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

Championship XG

2456711

Comments


  • Crazy how far ahead Leeds are
    I would suggest that’s probably unsustainable and we’ll see a mid season dip in form where the pack catches them as we saw last season...
  • edited August 2019
    Dazzler21 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.
    But unless you're a bit thick, any attending Charlton fan will tell you that up to this point, we are very clinical.

    It seems to me the stats just confirm what has happened and anyone with a modicum of common sense will already know.

    A bit similar to attendance stats, when the stats say we've had 14,000+ attendances this season in the league.

    Well, if you were there and had a bit of common sense, you already know.

  • Crazy how far ahead Leeds are
    I would suggest that’s probably unsustainable and we’ll see a mid season dip in form where the pack catches them as we saw last season...
    As Ander Herrera said when he was at Bilbao, they were usually done by February because that's when they were physically spent. So I'd agree.

  • Crazy how far ahead Leeds are
    Is that the chart for televised appearances?
  • edited August 2019
    Dazzler21 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.
    But unless you're a bit thick, any attending Charlton fan will tell you that up to this point, we are very clinical.

    It seems to me the stats just confirm what has happened and anyone with a modicum of common sense will already know.

    A bit similar to attendance stats, when the stats say we've had 14,000+ attendances this season in the league.

    Well, if you were there and had a bit of common sense, you already know.
    It's there to put a number on it, it's nothing about common sense. xG isn't gospel, it's a stat to be used in context to what you are saying. 
  • 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!
    Is that an outlier, or was XG just utterly wrong? I’ll admit I don’t fully understand XG, but I do find it very irritating when a team is measured against it as it seems to me that’s more about the prediction being wrong than a team over or under performing. 
  • The following season Reading went from 3rd to 20th
  • The following season Reading went from 3rd to 20th
    Which proves what?

    You are suggesting that XG was correct without factoring in changes in players, management, injuries, etc etc

    Also where does XG factor in the goalkeeper?  "Good" chances aren't always missed, often a keeper saves them.
  • Sponsored links:


  • As I recall their team didn't change that much, and as a consequence they reverted to the norm.  I got a degree of abuse on here for backing them to go down that season (they stayed up by 3pts).

    It does somewhat baffle me why people are so adverse to seeing hard data that can back up or be contrary to an opinion.  Tony Bloom has bought a football club and become a billionaire with an army of people building this sort of data in a building in Camden.  Matt Benham the Brentford owner is another one.  Think I'd rather be on their side!  
  • As I recall their team didn't change that much, and as a consequence they reverted to the norm.  I got a degree of abuse on here for backing them to go down that season (they stayed up by 3pts).

    It does somewhat baffle me why people are so adverse to seeing hard data that can back up or be contrary to an opinion.  Tony Bloom has bought a football club and become a billionaire with an army of people building this sort of data in a building in Camden.  Matt Benham the Brentford owner is another one.  Think I'd rather be on their side!  
    I’ve no idea how many others have tried and failed. If they’re the only people to try it, and it worked, then great. But I’d still need more evidence for an experiment to draw a conclusion. I learnt that much in my science GCSE. 

    I think the the main reason I don’t like it is, it sounds like hipster stuff that American Reddit pseudo-intellectuals spout. People who have never watched a live game, let alone played in one! Football is much harder to turn into an exact science than baseball. 
  • Dazzler21 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.
    But unless you're a bit thick, any attending Charlton fan will tell you that up to this point, we are very clinical.

    It seems to me the stats just confirm what has happened and anyone with a modicum of common sense will already know.

    A bit similar to attendance stats, when the stats say we've had 14,000+ attendances this season in the league.

    Well, if you were there and had a bit of common sense, you already know.
    The stat quantifies our performance rather than just saying 'we are very clinical'
  • JiMMy 85 said:
    As I recall their team didn't change that much, and as a consequence they reverted to the norm.  I got a degree of abuse on here for backing them to go down that season (they stayed up by 3pts).

    It does somewhat baffle me why people are so adverse to seeing hard data that can back up or be contrary to an opinion.  Tony Bloom has bought a football club and become a billionaire with an army of people building this sort of data in a building in Camden.  Matt Benham the Brentford owner is another one.  Think I'd rather be on their side!  
    I’ve no idea how many others have tried and failed. If they’re the only people to try it, and it worked, then great. But I’d still need more evidence for an experiment to draw a conclusion. I learnt that much in my science GCSE. 

    I think the the main reason I don’t like it is, it sounds like hipster stuff that American Reddit pseudo-intellectuals spout. People who have never watched a live game, let alone played in one! Football is much harder to turn into an exact science than baseball. 
    Got the syndicates in Asia doing the same thing too.  If anything xG and its variants success is so well proven now that the market is too saturated (these have been around for 20+ years) and any edge now is so minute you have to bet in the sizes of Blooms, Benhams etc... to make it pay.  I certainly can't find a significant edge in football these days to make it worth my while.
  • Is that the Brentford owner who sacked his manager when they were near the top of the division only to then see them flop?

    Reading did well and then did badly. XG doesn't prove or disprove that, it was already a proven fact.

    Stats are fine but they just stats, it's the almost religious faith placed in them by somevthat is illogical.

    Not everything, or even most things, in life fit neatly in statistical or other boxes.

    People seek order in a chaotic world hence religion, conspiracy theories and over reliance on statistical models.

    Football, far more than other sports, is random. That is its beauty.


