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

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
I think people are misinterpreting the xG statistic, if we got a higher xG than goals scored it doesnt mean that magically our strikers will get more goals, its not luck its about ability.

Thats why Aguero always has more goals that xG expected, he is world class and thats what worldclass players do. They get goals in situations other strikers dont. At the end of the day the xG is a value given from many different players under many different chances but if you have a striker who misses a lot if wont matter if a thousand players would had scored, we will still struggle.

The same applies for our defence.
No that's not it at all. Over the past 3 full seasons, Aguero is basically even with his xG.
2016/17 - 20 goals, 22.67 xG (-2.67)
2017/18 - 21 goals, 18.57 xG (+2.43)
2018/19 - 21 goals, 19.95 xg (+1.05)

In fact I'm pretty sure that was the basis of a lot of the initial studies. The best strikers are very frequently not actually anything special in their finishing relative to the chances they get. The thing that they excel at more than any other player is the ability to get in those quality chances, aka get a high xG. Scoring lots of goals 1 year with a low xG doesn't mean that player is world class, it means it's a statistical outlier for pretty much everyone and odds are that it will revert very close to the norm the following season. High xG players tend to be the highest actual goalscorers, and the ones that can actually sustain that year after year.

Also feel like you're referring to our own players like Rashford missing chances regularly, being poor finishers, etc. Rashford since he broke through:
2019/20 - 2 goals, 3.30 xG (-1.30)
2018/19 - 10 goals, 11.61 xG (-1.61)
2017/18 - 7 goals, 5.53 xG (+1.47)
2016/17 - 5 goals, 6.27 xG (-1.27)
2015/16 - 5 goals, 3.03 xG (+1.97)

So he's less than a goal off what his xG says it should be at, over ~3.5 seasons. Standard finishing. Just like Aguero's over the past 3 years. There is always variability, but in general with a large enough sample size, players tend to average out to their xG.
 

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
I think it's more thna gaffes though. The lethargic first half against Palace was very concerning and the lack of any organisation or generally having a clue in certain situations (crosses, set pieces) is really very poor. There are positive elements in it too which ties in to what you are saying. The second half of the Chelsea game, first half at Wolves, and periods of the Southampton game, I think we've looked quite good. It's unrealistic to expect us to look good for 90 minutes of every game but we need to start using these periods in games to win them by not completing switching off in other parts or ignoring basic fundamentals.

Think I've changed my opinion a bit. at the time I said the Palace and Southampton performances were too similar to under Jose, but on reflection I don't think they were. We are just having different kinds of problems. If we can iron them out then at the very least it'll be much easier to get behind this team, even if the results are still a bit hit and miss. if some of the rest of the team can show somewhere near the work ethic of James that will also help.
The set pieces are a thing I think we'll just have to accept... They are shockingly bad and I don't see the slightest bit of progress over the last few years, if anything they've regressed. I look at the side and the only player who IMO can take a good set piece is shit at just about everything else and rarely plays (Pereira), so even if we have a decent aerial threat for once in Maguire, it's still most likely going to result in nothing much from set pieces because we have fecking Shaw taking them from one side, and Rashford from the other (who just isn't good at corners, but the rest aren't better).
 

Amarsdd

Full Member
Joined
Jun 7, 2013
Messages
3,299
Do you know what can also give you a good reflection of teams' performance over long periods? Your eyeballs; your brain. Try using them. Some of you xG guys need to close your laptops, climb the stairs from your mother's basement and go have a beer, meet some people, get laid, do something, anything.
I really hope for your sake that you are not as stupid as your post there.
 

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
I'm not sure what this sort of stuff proves. We've dropped points the last two games because we concede ridiculously cheap goals that allow teams to score against us without needing to create chances themselves or dominate parts of the game.

That isn't something that's going to fix itself just because we have good XG stats.

