United and xG (now that Ole is gone will things change?)

TMDaines

Fun sponge.
Joined
Sep 1, 2014
Messages
13,997
United’s expected points after 27 games in each season since Ferguson retired:

20/21 (Ole): 45.10
19/20 (Ole): 49.16
18/19 (Mou/Ole): 43.78
17/18 (Mou): 46.81
16/17 (Mou): 52.88
15/16 (LVG): 41.98
14/15 (LVG): 43.74

We’re still going absolutely nowhere.

Source: Understat
 

Classical Mechanic

Full Member
Joined
Aug 25, 2014
Messages
35,216
Location
xG Zombie Nation
Looking at the early xG data and we're going to need a huge slice of fortune to win the league if we keep performing like this. City and Liverpool are way ahead of us on quality of chance creation and City significantly ahead defending against chance creation. If you were to bet on a side to win it then it would be City I'm afraid. For xGA only City are significantly ahead of us but for xG we're lacking considering who we've played so far. Interestingly our xG (7.6) and xGA (5.5) is exactly the same as Chelsea at the moment. DDG is only behind Allison for conceding less than expected so far but the worry is that he's still very weak on the other aspects of goalkeeping so we're relying on him protecting the goal with amazing saves. He has done it for a season before but it looks like we'll be relying on indvidual brilliance at both ends of the pitch to win the league if we continue in this vein.

A sober assessment would be that we probably won't win the league this season and that City and Liverpool are favourites in that order. 5 games is a small sample but the superior team play of City and Liverpool has been a recurring theme in recent seasons. They are better coached sides and this is an important difference in my opinion.
 

Borys

Statistics Wizard
Joined
May 10, 2013
Messages
9,094
Location
Bielsko Biala, Poland
Looking at the early xG data and we're going to need a huge slice of fortune to win the league if we keep performing like this. City and Liverpool are way ahead of us on quality of chance creation and City significantly ahead defending against chance creation. If you were to bet on a side to win it then it would be City I'm afraid. For xGA only City are significantly ahead of us but for xG we're lacking considering who we've played so far. Interestingly our xG (7.6) and xGA (5.5) is exactly the same as Chelsea at the moment. DDG is only behind Allison for conceding less than expected so far but the worry is that he's still very weak on the other aspects of goalkeeping so we're relying on him protecting the goal with amazing saves. He has done it for a season before but it looks like we'll be relying on indvidual brilliance at both ends of the pitch to win the league if we continue in this vein.

A sober assessment would be that we probably won't win the league this season and that City and Liverpool are favourites in that order. 5 games is a small sample but the superior team play of City and Liverpool has been a recurring theme in recent seasons. They are better coached sides and this is an important difference in my opinion.
Probably the best one-sentence description of the season so far. We're on the extreme end of both goalkeeping stat (PSxG+/-) and scoring stats (np:G-xG).
So basically if we want to maintain this run, De Gea, Ronaldo and Greenwood need to continue their purple patch.
 

AjaxCunian

vexingwijsneus
Joined
Mar 10, 2021
Messages
4,241
Supports
Ajax & United
I think the sample size is still a bit small and also the distribution of quality of teams played.

City lack a striker though and there finishers up front are erratic, despite their great xG they might struggle.
 

sullydnl

Ross Kemp's caf ID
Joined
Sep 13, 2012
Messages
34,063
Agree, the underlying data will be concerning if it keeps trending like that. Particularly as we've had easier fixtures than at least two of our rivals in that small sample size of games.

Last season Man City's xG difference was +42.0, Chelsea's (despite a good chunk of the season with Lampard as manager) was +31.2, Liverpool +27.3 and ours +18.0. I don't see how we can win the league without huge improvements there, particularly as we were already overperforming our xG by the second most in the league and Liverpool and particularly Chelsea are unlikely to underperform their xG as they did last season.

Yesterday's game tells its own little story. 27 shots but still ended up with just 1.5xG to their 1.4xG
 

Siorac

Full Member
Joined
Sep 1, 2010
Messages
23,815
Yesterday's game tells its own little story. 27 shots but still ended up with just 1.5xG to their 1.4xG
It tells a very old story: we're not very good at consistently creating high-quality chances.
 