  • Is that the Brentford owner who sacked his manager when they were near the top of the division only to then see them flop?

    Reading did well and then did badly. XG doesn't prove or disprove that, it was already a proven fact.

    Stats are fine but they just stats, it's the almost religious faith placed in them by somevthat is illogical.

    Not everything, or even most things, in life fit neatly in statistical or other boxes.

    People seek order in a chaotic world hence religion, conspiracy theories and over reliance on statistical models.

    Football, far more than other sports, is random. That is its beauty.


    Their accuracy is pretty unequivocal over time I'm afraid
  • Is that the Brentford owner who sacked his manager when they were near the top of the division only to then see them flop?

    Reading did well and then did badly. XG doesn't prove or disprove that, it was already a proven fact.

    Stats are fine but they just stats, it's the almost religious faith placed in them by somevthat is illogical.

    Not everything, or even most things, in life fit neatly in statistical or other boxes.

    People seek order in a chaotic world hence religion, conspiracy theories and over reliance on statistical models.

    Football, far more than other sports, is random. That is its beauty.


    Their accuracy is pretty unequivocal over time I'm afraid
    Accuracy of what?

    They are after the event stats?

    Yes, teams who take more of their chances win more often.
  • In so much as that it is unsustainable to outperform or vice versa. 

    I fear that we may be going round in circles with this debate, but ultimately in blindly dismissing data as incapable of explaining your beautiful game you are trying to fight the laws of probability.

    In starting this thread I merely wanted to put some meat on the bones of our performances...is our start sustainable?  Is it too good to be true? etc....

    The fact people seem so determined to ignore facts/data probably explains why Denise Coates is a billionaire.
  • Low-budget team with idiot owner finds itself at the top of the Championship... I mean... I just don’t need hipster stats to tell me that’s a surprise we can’t expect to maintain! 
  • JiMMy 85 said:
    Low-budget team with idiot owner finds itself at the top of the Championship... I mean... I just don’t need hipster stats to tell me that’s a surprise we can’t expect to maintain! 
    And they confirm your opinion.  Isn't that cool?!
  • Sponsored links:


  • Is that the Brentford owner who sacked his manager when they were near the top of the division only to then see them flop?

    Reading did well and then did badly. XG doesn't prove or disprove that, it was already a proven fact.

    Stats are fine but they just stats, it's the almost religious faith placed in them by somevthat is illogical.

    Not everything, or even most things, in life fit neatly in statistical or other boxes.

    People seek order in a chaotic world hence religion, conspiracy theories and over reliance on statistical models.

    Football, far more than other sports, is random. That is its beauty.


    With techniques that have emerged in the last 5-10 years most things now can be modelled using a machine learning algorithm. 
  • Is that the Brentford owner who sacked his manager when they were near the top of the division only to then see them flop?

    Reading did well and then did badly. XG doesn't prove or disprove that, it was already a proven fact.

    Stats are fine but they just stats, it's the almost religious faith placed in them by somevthat is illogical.

    Not everything, or even most things, in life fit neatly in statistical or other boxes.

    People seek order in a chaotic world hence religion, conspiracy theories and over reliance on statistical models.

    Football, far more than other sports, is random. That is its beauty.


    With techniques that have emerged in the last 5-10 years most things now can be modelled using a machine learning algorithm. 
    Really,?

    Read the thread on here on mental health and tell me a machine can model it.

    Love,hate,war, all algorithms or is it just that people want certainty in an uncertain world?

    By the way Coates, like Bloom, is a bookie, a professional that existed, and often thrived, long before Xg came along. My uncle taught me how to calculate the 100 as a kid.
  • Individual cases obviously cannot be modelled but over a large sample, I’m sure there are some very accurate studies with heaps of data regarding mental health.
  • edited August 2019
    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. 
    Last season’s data over the whole season... I’m sure xG was wildly wrong on specific individual occasions but OVER A LARGE ENOUGH SAMPLE, the data was accurate.
  • Our percentage of shots on target is currently sitting at almost 60%. The next closest teams are Sheff Wed and Bristol City at just a touch over 35%.
  • “If Reading can limit their shots”

    The idea of shooting less is that Charlton only shoot when a chance is incredibly good. It’s not shooting for the sake of it. It’s about creating a genuine chance to score a goal. Reading can’t “limit” us if we’re deciding what a golden chance is. That’s completely different to being shot-shy. They might as well say “Reading can win if Charlton don’t score, we’re so smart, can you believe how smart this system is. *Smug face*”
  • JiMMy 85 said:
    “If Reading can limit their shots”

    The idea of shooting less is that Charlton only shoot when a chance is incredibly good. It’s not shooting for the sake of it. It’s about creating a genuine chance to score a goal. Reading can’t “limit” us if we’re deciding what a golden chance is. That’s completely different to being shot-shy. They might as well say “Reading can win if Charlton don’t score, we’re so smart, can you believe how smart this system is. *Smug face*”
    I agree with you. We are definitely limiting our number of shots from outside of the area and this is a key contributing factor.

    83.3% (30/36) of all CAFC shots have come from inside the penalty area. We have also created a total of ten “big chances”.

    It basically highlights a clear philosophy of picking and choosing our moments and resisting all temptation to shoot on sight.

  • If charltons 2 best performing players get less chances they may not do so well

    .... you don’t need a graph to work that out, just fking common sense. 


Sign In or Register to comment.

Roland Out Forever!