If you take the stupid errors and pathetic set piece play out of our games the performances have actually been fairly good, barring the first half of the Palace game, which was terrible.
It can be used as a quantitative marker to point towards to show signs of progress and reasons to be optimistic rather than just saying oh new season, same old shit after 4 games. It's not, because we are creating more chances, we are creating better chances, our defence is conceding fewer chances than any other team in the league (only 4 games) and a recipe of creating a decent amount of chances while conceding few chances tends to result in good results eventually, even if you don't have the luck at first. It's also useful to judge IMO what a manager should have a decently big influence on, while ignoring what they really don't have much of an influence on. The way I see it, everything up until the actual shot at goal, the manager through training and tactics and everything else should have an influence on. At that point, it's on the player to just finish the chances, score the pens, or your goalkeeper to not feck up. I find it very hard to criticize a manager at any point if those points are the reason why a team lost a game. You can have a game where the opposition scores 3 goals from 3 shots from 30 yards out where they create nothing else (even if your goalkeeper could do nothing about it), and odds are that you'll lose that game even if you create 30 shots throughout the game and have an xG of over 2. It happens. Same with random big mistakes that just happen to anyone. XG can give perspective to how a game went, and while it's not always good to just look at it as a game by game basis in raw values, it's helpful to give a clearer picture if the results aren't doing it justice. Or it can point out trends that a team is doing well but keeps shooting themselves in the foot and there's a clear issue that needs sorting out (usually individual mistakes).
 

SAFMUTD

New Member
Joined
Mar 14, 2018
Messages
11,787
No that's not it at all. Over the past 3 full seasons, Aguero is basically even with his xG.
2016/17 - 20 goals, 22.67 xG (-2.67)
2017/18 - 21 goals, 18.57 xG (+2.43)
2018/19 - 21 goals, 19.95 xg (+1.05)

In fact I'm pretty sure that was the basis of a lot of the initial studies. The best strikers are very frequently not actually anything special in their finishing relative to the chances they get. The thing that they excel at more than any other player is the ability to get in those quality chances, aka get a high xG. Scoring lots of goals 1 year with a low xG doesn't mean that player is world class, it means it's a statistical outlier for pretty much everyone and odds are that it will revert very close to the norm the following season. High xG players tend to be the highest actual goalscorers, and the ones that can actually sustain that year after year.

Also feel like you're referring to our own players like Rashford missing chances regularly, being poor finishers, etc. Rashford since he broke through:
2019/20 - 2 goals, 3.30 xG (-1.30)
2018/19 - 10 goals, 11.61 xG (-1.61)
2017/18 - 7 goals, 5.53 xG (+1.47)
2016/17 - 5 goals, 6.27 xG (-1.27)
2015/16 - 5 goals, 3.03 xG (+1.97)

So he's less than a goal off what his xG says it should be at, over ~3.5 seasons. Standard finishing. Just like Aguero's over the past 3 years. There is always variability, but in general with a large enough sample size, players tend to average out to their xG.
Wow great post, didn't knew that. Hopefully we’ll start turning around then. But also the xG for Rashford is really low for someone we hope will get 20 goals a season.
 

Treble

Full Member
Joined
Sep 11, 2008
Messages
10,550
It can be used as a quantitative marker to point towards to show signs of progress and reasons to be optimistic rather than just saying oh new season, same old shit after 4 games. It's not, because we are creating more chances, we are creating better chances, our defence is conceding fewer chances than any other team in the league (only 4 games) and a recipe of creating a decent amount of chances while conceding few chances tends to result in good results eventually, even if you don't have the luck at first. It's also useful to judge IMO what a manager should have a decently big influence on, while ignoring what they really don't have much of an influence on. The way I see it, everything up until the actual shot at goal, the manager through training and tactics and everything else should have an influence on. At that point, it's on the player to just finish the chances, score the pens, or your goalkeeper to not feck up. I find it very hard to criticize a manager at any point if those points are the reason why a team lost a game. You can have a game where the opposition scores 3 goals from 3 shots from 30 yards out where they create nothing else (even if your goalkeeper could do nothing about it), and odds are that you'll lose that game even if you create 30 shots throughout the game and have an xG of over 2. It happens. Same with random big mistakes that just happen to anyone. XG can give perspective to how a game went, and while it's not always good to just look at it as a game by game basis in raw values, it's helpful to give a clearer picture if the results aren't doing it justice. Or it can point out trends that a team is doing well but keeps shooting themselves in the foot and there's a clear issue that needs sorting out (usually individual mistakes).
It's dubious to draw xG based conclusions after 4 games. Teams start the season in different form. Some start strongly and fade away, others are a bit slow but then get better. Give it at least 15 or even 20 games to make the xG discussion more meaningful. I think xG is a good tool but not on the basis of 4 games.
 