Pogue Mahone

The caf's Camus.
Joined
Feb 22, 2006
Messages
133,934
Location
"like a man in silk pyjamas shooting pigeons
Looking at the early xG data and we're going to need a huge slice of fortune to win the league if we keep performing like this. City and Liverpool are way ahead of us on quality of chance creation and City significantly ahead defending against chance creation. If you were to bet on a side to win it then it would be City I'm afraid. For xGA only City are significantly ahead of us but for xG we're lacking considering who we've played so far. Interestingly our xG (7.6) and xGA (5.5) is exactly the same as Chelsea at the moment. DDG is only behind Allison for conceding less than expected so far but the worry is that he's still very weak on the other aspects of goalkeeping so we're relying on him protecting the goal with amazing saves. He has done it for a season before but it looks like we'll be relying on indvidual brilliance at both ends of the pitch to win the league if we continue in this vein.

A sober assessment would be that we probably won't win the league this season and that City and Liverpool are favourites in that order. 5 games is a small sample but the superior team play of City and Liverpool has been a recurring theme in recent seasons. They are better coached sides and this is an important difference in my opinion.
You don’t need xG to know that’s the way this season will likely pan out.
 

Classical Mechanic

Full Member
Joined
Aug 25, 2014
Messages
35,216
Location
xG Zombie Nation
You don’t need xG to know that’s the way this season will likely pan out.
That's true but I think for most fans on the question of Ole there's the battle between the emotional self and the rational self. We may win a title with Ole with everything running for us in a given season but there's nothing to suggest we can sustain success with him in charge. We're at a point now where we have a very good squad and should be looking to match Liverpool and City. We need to look beyond Ole and for a new manager if we want to capitalise on the talent we now have in my opinion.

I would say that xG helps to better quantify just how far behind City and Liverpool we are when it comes to creating quality chances and how superior their coaching is in that regard.
 

MDFC Manager

Full Member
Joined
Dec 26, 2005
Messages
24,303
A sober assessment would be that we probably won't win the league this season and that City and Liverpool are favourites in that order. 5 games is a small sample but the superior team play of City and Liverpool has been a recurring theme in recent seasons. They are better coached sides and this is an important difference in my opinion
Agree with everything especially the conclusion. No amount of player upgrades will help us bridge the gap enough to be able to win the league as the difference will continue to be the quality gap in coaching.
 

EtH

Full Member
Joined
Aug 29, 2021
Messages
2,712
Not really surprising our xG was piss with Martial stood still upfront all match is it.

I am more interested in how we look in the coming months. With all the attacking options now at our disposal our xG and points tally should improve on last season.
 
Last edited:

ivaldo

Mediocre Horse Whisperer, s'up wid chew?
Joined
Nov 15, 2012
Messages
28,699
Another example of xG confirming what was obvious to anyone with functional eyes and a brain. We were all huff and puff with zero cutting edge. Despite what @ivaldo was trying to convince me.
How much did the Martial decision to pass instead of finish account for? How about the nailed on pen that we missed?
 

largelyworried

Full Member
Joined
Feb 10, 2021
Messages
2,101
But someone told me that Tuchel is the best manager in the game whereas Ole is the worst.
They've played away at Arsenal and Liverpool, and half of their game with Liverpool they had 10 men. So the fact their xG is the same both directions does not look good on us.
 

OleBoiii

New Member
Joined
Oct 4, 2019
Messages
6,021
They've played away at Arsenal and Liverpool, and half of their game with Liverpool they had 10 men. So the fact their xG is the same both directions does not look good on us.
You could also argue that it's easier to create big chances against teams that dare to play a more open game. The exception being if the opposing team is vastly superior(which isn't the case here).

