    Callumcafc Posts: 63,764
  • Callumcafc
    Callumcafc Posts: 63,764
  • Sponsored links:



  • Callumcafc
    Callumcafc Posts: 63,764
  • Callumcafc
    Callumcafc Posts: 63,764
  • MrOneLung
    MrOneLung Posts: 26,853
    So this is an arbitrary stat? 
    Who decides the ranking of each particular shot? One person may think an effort is an expected goal and another may have it at 50% chance of goal. 
  • el-pietro
    el-pietro Posts: 611
    MrOneLung said:
    So this is an arbitrary stat? 
    Who decides the ranking of each particular shot? One person may think an effort is an expected goal and another may have it at 50% chance of goal. 
    Its based on similar chances historically and how often they are converted.
  • PolzeathNick
    PolzeathNick Posts: 332
    As most would have spotted regardless of any stats, we were lucky on Saturday.  XG had us losing 0.5-2.3.

    The updated table sees that we should be sitting 10th, so still seemingly overachieving at this stage.  But as alluded to above we seem to be finding something extra in terms of being very clinical, and reaching very strong positions in play.

    Leeds still way out ahead on top.  Should be 5 from 5, and 4 points clear.  That 9/4 for the title is now long gone a best priced 5/4 and as short as 5/6 in a place!
  • SuedeAdidas
    SuedeAdidas Posts: 7,741
    cantersaddick said:
    I love it but them I'm an analyst by profession so bit of a nerd when it comes to sports analysis. 

    Don't get me started on fielding metrics in cricket.
    Ok - I won’t 😄
  • Dazzler21
    Dazzler21 Posts: 51,344
    PolzeathNick said:
    As most would have spotted regardless of any stats, we were lucky on Saturday.  XG had us losing 0.5-2.3.

    The updated table sees that we should be sitting 10th, so still seemingly overachieving at this stage.  But as alluded to above we seem to be finding something extra in terms of being very clinical, and reaching very strong positions in play.

    Leeds still way out ahead on top.  Should be 5 from 5, and 4 points clear.  That 9/4 for the title is now long gone a best priced 5/4 and as short as 5/6 in a place!
    So it was wrong... And continues to be wrong. 

    Are we overachieving? Or just underrated?

    Not a fan and not sure if it tells you anything accurately...
  • stackitsteve
    stackitsteve Posts: 12,102

  • Callumcafc
    Callumcafc Posts: 63,764
    edited August 2019
    Dazzler21 said:
    PolzeathNick said:
    As most would have spotted regardless of any stats, we were lucky on Saturday.  XG had us losing 0.5-2.3.

    The updated table sees that we should be sitting 10th, so still seemingly overachieving at this stage.  But as alluded to above we seem to be finding something extra in terms of being very clinical, and reaching very strong positions in play.

    Leeds still way out ahead on top.  Should be 5 from 5, and 4 points clear.  That 9/4 for the title is now long gone a best priced 5/4 and as short as 5/6 in a place!
    So it was wrong... And continues to be wrong. 

    Are we overachieving? Or just underrated?

    Not a fan and not sure if it tells you anything accurately...
    I think both those suggestions are wrong. The data shows that we are, up to this point, very clinical.

    xG cannot predict every score every time but over a long period, it will get it right.

    Nothing more, nothing less.
  • PolzeathNick
    PolzeathNick Posts: 332
    stackitsteve said:

    Crazy how far ahead Leeds are

Categories

“Marketplace”