Sky1981

Fending off the urge
Joined
Apr 12, 2006
Messages
30,002
Location
Under the bright neon lights of sincity
If you're expected to score 20 assuming those 20 are a good chance and you only score 3. That's not progress, that's bad coaching.

If you think 3 wins is 15 games is progress. It's not.

For 3 wins in 15 to be considered a progress the opposition must be making 5 world class saves a game, we should hit the woodwork 20 times, constantly attacking 90 minutes. Then we can say it's a progress.

If this was jose with 3 wins in 15 he'll be booed across old trafford.
 

Mindhunter

Full Member
Joined
Jul 26, 2013
Messages
3,629
No that's not it at all. Over the past 3 full seasons, Aguero is basically even with his xG.
2016/17 - 20 goals, 22.67 xG (-2.67)
2017/18 - 21 goals, 18.57 xG (+2.43)
2018/19 - 21 goals, 19.95 xg (+1.05)

In fact I'm pretty sure that was the basis of a lot of the initial studies. The best strikers are very frequently not actually anything special in their finishing relative to the chances they get. The thing that they excel at more than any other player is the ability to get in those quality chances, aka get a high xG. Scoring lots of goals 1 year with a low xG doesn't mean that player is world class, it means it's a statistical outlier for pretty much everyone and odds are that it will revert very close to the norm the following season. High xG players tend to be the highest actual goalscorers, and the ones that can actually sustain that year after year.

Also feel like you're referring to our own players like Rashford missing chances regularly, being poor finishers, etc. Rashford since he broke through:
2019/20 - 2 goals, 3.30 xG (-1.30)
2018/19 - 10 goals, 11.61 xG (-1.61)
2017/18 - 7 goals, 5.53 xG (+1.47)
2016/17 - 5 goals, 6.27 xG (-1.27)
2015/16 - 5 goals, 3.03 xG (+1.97)

So he's less than a goal off what his xG says it should be at, over ~3.5 seasons. Standard finishing. Just like Aguero's over the past 3 years. There is always variability, but in general with a large enough sample size, players tend to average out to their xG.
Great post. Thanks l, I learned a lot.
 

Pagh Wraith

Full Member
Joined
May 2, 2011
Messages
4,361
Location
Germany
Just to add to that, these are Messi's stats:

2018/19 - 36 goals, 26.00 xG (+10.00)
2017/18 - 34 goals, 28.95 xG (+5.05)
2016/17 - 37 goals, 26.89 xG (+10.11)
2015/16 - 26 goals, 27.10 xG (-1.10)
2014/15 - 43 goals, 35.89 xG (+7.11)

Understat's xG model is not the most sophisticated model out there. For example, it doesn't take defensive pressure enough into account (i.e. number of defenders between shooter and goal line and their proximity) whereas other models do (which explains why Atletico outperform their xG conceded beyond normal margins). But it gives us something to work with. Messi has been outperforming his xG by an average of 13% over the last five seasons. If we agree on him being the best finisher in football, it makes sense to set this number as the upper limit of what is possible. This is in line with what I found a while back when I analysed the numbers across the major leagues. Even the top teams rarely outperform their xG by more than 10% longterm showing how useful the stat is. Anything beyond that is a statistical outlier, such as Dortmund's freak season in 2014/15 where they should have been comfortably in the top 4 all season but were fighting relegation for the most part.
 

JustAGuest

Full Member
Joined
Jun 19, 2013
Messages
742
Do you know what can also give you a good reflection of teams' performance over long periods? Your eyeballs; your brain. Try using them. Some of you xG guys need to close your laptops, climb the stairs from your mother's basement and go have a beer, meet some people, get laid, do something, anything.
Imagine getting this upset over a statistic. :lol:
 

Grande

Full Member
Joined
Jun 22, 2007
Messages
6,205
Location
The Land of Do-What-You-Will
I think it's more thna gaffes though. The lethargic first half against Palace was very concerning and the lack of any organisation or generally having a clue in certain situations (crosses, set pieces) is really very poor. There are positive elements in it too which ties in to what you are saying. The second half of the Chelsea game, first half at Wolves, and periods of the Southampton game, I think we've looked quite good. It's unrealistic to expect us to look good for 90 minutes of every game but we need to start using these periods in games to win them by not completing switching off in other parts or ignoring basic fundamentals.