So far we're pretty much repeating the pattern of seasons past under Ole.
I hope that we do. If we continue with the same pattern then we will finish the season with 99 points and 99 goals :p
 

cyberman

Full Member
Joined
May 26, 2010
Messages
37,331
Isn’t this down to simply having a front 4 who can score from every angle and are extremely clinical? Sometimes these stats doesn’t actually reveal anything.
Son is the perfect example
https://www.independent.co.uk/sport...ham/son-heung-min-xg-harry-kane-b1782968.html
Greenwood is above XG as well
https://understat.com/player/7490
It looks as if the cut in and score from tighter angles bluffs xg a bit and doesn’t tell the full story?
Salah over performs as well
https://understat.com/player/1250
It’s hard for me to to find a top wide player who dives under xg. Even Rashford is wasteful and he is more or less bang on with his.

Then add in the fact we have scored the majority of our goals from players who dont play centrally or come from deep then technically we score a lot of goals from sub optimal situations?
My theory is it’s down to Bruno. He is a unique player in the AM position with his goal rate and that adds an extra finisher to the front line that other sides dont have so it’s less chances we have to create so we can score.
I expect that to change with a killer like Ronaldo in the middle though.
 

Tarrou

Full Member
Joined
May 13, 2013
Messages
25,635
Location
Sydney
Probably the best one-sentence description of the season so far. We're on the extreme end of both goalkeeping stat (PSxG+/-) and scoring stats (np:G-xG).
So basically if we want to maintain this run, De Gea, Ronaldo and Greenwood need to continue their purple patch.
Or we could just improve
 

cyberman

Full Member
Joined
May 26, 2010
Messages
37,331
Probably the best one-sentence description of the season so far. We're on the extreme end of both goalkeeping stat (PSxG+/-) and scoring stats (np:G-xG).
So basically if we want to maintain this run, De Gea, Ronaldo and Greenwood need to continue their purple patch.
Yes, we will need our strikers to score goals if we want to win the league?
 

Borys

Statistics Wizard
Joined
May 10, 2013
Messages
9,094
Location
Bielsko Biala, Poland
Yes, we will need our strikers to score goals if we want to win the league?
To be honest I would not be surprised if we ended the season on extreme (positive) end of goals scored - xG ratio, in a team of Ronaldo Greenwood and Bruno that's possible. I wouldn't expect the same trend to continue on the other side of the pitch.
But at the same time, McTominay is back so I'd expect to see some improvement as we won't be forced to play makeshift midfields that often.
 

Zlatan 7

We've got bush!
Joined
May 26, 2016
Messages
11,797
Isn’t this down to simply having a front 4 who can score from every angle and are extremely clinical? Sometimes these stats doesn’t actually reveal anything.
Son is the perfect example
https://www.independent.co.uk/sport...ham/son-heung-min-xg-harry-kane-b1782968.html
Greenwood is above XG as well
https://understat.com/player/7490
It looks as if the cut in and score from tighter angles bluffs xg a bit and doesn’t tell the full story?
Salah over performs as well
https://understat.com/player/1250
It’s hard for me to to find a top wide player who dives under xg. Even Rashford is wasteful and he is more or less bang on with his.

Then add in the fact we have scored the majority of our goals from players who dont play centrally or come from deep then technically we score a lot of goals from sub optimal situations?
My theory is it’s down to Bruno. He is a unique player in the AM position with his goal rate and that adds an extra finisher to the front line that other sides dont have so it’s less chances we have to create so we can score.
I expect that to change with a killer like Ronaldo in the middle though.
Nice post. I hate stats of expected goals from certain positions on the field. Every team plays and creates chances differently.
 

Revan

Assumptionman
Joined
Dec 19, 2011
Messages
49,635
Location
London
You could also argue that it's easier to create big chances against teams that dare to play a more open game. The exception being if the opposing team is vastly superior(which isn't the case here).
You could also argue that the world is flat, to be fair. Just that it is a poor argument.
 

Revan

Assumptionman
Joined
Dec 19, 2011
Messages
49,635
Location
London
Isn’t this down to simply having a front 4 who can score from every angle and are extremely clinical? Sometimes these stats doesn’t actually reveal anything.
Son is the perfect example
https://www.independent.co.uk/sport...ham/son-heung-min-xg-harry-kane-b1782968.html
Greenwood is above XG as well
https://understat.com/player/7490
It looks as if the cut in and score from tighter angles bluffs xg a bit and doesn’t tell the full story?
Salah over performs as well
https://understat.com/player/1250
It’s hard for me to to find a top wide player who dives under xg. Even Rashford is wasteful and he is more or less bang on with his.

