or to join or start a new Discussion

Articles/all comments
These 1090836 comments are related to an article called:

Good News

Page 43622 of 43634

posted 6 days, 23 hours ago

comment by Phenom (U20037)
posted 46 seconds ago
does the sterlingarea not count as north of edinburgh
----------------------------------------------------------------------

It does tbf, there may be 1

posted 6 days, 23 hours ago

comment by Hengy (U9129)
posted 9 hours, 11 minutes ago
I make YouTube videos with ads and make money
----------------------------------------------------------------------
What's the YouTube videos about?

posted 6 days, 23 hours ago

I think Alton Towers > Thorpe Park

But i havent been to either in like 10 years

posted 6 days, 23 hours ago

comment by Bãleș left boot - Palhinha my goat (U22081)
posted 1 minute ago
comment by Hengy (U9129)
posted 9 hours, 11 minutes ago
I make YouTube videos with ads and make money
----------------------------------------------------------------------
What's the YouTube videos about?
----------------------------------------------------------------------
matth

comment by Hengy (U9129)

posted 6 days, 23 hours ago

comment by Bãleș left boot - Palhinha my goat (U22081)
posted 2 minutes ago
comment by Hengy (U9129)
posted 9 hours, 11 minutes ago
I make YouTube videos with ads and make money
----------------------------------------------------------------------
What's the YouTube videos about?
----------------------------------------------------------------------
Subscribe, like and hit that notification bell

posted 6 days, 23 hours ago

Doakes memes absolutely everywhere now

comment by Phenom (U20037)

posted 6 days, 23 hours ago

comment by Cole Blarmer (U14547)
posted 8 minutes ago
Doakes memes absolutely everywhere now
----------------------------------------------------------------------
whos Doakes? must be an upcoming england cricketer with that name

posted 6 days, 22 hours ago

comment by Phenom (U20037)
posted 20 minutes ago
comment by Cole Blarmer (U14547)
posted 8 minutes ago
Doakes memes absolutely everywhere now
----------------------------------------------------------------------
whos Doakes? must be an upcoming england cricketer with that name
----------------------------------------------------------------------
he was the Bay Harbor Butcher

posted 6 days, 22 hours ago

More of a Clay Davis man than Doakes, but the geezer who did the impression of Doakes saying small fries amoungst other variations was very good.


Shiiiiieeeeeeeeed

posted 6 days, 22 hours ago

comment by Cole Blarmer (U14547)
posted 20 minutes ago
comment by Phenom (U20037)
posted 20 minutes ago
comment by Cole Blarmer (U14547)
posted 8 minutes ago
Doakes memes absolutely everywhere now
----------------------------------------------------------------------
whos Doakes? must be an upcoming england cricketer with that name
----------------------------------------------------------------------
he was the Bay Harbor Butcher
----------------------------------------------------------------------
Nah Dexter framed him

posted 6 days, 19 hours ago

comment by Blackpolespur (U9242)
posted 2 hours, 21 minutes ago
comment by Cole Blarmer (U14547)
posted 20 minutes ago
comment by Phenom (U20037)
posted 20 minutes ago
comment by Cole Blarmer (U14547)
posted 8 minutes ago
Doakes memes absolutely everywhere now
----------------------------------------------------------------------
whos Doakes? must be an upcoming england cricketer with that name
----------------------------------------------------------------------
he was the Bay Harbor Butcher
----------------------------------------------------------------------
Nah Dexter framed him
----------------------------------------------------------------------
Wild conspiracy theory

posted 6 days, 18 hours ago

Tifo had a long chat about whether you guys are good or not this season, was quite interesting

posted 6 days, 17 hours ago

Meh I thought the analysis was alright but approached only from an negative angle, attempting to justify the stats. I mean, it was briefly mentioned that we're overperforning xg so much while in the next breath saying how bad the forwards are aside from Kudus.

For me it's indicative of the modern trend of having to be right. A more interesting piece to me would be why these 'underlying numbers' don't reflect actual football. No one watching the Leeds game would say we didn't deserve to win yet it's made out on the podcast as if we were lucky or summink

posted 6 days, 17 hours ago

I dont think the expected numbers matter much over such small samples personally

posted 6 days, 17 hours ago

like especially the xG reading in a single game shouldnt be focused on much

comment by T-Bag (U11806)

posted 6 days, 16 hours ago

Unless you're an Arsenal fan bemoaning how lucky Liverpool are

posted 6 days, 16 hours ago

different xG models being different annoys me too

posted 6 days, 15 hours ago

Well when it’s less quantifiable a data point it means that models have to set their own parameters. It’s surprising common in a lot of metrics you will take for granted.

posted 6 days, 15 hours ago

I often think xG doesn’t model well with counter attacking teams who create most their chances on the counter

posted 6 days, 15 hours ago

Cant they just copy the Opta modelling

comment by T-Bag (U11806)

posted 6 days, 15 hours ago

comment by Edinspur (U1109)
posted 35 minutes ago
I often think xG doesn’t model well with counter attacking teams who create most their chances on the counter
----------------------------------------------------------------------
Does it take into account who is crossing the ball, etc?

posted 6 days, 14 hours ago

comment by Cole Blarmer (U14547)
posted 34 minutes ago
Cant they just copy the Opta modelling
----------------------------------------------------------------------

I think it’s statsbomb that have the most ‘complete’ model, but it’s behind a paywall

posted 6 days, 14 hours ago

comment by T-Bag (U11806)
posted 24 minutes ago
comment by Edinspur (U1109)
posted 35 minutes ago
I often think xG doesn’t model well with counter attacking teams who create most their chances on the counter
----------------------------------------------------------------------
Does it take into account who is crossing the ball, etc?
----------------------------------------------------------------------

That would be counterintuitive. The whole point of the model is that it creates a baseline average for all the situations to accurately create an appropriate data set.

The more advanced models (like statsbomb) take into account bodies between shooter and gk, defensive positioning etc

posted 6 days, 14 hours ago

the typical models dont give much weight to chances that come via crosses yeah? which is probably a valid point when looking at all the data

posted 6 days, 14 hours ago

comment by Edinspur (U1109)
posted 1 hour, 9 minutes ago
I often think xG doesn’t model well with counter attacking teams who create most their chances on the counter
----------------------------------------------------------------------

To further my point with this, in a really basic simplistic way. Teams who effectively create good xG through counters do so by few, high quality chances; whereas teams who do it through dominating possession tend to have lots of chances of ‘lower’ quality.

As a result a misplaced final ball in a counter attacking holds so much more potential than another pass in a high possession system. I’ve found it is something that can’t be modelled and why so many good counter teams ‘overperform’ their xG

Page 43622 of 43634

Sign in if you want to comment