xG (maybe things aren't as bad as they look?) (4 Viewers)

skybluecam · Sep 22, 2024

Opp	Result	xG	xGA	G	GA	Cov +/-	Opp +/-
Stoke	L 0-1	0.7	0.9	0	1	-0.7	0.1
Oxford	W 3-2	2.9	0.9	3	2	0.1	1.1
Bristol	D 1-1	0.6	0.7	1	1	0.4	0.3
Norwich	L 0-1	1.1	0.4	0	1	-1.1	0.6
Watford	D 1-1	2.2	1.8	1	1	-1.2	-0.8
Swansea	L 0-1	1.6	0.7	1	2	-0.6	1.3
	Totals	9.1	5.4	6	8	-3.1	2.6

Above is a table of our results with xG for/against stats.

In 4 of our games we've generated more xG than our opponent. In 2 we have generated less.
In 4 of our games we've under scored our xG. 2 over.

We've generated 9.1 xG and scored 6. We've allowed 5.4 xGA and conceded 8.

Since Oxford we've scored 2 goals off 5.5 xG.

Without doing a proper xPts simulation, I would estimate our expected points at around 11 (W: Swansea, Norwich, Oxford. D: Watford, Bristol. L: Stoke)

Expected Tables:

Source	Pos	xG	xGA	xGD	xPts
FotMob	6th	9.4	5.5	3.9	11
TheFishy	7th	7.7	5.3	2.3	9.5
footystats	3rd	8.9	6.8	2.1	-
FBRef/Opta	6th	9.2	5.4	3.8	-

Gathered a few expected tables. Bear in mind only FotMob seems to be doing a proper xPts calculation.

It's also worth noting that our opponents have converted 5.4 xG into 7.5 PSxG - suggesting we're the victim of some bad luck in terms of opposition players striking better than average shots.

Obviously we've still got issues. But maybe it's not as bad as it looks. Thoughts?

Evo1883 · Sep 22, 2024

skybluecam said:
Opp Result xG xGA G GA Cov +/- Opp +/-
Stoke L 0-1 0.7 0.9 0 1 -0.7 0.1
Oxford W 3-2 2.9 0.9 3 2 0.1 1.1
Bristol D 1-1 0.6 0.7 1 1 0.4 0.3
Norwich L 0-1 1.1 0.4 0 1 -1.1 0.6
Watford D 1-1 2.2 1.8 1 1 -1.2 -0.8
Swansea L 0-1 1.6 0.7 1 2 -0.6 1.3
Totals 9.1 5.4 6 8 -3.1 2.6

Above is a table of our results with xG for/against stats.

In 4 of our games we've generated more xG than our opponent. In 2 we have generated less.
In 4 of our games we've under scored our xG. 2 over.

We've generated 9.1 xG and scored 6. We've allowed 5.4 xGA and conceded 8.

Since Oxford we've scored 2 goals off 5.5 xG.

Without doing a proper xPts simulation, I would estimate our expected points at around 11 (W: Swansea, Norwich, Oxford. D: Watford, Bristol. L: Stoke)

Expected Tables:

Source Pos xG xGA xGD xPts
FotMob 6th 9.4 5.5 3.9 11
TheFishy 7th 7.7 5.3 2.3 9.5
footystats 3rd 8.9 6.8 2.1 -
FBRef/Opta 6th 9.2 5.4 3.8 -

Gathered a few expected tables. Bear in mind only FotMob seems to be doing a proper xPts calculation.

It's also worth noting that our opponents have converted 5.4 xG into 7.5 PSxG - suggesting we're the victim of some bad luck in terms of opposition players striking better than average shots.

Obviously we've still got issues. But maybe it's not as bad as it looks. Thoughts?

I agree with you but a few won't, but last season early doors showed the xg tables that city were underperforming early on , we turned It around obviously but fell away at the end , which at the time most on here agreed was down to the small squad playing too many games.

Ironically I was one of the only ones angry at the way we fell away I have no doubt we will turn it around but Leeds is a big ask to start the turnaround , I think it's more about the performance at Leeds

skybluecam · Sep 22, 2024

Evo1883 said:
I agree with you but a few won't, but last season early doors showed the xg tables that city were underperforming early on , we turned It around obviously but fell away at the end , which at the time most on here agreed was down to the small squad playing too many games.

Ironically I was one of the only ones angry at the way we fell away I have no doubt we will turn it around but Leeds is a big ask to start the turnaround , I think it's more about the performance at Leeds

I’m expecting a joker lineup from Robins at Leeds tbh, like last year. So might not tell us anything. Although being hard to beat defensively would be a good start.

LilleSkyBlue · Sep 22, 2024

Evo1883 said:
I agree with you but a few won't, but last season early doors showed the xg tables that city were underperforming early on , we turned It around obviously but fell away at the end , which at the time most on here agreed was down to the small squad playing too many games.