Then add in the fact we have scored the majority of our goals from players who dont play centrally or come from deep then technically we score a lot of goals from sub optimal situations?
My theory is it’s down to Bruno. He is a unique player in the AM position with his goal rate and that adds an extra finisher to the front line that other sides dont have so it’s less chances we have to create so we can score.
I expect that to change with a killer like Ronaldo in the middle though.
xG accounts the quality of chances. You outperform xG if you are lucky, or your strikers are clinical. We have arguably the most clinical player in the world, and Bruno/Greenwood are quite clinical too, so it is likely that we are gonna outperform xG. At the same time, the lack of high-quality chances is gonna catch us, and as usual we will finish 10+ points below the champions. Very hard to be lucky for an entire season.

There was the same argument when Ole came as manager. xG is for nerds, table doesn’t lie, we have great strikers etc, while everyone who knows something about this was urging that it is gonna change soon. And then we want to crap soon after.
 

Zlatan 7

We've got bush!
Joined
May 26, 2016
Messages
11,797
xG accounts the quality of chances. You outperform xG if you are lucky, or your strikers are clinical. We have arguably the most clinical player in the world, and Bruno/Greenwood are quite clinical too, so it is likely that we are gonna outperform xG. At the same time, the lack of high-quality chances is gonna catch us, and as usual we will finish 10+ points below the champions. Very hard to be lucky for an entire season.

There was the same argument when Ole came as manager. xG is for nerds, table doesn’t lie, we have great strikers etc, while everyone who knows something about this was urging that it is gonna change soon. And then we want to crap soon after.
What is a high quality chance
 

Revan

Assumptionman
Joined
Dec 19, 2011
Messages
49,635
Location
London
What is a high quality chance
Chances that are easy to score. For example Noble’s opportunity at the end of the match yesterday, or one on one striker vs keeper cases, or the million tap ins City score every season.
 

alexanderplatz

Full Member
Joined
Jan 24, 2010
Messages
757
Location
Ireland
That's true but I think for most fans on the question of Ole there's the battle between the emotional self and the rational self. We may win a title with Ole with everything running for us in a given season but there's nothing to suggest we can sustain success with him in charge. We're at a point now where we have a very good squad and should be looking to match Liverpool and City. We need to look beyond Ole and for a new manager if we want to capitalise on the talent we now have in my opinion.

I would say that xG helps to better quantify just how far behind City and Liverpool we are when it comes to creating quality chances and how superior their coaching is in that regard.
completely agree with this. I actually think Ole could be a good DOF at United but sadly if he gets hooked I’m not sure he would take such a position.
The play last night was pedestrian and I get frustrated watching slow, thoughtful buildup that goes nowhere. We are still stuck passing backwards and to the side. Hard to watch so many sides bring the intensity and speed and yet United lack in that respect. I expect xG would improve significantly if we played more one touch football and made more use of our speed on the pitch. Feel like this starts with the manager
 

cyberman

Full Member
Joined
May 26, 2010
Messages
37,331
xG accounts the quality of chances. You outperform xG if you are lucky, or your strikers are clinical. We have arguably the most clinical player in the world, and Bruno/Greenwood are quite clinical too, so it is likely that we are gonna outperform xG. At the same time, the lack of high-quality chances is gonna catch us, and as usual we will finish 10+ points below the champions. Very hard to be lucky for an entire season.