Think I've changed my opinion a bit. at the time I said the Palace and Southampton performances were too similar to under Jose, but on reflection I don't think they were. We are just having different kinds of problems. If we can iron them out then at the very least it'll be much easier to get behind this team, even if the results are still a bit hit and miss. if some of the rest of the team can show somewhere near the work ethic of James that will also help.
James is showing good spirit and bravery, and should be a lead to follow for others. Regarding the set piece/crosses thing (which we were bad at under Jose), I have noticed we are very up and down. It seems most times we are well organized, as evidenced by the fact that even teams that target these situations (Wolves, Palace, Sot’on) have not made many dangerous situations from them.

It’s more like we seem to panic in parts of the game and go into complete dissarray. This happened a part of the game vs Chelsea (first half), and in two five minute periods in the second halves vs Wolves and So’ton. It actually seems we are pretty well organized, but the players aren’t that secure about the organization yet, so if shit happens fast, they blank out easily. Hopefully this comes with practice. It’s worth noting that the only candidate for defensive leadership is the only one who missed out on preseason and is new to the team. When Maguire knows by heart how Solskjær wants us to defend in different situations, I expect us to improve in that regard.
 

Tom Cato

Godt nyttår!
Joined
Jan 3, 2019
Messages
7,565
Do you know what can also give you a good reflection of teams' performance over long periods? Your eyeballs; your brain. Try using them. Some of you xG guys need to close your laptops, climb the stairs from your mother's basement and go have a beer, meet some people, get laid, do something, anything.
What sort of troglodyte of a brexit neckbeard are you?

You know what doesn't give you a good reflection of a teams performance over time? Your Brain. Because we all have confirmation biases. You see what you want to believe. And then you go on the internet and somehow tell people to both use their brains, but don't use math, because anyone who's into numbers surely must be single, live with their parents and be too afraid to partake in social gatherings.

You know, if you DID want to use your brain, you could stop clowning around and commit to earning a Bachelors, Masters, and eventually a PhD in statistical analysis. Get yourself a PhD in a field, then you can really tell your peers to use their brains when they talk about something you actually know something about, instead of coming into this thread to belittle people who discuss numbers you don't have the faintest idea how are even relevant, or what they mean for that matter.

Now excuse me, I'll go lay with your mother and drink your beer.
 

Leftback99

Might have a bedwetting fetish.
Joined
Jan 11, 2015
Messages
13,966
If you're expected to score 20 assuming those 20 are a good chance and you only score 3. That's not progress, that's bad coaching.

If you think 3 wins is 15 games is progress. It's not.

For 3 wins in 15 to be considered a progress the opposition must be making 5 world class saves a game, we should hit the woodwork 20 times, constantly attacking 90 minutes. Then we can say it's a progress.

If this was jose with 3 wins in 15 he'll be booed across old trafford.
You don't get to Premier league level and rely on coaching be able to finish chances.

Has anyone said 3 wins in 15 is progress?
 

midnightmare

Full Member
Joined
Sep 22, 2015
Messages
1,228
Location
Midian
Cold statistical analysis does help - to an extent, but they need to be viewed properly; in statistical analysis, this literally means "when there is a representative body of data to study". It's way too soon to analyse xG for James for example, as it's been just 3 and a bit games. It's not, however, too soon to assess Rashford's PL performances or Martial's, since their debuts. That's how statistics work! You can't take isolated data points or small sets and try to extrapolate. Once you have a body of data over time though, statistics make a lot more sense - and then they start to tell a tale.