Ironically I was one of the only ones angry at the way we fell away I have no doubt we will turn it around but Leeds is a big ask to start the turnaround , I think it's more about the performance at Leeds

I was also unhappy at the way we fell away. One point in the last six matches - and that was the dreadful 0-0 game away at Blackburn. I thought then that results like that are apt to mess with players' heads, or the collective spirit of the club, or something intangible like that.

I don't like the condemning of large swathes of games on the basis that the cup run knackered them, or the squad is small, or whatever else. It was still a poor run and the ripples must come to rest somewhere.

GaryJones · Sep 22, 2024

Xg pxsg etc etc blah blah blah - It’s as bad as we can see with our own eyes - it’s as bad as the league table says it is.
You can analyse the shit out of it but it’s as bad as we all know it it is!!

robbiekeane · Sep 23, 2024

GaryJones said:
Xg pxsg etc etc blah blah blah - It’s as bad as we can see with our own eyes - it’s as bad as the league table says it is.
You can analyse the shit out of it but it’s as bad as we all know it it is!!

a beautifully articulated and well reasoned response

Jamesimus · Sep 23, 2024

GaryJones said:
Xg pxsg etc etc blah blah blah - It’s as bad as we can see with our own eyes - it’s as bad as the league table says it is.
You can analyse the shit out of it but it’s as bad as we all know it it is!!

Okay then Gary, explain your counter argument using a lovely detailed explanation with stats like the OP did and I then will listen…

Jamesimus · Sep 23, 2024

For what it’s worth, we seem to have a few players who lack composure in front of goal; not necessarily even the bad performers, but players like Rudoni, BTA, Eccles have squandered chances and they all have a history of doing so, for us and their previous clubs.

fernandopartridge · Sep 23, 2024

GaryJones said:
Xg pxsg etc etc blah blah blah - It’s as bad as we can see with our own eyes - it’s as bad as the league table says it is.
You can analyse the shit out of it but it’s as bad as we all know it it is!!

Thanks for that input

Brighton Sky Blue · Sep 23, 2024

Did we really generate 1.6 xG against Swansea? Don’t remember too many clear chances except that where we hit the bar.

shmmeee · Sep 23, 2024

robbiekeane said:
a beautifully articulated and well reasoned response

Not all that counts can be counted and not all that can be counted counts.

shmmeee · Sep 23, 2024

Brighton Sky Blue said:
Did we really generate 1.6 xG against Swansea? Don’t remember too many clear chances except that where we hit the bar.

We had more shots on goal than anyone else in the division this weekend believe it or not.

Brighton Sky Blue · Sep 23, 2024

shmmeee said:
We had more shots on goal than anyone else in the division this weekend believe it or not.

Fair, it didn’t look like we created that many decent chances to be honest

shmmeee · Sep 23, 2024

Brighton Sky Blue said:
Fair, it didn’t look like we created that many decent chances to be honest

No I agree. I was shocked to see that stat on Sunday

Frostie · Sep 23, 2024

shmmeee said:
No I agree. I was shocked to see that stat on Sunday

Lots of low xG shots + 1 very high one which was missed (Bassette).

Fergusons_Beard · Sep 23, 2024

skybluecam said:
I’m expecting a joker lineup from Robins at Leeds tbh, like last year. So might not tell us anything. Although being hard to beat defensively would be a good start.

That Joker line up(I couldn’t believe it either) got us an unlikely point last year.

Maybe Robins does know what he’s doing after all!

Sent from my iPhone using Tapatalk Pro

fernandopartridge · Sep 23, 2024

skybluecam said:
Opp Result xG xGA G GA Cov +/- Opp +/-
Stoke L 0-1 0.7 0.9 0 1 -0.7 0.1
Oxford W 3-2 2.9 0.9 3 2 0.1 1.1
Bristol D 1-1 0.6 0.7 1 1 0.4 0.3
Norwich L 0-1 1.1 0.4 0 1 -1.1 0.6
Watford D 1-1 2.2 1.8 1 1 -1.2 -0.8
Swansea L 0-1 1.6 0.7 1 2 -0.6 1.3
Totals 9.1 5.4 6 8 -3.1 2.6

Above is a table of our results with xG for/against stats.

In 4 of our games we've generated more xG than our opponent. In 2 we have generated less.
In 4 of our games we've under scored our xG. 2 over.

We've generated 9.1 xG and scored 6. We've allowed 5.4 xGA and conceded 8.

Since Oxford we've scored 2 goals off 5.5 xG.