There was the same argument when Ole came as manager. xG is for nerds, table doesn’t lie, we have great strikers etc, while everyone who knows something about this was urging that it is gonna change soon. And then we want to crap soon after.
We have been scoring since Bruno joined us and haven’t slowed down since? So the XG naysayers have been right so far?
I still think people out too much stock into it. Barcelona drew 1-1 on Monday. Scored a header that shouldn’t be missed yet missed a header from a corner when a player was free, 4 yards out yet XG was 1. Something.
The XG crowd that claim it’s the be all and end all are just as bad as the crowd that tells us it holds no value
 

Zlatan 7

We've got bush!
Joined
May 26, 2016
Messages
11,797
Chances that are easy to score. For example Noble’s opportunity at the end of the match yesterday, or one on one striker vs keeper cases, or the million tap ins City score every season.
Does it take into account the player one on one with keeper or his style of play? Or how the tap in came about? Curious
 

Borys

Statistics Wizard
Joined
May 10, 2013
Messages
9,094
Location
Bielsko Biala, Poland
What is a high quality chance
In terms of xG algorithm, any chance where shot is taken centrally from the goal and close to it (and not a header). So like Revan mentioned, City tend to play football which result in high number of high quality chances. It doesn't require a striker to convert those. They often end up north of xG=2, which is pretty high.
The bigger the distance and angle, the lower the xG.
Does it take into account the player one on one with keeper or his style of play? Or how the tap in came about? Curious
No, it's average (hence the expected). So if you have a good strikers they can turn low quality chances into goals, like we do. But you'll struggle when they are out. That's why people mention we rely a lot on individual brilliance.
Statistically a shot of xG=0,3 will be converted in to goal once every 3 shots by average player, that's the way the xG works. Ronaldo, Bruno, and Greenwood tend to be on the positive side of goals to xG ratio but that won't happen every game.
 
Last edited:

Pogue Mahone

The caf's Camus.
Joined
Feb 22, 2006
Messages
133,934
Location
"like a man in silk pyjamas shooting pigeons
We really need to bring in an xG specialist
We did. Bruno. And he made a massive difference. Our problem is that our ability to create goals (our xG, if you will) is totally reliant on someone who can be a bit flaky and can’t possibly play every game.

This “patterns of play” stuff gives me a ball ache but there has to be some truth in it. And it’s most apparent whenever we try and play anything other than our strongest possible line-up. City and Liverpool can slot in different players but the system remains exactly the same. We look like a team of strangers trying to improvise on the hoof. That’s not going to be good enough to win the league against teams managed by Klopp and Pep, sadly.
 

Revan

Assumptionman
Joined
Dec 19, 2011
Messages
49,635
Location
London
We have been scoring since Bruno joined us and haven’t slowed down since? So the XG naysayers have been right so far?
I still think people out too much stock into it. Barcelona drew 1-1 on Monday. Scored a header that shouldn’t be missed yet missed a header from a corner when a player was free, 4 yards out yet XG was 1. Something.
The XG crowd that claim it’s the be all and end all are just as bad as the crowd that tells us it holds no value
It is not the be all and end all, but it is the best stat we have. It shows a much better picture than ball possession, or number of shots. It also confirms what we can see with eyes, that we are not very good at creating chances in a systematic manner.
Does it take into account the player one on one with keeper or his style of play? Or how the tap in came about? Curious
For most part no. Which is why teams with great attackers outperform their xG and teams with great keepers outperform their xGA. If Ronaldo and De Gea continue their form (both doable), we might outperform both xG and xGA. For example, in Mourinho’s second season, Caf was claiming that our defense was great, which it wasn’t. We saw teams repeatedly being in good positions to score, but De Gea saving us repeatedly. xGA confirmed that, De Gea outperformed more than any other keeper their xGA.

So in simple terms, xG measures how good a team is at creating chances, while xGA measures how good a team is at denying the other team to create good chances.
 

Revan

Assumptionman
Joined
Dec 19, 2011
Messages
49,635
Location
London
Hardly comparable examples.
Pretty much every team in average creates less chances, scores less goals and gets fewer points when playing a top opposition team compared to playing a bad team. Regardless of how the bad team defends more.
 

largelyworried

Full Member
Joined
Feb 10, 2021
Messages
2,101
You could also argue that it's easier to create big chances against teams that dare to play a more open game. The exception being if the opposing team is vastly superior(which isn't the case here).
I mean you could argue its easier to play at Anfield with 10 men if you want, but I don't think you'd get very far.