2018/2019 Premier League Discussion Part III

HÖÖÖÖÖÖÖÖÖÖÖÖÖÖÖÖÖÖJBEEEEERGGGG

HOLY SHIT

oh man

Sterling with one of the luckiest goals imaginable

sigh

and Southampton concede another one just before HT.. 1:3

That GK from Southampton is quite terrible

I don't get some PL stadium audiences

So City are battering Southampton with brutal pressure first 15 minutes, the home team then do really, really well to fight their way into the game, even get some strong possession vs City, get a fantastic equalizer, the crowd is there all of a sudden.

Then, 10 minutes later Southampton concede an unlucky deflected goal and the crowd goes COMPLETELY quiet.
No chants of encouragement, no defiance, nothing. Stadium goes completely flat even though their team is still in the game.

No wonder they concede easily again in that period

Mr Nick09 wrote:That GK from Southampton is quite terrible

Made 2 unbelievable stops early to keep it at 1:0. Ultimately, City create too much quality chances.

20 mins until we see Ole outclass Howe



proper midtable clash

5 pts off Chelsea if we win, you lads think we can sneak top 4 this season?

No.

Chelsea and Arsenal are poor and the cancer is gone

the correct answer was yes

We'll finish top 3 tbh

especially now Pogbito is focused (#PogMoji in app stores now)

You should be taking it step by step

for example, I think you'll be lucky to win today

the high point of my season, aside from taking a steaming shit on Surag and enjoying the weekly comedy show that is AC Milan, is likely to be this 3 game stretch we've had against Cardiff, Huddersfield and Bournemouth

allow my braggadocio

You'll be 8, not 5, points off Chelsea if you win today. Your crappadocio isn't even tied to reality.

I'll still allow it

Art Morte wrote:You'll be 8, not 5, points off Chelsea if you win today. Your crappadocio isn't even tied to reality.

I've already added + 3 points from when we beat Chelsea though

also, Pogbitoooooooooooooooooo

BEST IN THE FUCKING WORLD

lad is on fire post Mou

lmao it's amazing how liberated this Man Utd team comes across

like a leaden vest has been lifted off everyone

playing like young eager puppies, creative, quick, attacking

I'd be amazed if Mourinho is ever allowed to work again, based on what we have witnessed here in the last couple of years

What's up with this importance placed on xG like its gospel? We are back to whoscored ratings I guess.

To think that we rejected the chance to sign Martial in summer, with Rummenigge claiming we'd be covered in those positions having Gnabry, Coman and Robbéry

He's going to be an absolute monster in no time now that Mourinho is gone

That clown wasted 2.5 years of his career.

rincon wrote:What's up with this importance placed on xG like its gospel? We are back to whoscored ratings I guess.

There's a world of difference between a statistic used by professional companies and gamblers (xG) and a single website's rating.

xG is by no means a perfect stat and there's plenty that needs refining (the name itself, for starters, doesn't communicate entirely what the stat is). However, it has so far proved to be a better predictive stat than any other stat used in football.

There's a logic behind it that I agree with. It assigns a value to chance quality, which can then be used to assess how good teams are at creating high quality chances and defending against those chances. It follows that the best teams create a high volume of good chances and concede very few of the same chances.

So when you see a team like Arsenal who have an xG of around 30 but who have actually scored 42 goals, the suggestion is that Arsenal's attackers have over-performed and/or gotten lucky relative to the actual quality of their play. Over-performance on that scale is concerning because it's not sustainable. Luck runs out, form dips, and so on, and Arsenal reverting back to their supposed level is the difference between 3rd and 6th place.

xG isn't prophetic, but it's a useful measure of how a team is actually performing.

Pogbito with another

Just happy to see Martial and Pogba actually looking like they're enjoying themselves now

Mourinho must have been hell

farfan wrote:That clown wasted 2.5 years of his career.

actually made Martial lose out on winning a World Cup

unbelievable stuff

United are back?

Will be beautiful to see the "Solkjaer" vote rise and rise in the RedCafe Next Manager poll during the coming days and weeks

we still look very shaky at the back tbh but Top 4 doesn't look unobtainable

fantastic assist from Martial

Jay29 wrote:

rincon wrote:What's up with this importance placed on xG like its gospel? We are back to whoscored ratings I guess.

There's a world of difference between a statistic used by professional companies and gamblers (xG) and a single website's rating.

xG is by no means a perfect stat and there's plenty that needs refining (the name itself, for starters, doesn't communicate entirely what the stat is). However, it has so far proved to be a better predictive stat than any other stat used in football.

There's a logic behind it that I agree with. It assigns a value to chance quality, which can then be used to assess how good teams are at creating high quality chances and defending against those chances. It follows that the best teams create a high volume of good chances and concede very few of the same chances.

So when you see a team like Arsenal who have an xG of around 30 but who have actually scored 42 goals, the suggestion is that Arsenal's attackers have over-performed and/or gotten lucky relative to the actual quality of their play. Over-performance on that scale is concerning because it's not sustainable. Luck runs out, form dips, and so on, and Arsenal reverting back to their supposed level is the difference between 3rd and 6th place.

xG isn't prophetic, but it's a useful measure of how a team is actually performing.

I don't understand how when seeing a 50% error (Arsenal) the suggestion is that reality is so skewed, instead of the model being inaccurate. It's a number counting the amount of chances being created and assigning value to the perceived quality of those chances based on average results. So it's inherently flawed. It removes finishing from the game. Finishing isn't luck or an accident, clubs pay hundred of millions for strikers because it's a skill. So is a "quality chance" the same quality for Icardi as it is for Pepe? No. An average chance for Ronaldo can be a difficult chance for Morata and as such chances should be weighted differently depending on the players taking them as well as the GKs trying to save them. But for that, you need a much more advanced model than the current one.

That's of course assuming that the way to quantify chances and assign quality is close enough to being perfect. A big assumption as a chance for Robben looks very different than a chance for Lukaku for example, but that's ok for the most part.

It's good for averages and for fun, but to pin so much importance to it in a place like GL it's weird. How much more valuable is it to look at shoots/goal than it is to look at xG? it's similar.