Without doing a proper xPts simulation, I would estimate our expected points at around 11 (W: Swansea, Norwich, Oxford. D: Watford, Bristol. L: Stoke)

Expected Tables:

Source Pos xG xGA xGD xPts
FotMob 6th 9.4 5.5 3.9 11
TheFishy 7th 7.7 5.3 2.3 9.5
footystats 3rd 8.9 6.8 2.1 -
FBRef/Opta 6th 9.2 5.4 3.8 -

Gathered a few expected tables. Bear in mind only FotMob seems to be doing a proper xPts calculation.

It's also worth noting that our opponents have converted 5.4 xG into 7.5 PSxG - suggesting we're the victim of some bad luck in terms of opposition players striking better than average shots.

Obviously we've still got issues. But maybe it's not as bad as it looks. Thoughts?

Using the Fbref data: I am assuming out xGD/90 is 3.8/6 = 0.633 - extrapolate over the season is an xGD of 29 - the only teams that got higher than that last season were Leicester, Leeds and Southampton.

skybluecam · Sep 23, 2024

Brighton Sky Blue said:
Fair, it didn’t look like we created that many decent chances to be honest

Bassette’s header on to the bar was 0.49.

Rudoni had a chance on the rebound that was 0.24.

We then had 4 other shots that were all above 0.1.

1.6 xg was what Ipswich averaged last season and only Leicester, Leeds and Southampton averaged more.

napolimp · Sep 23, 2024

This was the XG table last season vs the final standings. Goes to show there's not that much correlation between the 2, outside the very top of the table.

skybluecam · Sep 23, 2024

napolimp said:
This was the XG table last season vs the final standings. Goes to show there's not that much correlation between the 2, outside the very top of the table.

View attachment 38594

That’s only taking in to account your own xG, not xGA so it’s only half the puzzle.

And at the end of the day a season is just a sample.

Calista · Sep 23, 2024

It's just a way of quantifying what seems obvious to me. Our first 6 games have all been pretty tight, as you'd expect in this league. And again as you'd expect, if you don't take your chances (through bad luck and/or poor finishing) and the other team does, you are going to be lower half of the table. Our general performances are nowhere near as bad as people are making out (hence the xG stats, which don't surprise me at all), but in both penalty areas we are suffering.

napolimp · Sep 23, 2024

skybluecam said:
That’s only taking in to account your own xG, not xGA so it’s only half the puzzle.

And at the end of the day a season is just a sample.

Ah right, got it, that makes sense. So I guess your likes of West Brom and Norwich had low XG against.

stupot07 · Sep 23, 2024

napolimp said:
This was the XG table last season vs the final standings. Goes to show there's not that much correlation between the 2, outside the very top of the table.

View attachment 38594

Yep, football is won or lost in moments, xG is largely bollocks, as it doesn't take into account the ebb and flow of the game, you could argue that Norwich came to keep it tight and try and grab one on the break, and once their got their goal, they had a game plan of sitting back, defend their lead and soak up the pressure. xG (0.4) says they were rubbish and hardly threatened.

Mucca Mad Boys · Sep 23, 2024

stupot07 said:
Yep, football is won or lost in moments, xG is largely bollocks, as it doesn't take into account the ebb and flow of the game, you could argue that Norwich came to keep it tight and try and grab one on the break, and once their got their goal, they had a game plan of sitting back, defend their lead and soak up the pressure. xG (0.4) says they were rubbish and hardly threatened.

Funnily enough, they do measure xG on game state (i.e. when winning or losing) too now.

Looking at the Man City v Arsenal match, Man City’s xG was only 1.5-1.6 in the 2nd half v Arsenal. That’s accumulated because they had 29 shots in that half.

Back to that Norwich game, we didn’t fashion a clear cut chance but that 0.4 isn’t perhaps as bad you think since a penalty is about 0.6 xG. For context, our game v Bristol City was 0.66 (Bristol) versus our 0.64 for the whole match.

We look like a better side when we’re direct and putting the ball into space for Haji, Simms, ETA and even EMC. We just make it a bit too easy for teams to get set and so far, I don’t back Eccles, Torp or even Sheaf to make those progressive passes to split set defences up.

skybluecam · Sep 23, 2024

stupot07 said:
Yep, football is won or lost in moments, xG is largely bollocks, as it doesn't take into account the ebb and flow of the game, you could argue that Norwich came to keep it tight and try and grab one on the break, and once their got their goal, they had a game plan of sitting back, defend their lead and soak up the pressure. xG (0.4) says they were rubbish and hardly threatened.

Game state is important yes.

But looking at Norwich’s chances up to and including the goal statistically you’d only expect them to score 1+ goals around 27% of the time. So in that sense we were just unlucky.

xG (maybe things aren't as bad as they look?) (4 Viewers)

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Users who are viewing this thread

Privacy & Transparency

Privacy & Transparency