Under Jose, there were folks that highlighted that our good run saw insane efficiency (as indeed under LvG) and that we were ahead of where xG would have us. People predicted a reversion to the mean and a crash in results, unless we started creating more chances - and it came to pass (and how!). The same transpired during Ole's good run - and the resulting dip was similar. We're now playing better (to the eye) and are under-performing compared to xG. Logic dictates that it will even out - and I trust it will begin soon. I don't place much stock in xG to analyse a single game, but such statistics when viewed over a (substantial) period of time almost invariably give valuable insights into what we should expect moving forward.
 

Withnail

Full Member
Joined
Jan 5, 2019
Messages
29,330
Location
The Arena of the Unwell
See, that's where I disagree. Our finishing has been poor.

vs. Wolves - we win possession 66/34, we win shots 9/6, we tie shots on target 2/2 - game ends in draw

vs. Palace - we win possession 71/29, we win shots 22/5, we tie shots on target 3/3 - we lose 2-1

vs. Southhampton - we win possession 59/41, we win shots 21/10, we win shots on target 8/2 - game ends in draw

In three matches, we've done no worse than 59% possession, taken 52 shots with 21 shots on target and have 4 goals to show for it.

That's a lack of finishing. For all the criticism of what's going on on the field, 4 of 21 is the problem.

Lots of those shots were speculative pops from long range, often when a better pass was on.

United aren't creating enough clear cut chances. Two of James goals were in the goal of the month and there's no guarantee he's gonna keep that up.

Although, I do agree with the wider point that the team have been playing better than the poor run at the end of last season. I think there is reason to be optimistic they'll continue to improve and turn performances into results.
 

Leftback99

Might have a bedwetting fetish.
Joined
Jan 11, 2015
Messages
13,966
If we got beat against top teams, then yes probably it's a slip. But when half the league have our numbers and knew exactly how to hurt us i dont think that as a progress at all.

What is it we do better this season?
So far create more chances than the opposition and concede very few chances, statistically up there with the best in the league. IF we continue doing this results will follow.
 

Tom Cato

Godt nyttår!
Joined
Jan 3, 2019
Messages
7,565
If we got beat against top teams, then yes probably it's a slip. But when half the league have our numbers and knew exactly how to hurt us i dont think that as a progress at all.

What is it we do better this season?
Pretty much everything except for scoring on our chances. So thats good.
 

Tom Cato

Godt nyttår!
Joined
Jan 3, 2019
Messages
7,565
Including defence? Off the ball movement? Pacing of the game? Subs? Conceding silly setpieces?

Success?

Dear me
We concede fewer goals than last season.
Defence is better than last season.
Pacing of the game is a lot better. We have a greater deal of posession.

We have a lack of results, not lack of form. There are a lot of very tiny "what ifs" between us and 12 points. I know it's not the best results, but just loook at the underlying numbers and there's more than ground for optimism.

Plus the mangers 3 purchases have all been among the best players to start the season.
 

Treble

Full Member
Joined
Sep 11, 2008
Messages
10,550
We concede fewer goals than last season.
Defence is better than last season.
Pacing of the game is a lot better. We have a greater deal of posession.

We have a lack of results, not lack of form. There are a lot of very tiny "what ifs" between us and 12 points. I know it's not the best results, but just loook at the underlying numbers and there's more than ground for optimism.

Plus the mangers 3 purchases have all been among the best players to start the season.
You are comparing 4 and 38 games. Which is meaningless. The first 4 games this season are better than the first or the last 4 games last season. But there were series of 4 games last season which were better than the current one: both in terms of actual outcomes and xG. So, to say "we do x better than last season" is a premature generalisation. The sample is too small for such inferences.
 

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
It's dubious to draw xG based conclusions after 4 games. Teams start the season in different form. Some start strongly and fade away, others are a bit slow but then get better. Give it at least 15 or even 20 games to make the xG discussion more meaningful. I think xG is a good tool but not on the basis of 4 games.
Oh I agree fully. But the basis of using it over this small period is pretty much "assuming we continue playing like this, the results will pick up". And even using it over the entirety of Ole's time here has us as 3rd in the league (but I definitely think last season isn't useful for a lot of reasons, both for United and others).
 

