Bruce with Roy Keane in the Roy Evans role.
If we sack him (I'm still not sure that they actually will do it) then I can see it being Carrick until the end of the season and then someone like Ten Hag.
Bruce would be the dream.
Brain dead pass of the year.
I'm still a right old dinosaur with this playing out from the back business. You've got to have such a high rate of not-fucking-up to avoid gifting goals from it.
One of Altrincham's equalisers on Tuesday was from our keeper fannying about. There's a level to play that kind of stuff and it's not Conference North.
It's probably better than a hopeful punt that comes straight back at you.
Arsenal's problem is that they play it short, but then the 3rd pass tends to be a massive lump up field to van Dijk or Matip.
I've just seen the Man Utd highlights.
A relegation level performance.
I'm not sure the statswankers have enough data on the value of retaining possession versus the likelihood of gifting goals through incompetence.
This is my problem with xG. Could the xG lovers tell me who fits the mould of this supposed ‘average player’ we’re basing the stats on?
As someone who has witnessed the change from 'clear the ball at every opportunity for 90 minutes' to 'fanny about', the results might not be better but I'd take the latter any day.
Having no prospects either way makes that easier though.
Super Taki
It's easier to spot mistakes that lead to a goal rather than marginal battles lost over possession that lead to a goal. That's why cretins still believe we need to pump balls aimlessly up and show GRIT and PASSION for our BATTLES.
I hate that we played kinda well and are going to get pummelled because of the optics of the score line.
It just seems dumb to me that you’re almost punished more for actually still trying to win when 2-0 down instead of just giving up.
All the result does is show where Arsenal are at the moment, much as the fistings they took off City and Chelsea did, and perhaps their otherwise good run does too. They're miles off the top teams, but improving to be at the head of the midfield chasers, maybe. Still think they're pretty average.
Newcastle at home then Man Utd and Everton away. If we play to that level today I think we come away with 9 points and sit in 4th.
37% possession and 0.31 xG. Played well, aye.
Just seen Burnley scores 3 with a 0.9 xG. Seems a really good and useful stat.
All these xStats are complete dung built for an era where people only watch goals on their phone.
Rio Ferdinand, "I could be a Sporting Director. I'd be amazing at it."
Jake Humphrey, "Who should Man Utd bring in as manager?"
Rio Ferdinand, ".....I dunno."
The correct answer by the way…
Steve Clarke.
Can't wait for the apology tweets.
People said Arsene Wenger was soft on the players but Phillipe Senderos cried after a loss and never played for Arsenal again.
Arsenal had us at arms length really until the touchline kick off. Brought the crowd back into it and with it, a bit of zip into the play.
After 1-0 they really melted down against the press.
Rio is the thickest bloke in football.
What player though? Isn’t the average player in like League One. It doesn’t work as a system.
I don't think Ole is up to it. Shame the board can't see something obvious the useless cunts.
Sergio Aguero is retiring because of his heart issue.
The xG value for a given shot is based on historic data, different models use different datasets.
The value of an individual shot will have an error bar on it because the model can't know everything about it, but if you have a big enough sample for a team or player, you can come to some fairly straightforward insights like "Messi is 25% better at finishing than the average player".
What do you want from it for it to work? Like any metric, it comes with caveats and requires context for it to be meaningful.
But who is the average player that Messi is 25% better than. That’s my question. That’s the context that is missing from all this. That’s my point.
Quincy proving my point by repeatedly posting obvious trolls and taking them as facts.
This is vastly oversimplifying but if Messi takes a shot that based on historical data has a 10% chance of resulting in a goal, he is 12.5% likely to score.
There isn't an average player, though the vast majority of players perform pretty close to expectation over long term. Aubameyang has 135 goals from shots that amount to 134 expected goals since 2014 so I suppose it could be him.
The model generates an average from all players, surely? Then people are individually compared to that average.
Whether it's necessary is one thing, but it seems to be fairly well put together.
Also isn’t a difference between 10 and 12% when you average 3-4 chances a game almost minuscule? It just seems like trying to turn soccer into baseball in a way that’s truly impossible.
I've explained it fine, and it's hardly some underground secret that doesn't hasn't had a million explainers written on it on BBC sport. You just seem desperate to not understand it and/or not read my replies.
I have already fielded both of these.
Last edited by -james-; 20-11-2021 at 10:02 PM.