The Championship 2020-21

Discussion in 'The Hornets' Nest - Watford Chat' started by Markoa$, Jul 26, 2020.

  1. ITK platypus

    ITK platypus Squad Player

    When I saw Warnock claim the player could have 'lost an eye' I sighed as a I really hate exaggerations like this. However, I then watched the challenge and it was genuinely horrifying. Would have done real damage.
     
    PowerJugs and wfc4ever like this.
  2. Steve Leo Beleck

    Steve Leo Beleck Squad Player

    Photos of the injury are horrifying too.
     
  3. Forzainglese

    Forzainglese Reservist

    So what is the referee thinking, then?
     
  4. UEA_Hornet

    UEA_Hornet First Team Captain

    "Hmmm, I wonder what's for tea? Ooo...wait a minute, what's happened there?"
     
  5. onion8837

    onion8837 Reservist

    Liked for being (1) amusing and (2) painfully accurate
     
    wfc4ever likes this.
  6. wfc4ever

    wfc4ever Administrator Staff Member

    Hopefully results stay as they are .

    Swansea Brentford 0-0
    Boro a goal down.
     
  7. I Blame Pozzo

    I Blame Pozzo First Team

    Swans down to 10,Naughty Naughton.
     
  8. I Blame Pozzo

    I Blame Pozzo First Team

    And 1-0 Bees.
     
  9. hornetfan

    hornetfan Academy Graduate

    Now 1-1. Swans equalized despite being down to 10.
     
  10. I Blame Pozzo

    I Blame Pozzo First Team

    Yes Conor Hourihane and the Nolan Sisters with the assist.
     
  11. IRB

    IRB THe artist formally know as ImRonBurgundy?

    Draw probably the best result for us
     
  12. Chumlax

    Chumlax Squad Player

    Oh wow, it's almost over, please god let it stay this scoreline.
     
  13. onion8837

    onion8837 Reservist

    Boro 2 down now - is that 3 home defeats on the spin without scoring ?
     
  14. wfc4ever

    wfc4ever Administrator Staff Member

    Brentford will be disappointed.

    A goal and man up but don't win.

    Notice Swansea scored from a free kick with a good loan signing.

    :rolleyes:
     
    Jumbolina and SkylaRose like this.
  15. Hogg-DEENEY!!!

    Hogg-DEENEY!!! Squad Player

    Hard to say exactly how it'll pan out, but on paper, a good result for us. Mind you, if we don't get some signings through the door in the next few days, the Middlesbrough result might more more relevant...
     
    SkylaRose likes this.
  16. SkylaRose

    SkylaRose Administrator Staff Member

    Good results. Boro are in a rut atm but still a long way to go. Sort of happy Barnsley didn't win either, dark horses maybe?
     
  17. IRB

    IRB THe artist formally know as ImRonBurgundy?

    Brentford... a goal up and a man up... didn’t win

    I’m getting deja vu here
     
    UEA_Hornet likes this.
  18. I Blame Pozzo

    I Blame Pozzo First Team

    Yes.
    Birmingham,PNE and RUFC.
    Warming up nicely for the Budgies!
     
  19. lowerrous

    lowerrous First Team

    He's no James Garner!
     
  20. LeedsOrn

    LeedsOrn Reservist

    A lot of discussion on here on whether our form is sustainable. On this evidence, it seems that of all our rivals, Swansea’s form is least sustainable and that if anything our results haven’t matched our performances (in terms of xG and xGconceded).
    https://twitter.com/eflstats/status/1354563783084093443?s=21

    I think that the fact that the style is quite turgid masks the fact that every game we create several big chances and concede none or very few.
     
  21. Chumlax

    Chumlax Squad Player

    We've been over the shortfall between xG and actual goals scored on here before quite recently, and there is one very simple answer...

    Louorns did points out under the line that this xG table includes penalties, though, which (if true) would indeed seem to skew it quite un-representatively in our favour.
     
  22. LeedsOrn

    LeedsOrn Reservist

    Definitely our finishing is a problem. And it was last season too. But creating chances and not conceding them is a sustainable way to grind out results.

    I didn’t really get @LouOrns’ point on Twitter. I don’t think we’ve received more unjust penalties than any other teams in the division. In fact, we’ve had quite a few good penalty shouts rejected and I think if we keep getting our dangerous players in and around the box, they’ll be able to keep drawing fouls. If you think that penalty distribution is purely random then yes, variance will come to bite us. But it isn’t - it’s a product of certain types of dangerous attacking play.

    Furthermore, even if you take out our penalties from the equation - 7 penalties is equivalent to around 5 xG - we would still have one of the highest xG and and highest xGD in the league.

    Lastly, the metric includes penalties for all teams, not just for us. Norwich have had the same number of penalties as us and have a worse xG, xGA and XGD - are their results unsustainable?Brentford have had 4, so 3 less. Remove penalties from the equation and we’d still have more xG than Norwich and be equal with Brentford (though I think Boro and Bournemouth would then have more xG).

    For what it’s worth I think xPoints, which the table also shows, is not a very useful metric as it doesn’t really show/factor in game state: our 2-0 defeat at Huddersfield, for instance, would have earned us 3 xPoints, since we created good chances while chasing the game. But I think xG, xGA and xGD are helpful for showing performance across the season, and we’ve been trending up under Xisco too. I think it’s been turgid viewing at times this season but theses stats suggest that our points output seems as sustainable as all of our rivals bar Brentford.
     
  23. Chumlax

    Chumlax Squad Player

    Sure, I'm not rubbishing your point at all and I do totally accept that to some extent earning penalties is the result of dangerous attacking play.