Treble

Full Member
Joined
Sep 11, 2008
Messages
10,550
Oh I agree fully. But the basis of using it over this small period is pretty much "assuming we continue playing like this, the results will pick up". And even using it over the entirety of Ole's time here has us as 3rd in the league (but I definitely think last season isn't useful for a lot of reasons, both for United and others).
Agreed.
 

adexkola

Doesn't understand sportswashing.
Joined
Mar 17, 2008
Messages
48,028
Location
The CL is a glorified FA Cup set to music
Supports
orderly disembarking on planes
Oh I agree fully. But the basis of using it over this small period is pretty much "assuming we continue playing like this, the results will pick up". And even using it over the entirety of Ole's time here has us as 3rd in the league (but I definitely think last season isn't useful for a lot of reasons, both for United and others).
Yes, you're using the tool in a valid way.
 

Greck

Full Member
Joined
Dec 1, 2016
Messages
7,099
Oh I agree fully. But the basis of using it over this small period is pretty much "assuming we continue playing like this, the results will pick up". And even using it over the entirety of Ole's time here has us as 3rd in the league (but I definitely think last season isn't useful for a lot of reasons, both for United and others).
Agree. Follow advanced analytics in basketball and other sports. Its use in football is still a bit crude but Xg seems like a step forward. One thing though, it's commonly agreed across those sports that they shouldn't be used in small samples because single games still have a huge impact on the averages.
 

Physiocrat

Has No Mates
Joined
Jun 29, 2010
Messages
8,913
Just to add to that, these are Messi's stats:

2018/19 - 36 goals, 26.00 xG (+10.00)
2017/18 - 34 goals, 28.95 xG (+5.05)
2016/17 - 37 goals, 26.89 xG (+10.11)
2015/16 - 26 goals, 27.10 xG (-1.10)
2014/15 - 43 goals, 35.89 xG (+7.11)
:eek:
 

Eckers99

Michael Corleone says hello
Joined
Aug 9, 2014
Messages
6,117
We concede fewer goals than last season.
Defence is better than last season.
Pacing of the game is a lot better. We have a greater deal of posession.

We have a lack of results, not lack of form. There are a lot of very tiny "what ifs" between us and 12 points. I know it's not the best results, but just loook at the underlying numbers and there's more than ground for optimism.

Plus the mangers 3 purchases have all been among the best players to start the season.
We're not clinical enough, lack an incisive central midfielder, buckle under pressure and lack a mobile DM to take the strain off the overly exposed back 4. You don't need stats to confirm any of those things, just a pair of eyes, access to the game and a moderate understanding of what you're watching.
 

chromepaxos

Full Member
Joined
Jul 29, 2015
Messages
192
We're not clinical enough, lack an incisive central midfielder, buckle under pressure and lack a mobile DM to take the strain off the overly exposed back 4. You don't need stats to confirm any of those things, just a pair of eyes, access to the game and a moderate understanding of what you're watching.
I swear to God I will never understand people who think it is better not to use tools when they are available.

Hey Galileo! Put that telescope down. All you need in order to know that there's a moon is your own eyes and a moderate understanding that the Sun goes round the Earth!

It's as if these folk want to be ignorant and lacking deeper insight.
 

Classical Mechanic

Full Member
Joined
Aug 25, 2014
Messages
35,216
Location
xG Zombie Nation
1.10 - 0.46

Again another tight display at the back but weak going forward.

Interestingly Spurs vs Palace was

1.05 - 0.58 so Spurs still not really dominating general play.
 

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
1.10 - 0.46

Again another tight display at the back but weak going forward.

Interestingly Spurs vs Palace was

1.05 - 0.58 so Spurs still not really dominating general play.
No surprise given we lacked our 2 most creative and talented attacking players, and Leicester are actually a very good defensive side this season.
 

Wednesday at Stoke

Full Member
Joined
Feb 11, 2014
Messages
21,629
Location
Copenhagen
Supports
Time Travel
Tight game isn’t really surprising when pogba who is the main attacking outlet isn’t playing. Add Martial to the absentees and there’s not a lot of creativity in attack.
 

bosnian_red

Worst scout to ever exist
Joined
Aug 13, 2011
Messages
57,578
Location
Canada
Best defensive side in the league after 5 games in xG and real goals (and a few other stats I think). What a difference Maguire and Wan Bissaka have made.