    I do think it's very interesting how much of a disparity there is between this xG rating and 538's non-shot xG rating/graph, which I posted here a couple of weeks ago. I imagine the Ivic-less weeks between then and now would explain some of the impact, but I'd be interested to see an up-to-the-minute non-shot xG rating - I can't see it featuring so wildly optimistic an upturn as this one just because Ivic is no longer here.
     
  24. LeedsOrn

    LeedsOrn Reservist

    One of the issues that I have with xG is that different models often produce substantially different outcomes. I don’t know if that is a problem of differing methodologies or data collection issues but it leaves me wondering which to trust and muddies the waters a little bit.

    With 538, something I’ve noticed is that their match by match stuff shows us often well clear of our opponents on shot-based xG and less stellar on non-shot xG. Don’t really know what to make of that.

    That said, 538 has their team ranking metric which is reflective of what they believe is the underlying quality of each squad based on their performances. They consider Brentford to be clearly the best team in the league, and is, Norwich and Bournemouth roughly equal. Swansea are then considered fifth best, but quite a bit behind.
     
  25. lendal

    lendal Reservist

    A bit slow here....WTF is 'XG' all about?


    Sent from my iPad using Tapatalk
     
    The Voice of Reason likes this.
  26. PowerJugs

    PowerJugs Doyley Fanatic

    xG = Expected goals.
    Quoting @reids from another platform: "Basically nerds went through and looked at 500,000 shots and determined the chance of a goal from a specific location. So penalties for instance count as 0.76 xG as on average there's a 76% chance of scoring a penalty."
     
    lendal likes this.
  27. reids

    reids First Team

    This explains it in a bit more depth and with a nice little quiz at the end: https://www.bbc.co.uk/sport/football/40699431
     
    lendal likes this.
  28. Jumbolina

    Jumbolina First Team

    Is it calculated just from the area of the pitch you shoot from or does it account for if you are unmarked/off balance and if the ball is stationary or coming at you full pelt? Presume it can only do the former?
     
  29. reids

    reids First Team

    Depends on the complexity of the model (as lots of different companies have their own xG model which tracks different things and subsequently gives different results) but a majority just do the former, I imagine some of the non-public xG stuff will track more detailed things like the stuff you mentioned. It's a really useful metric with some interesting use cases, people downplay it a lot: "I don't care about xG, I care about actual goals" or "football isn't played on a spreadsheet" are common comments on social media whenever xG is mentioned, but the facts are that every pro team will use xG in some way - whether it be in analysis, recruitment or even the boardroom, and if every pro team is using it then that would suggest that it isn't an entirely pointless statistic.

    It's had an interesting influence on the game as well, teams are (over the past 5 years or so) taking significantly less long-range shots than previously which has correlated well with the introduction and principles of xG (taking low quality shots = bad)
     
  30. WillisWasTheWorst

    WillisWasTheWorst Its making less grammar mistake's thats important

    I always think 76% chance of scoring a penalty is very low for an established penalty taker. If they didn't do all the stop/start, skip and Panenka stuff etc., I'm sure it would be up at about 90%.
     
  31. reids

    reids First Team

    Yeah, Deeney is on 85% (36 out of 42 taken)!
     
  32. Steve Leo Beleck

    Steve Leo Beleck Squad Player

    This is my issue with it. Just looking at zone shot was taken and body part used is never going to tell you much at all. Football is a dynamic, fluid game that doesn't lend itself well to this kind of thing - there are tens of other factors in addition to the ones you listed - spin on the ball, position of defender, goalkeeper, how much the attacker is stretching to make contact, whether it's a volley, half volley, on the ground, behind you. One of the factors that seems to be considered in xG models is whether or not it's a "big chance", which has to be subjective (even if only on setting the parameters of what constitutes a big chance).

    I suppose if you have a large enough sample you can draw some conclusions and the example that Reids gives about not taking as many long shots is interesting. But looking at individual matches or small samples of xG seems pretty flawed to me.
     
  33. UEA_Hornet

    UEA_Hornet First Team Captain

    Interesting. As a fan at a game I'd much rather watch my team have 10 wild shots from range and at least have the statistically improbable hope of 1 going in, than watch an overly cautious midfielder turn back inside again and trigger another backwards passing move into and across the defence. Football has tilted far too far towards the latter in my view.
     
  34. reids

    reids First Team

    Definitely, it's getting more and more advanced though (the Statsbomb xG model is one of the best out there and they've added defender (and goalkeeper!) positioning and shot height into their model - https://statsbomb.com/2020/07/statsbomb-release-expected-goals-with-shot-impact-height/) which has changed the values quite significantly.

    Small sample sizes is one of the biggest flaws and misconceptions around xG and a majority of the analytics community don't use xG for individual matches or small sample sizes as variance and other factors can have a big effect that can be misleading.
     
    PowerJugs likes this.
  35. Chumlax

    Chumlax Squad Player

    Presumably it means that the general pattern is that we manage 'somehow' to create a certain amount of incredibly high xG chances, that Gray then fu cks up royally, as a result of just enough influence from the overarching quality of our attacking players, but create very little outside that/it's limited to those few major moments, with/because of our overall player being quite undefined and lacking in cohesion in an attacking sense? It's no secret that we clearly struggle to arrange any kind of cohesive attacking plan in the final third - it's been an issue for years.

    The Swansea thing is interesting - presumably that is down to just how much they rely on their defensive prowess, and how relatively unproductive their attack is.
     

Share This Page