Expected Goals Table

[hogesar 10,766]

I know some on here aren't a fan of statistics but the expected goals stat is one of the more interesting ones. As always context is important but for those who are saying we've not been able to break down teams, this helps reinforce my belief that our poor finishing has been the problem. Playing wise I think we've been good.

 

[hogesar 10,766]

Additionally, the match yesterday was a deserved win, contrary to what some on here believe - if we go by things like chances created, possession, shots, passes, goals..pretty much everything...

 

[Crabbycanary3 994]

Does that mean that Reading are Watford are lucky so far and their luck may well run out? 🙂 

[hogesar 10,766]

3 minutes ago, Crabbycanary3 said:

Does that mean that Reading are Watford are lucky so far and their luck may well run out? 🙂 

I mean, at the least it shows they've been very very clinical without actually creating many good chances. With Watford you could put that down to the quality of player meaning they're taking half chances more often than other teams, I guess.

[Nuff Said 5,965]

Didn’t we outperform our xG in the last promotion season?

[king canary 8,758]

17 minutes ago, Crabbycanary3 said:

Does that mean that Reading are Watford are lucky so far and their luck may well run out? 🙂 

Either that or they have very clinical finishers.

[TheGunnShow 7,377]

1 hour ago, Crabbycanary3 said:

Does that mean that Reading are Watford are lucky so far and their luck may well run out? 🙂 

That is possible, especially considering that we've just started the season. It does hint that unless they improve, their results will probably fall back towards mid-table. 

Wycombe had more shots than Reading but lost 1-0 and even though we never really had them under control, they still had more shots against Reading than they did against us.

[Wacky Waving Inflatable Arm Flailing Tube Man 4,620]

Good table. I know a lot of people are anti-xG but it often raises a lot of different points.

I'm surprised to see us that high for xG personally, as we don't seem to have created a huge amount of chances, certainly before the international break. 

Interesting to see that Reading have the lowest xG in the league. Obviously it shows they've been very clinical, but it's often a sign that they've been a bit fortunate and their luck will run out sooner rather than later. But then again, they've got the lowest xG against, so that balances it out I suppose...

 

Edited October 25, 2020 by Wacky Waving Inflatable Arm Flailing Tube Man

[sonyc 6,145]

Reading look very good. Only 1 conceded I believe in 7 games.

[Block Y Seat 176 80]

1 hour ago, hogesar said:

I know some on here aren't a fan of statistics but the expected goals stat is one of the more interesting ones. As always context is important but for those who are saying we've not been able to break down teams, this helps reinforce my belief that our poor finishing has been the problem. Playing wise I think we've been good.

 

Which means 0-0 Tuesday night!

[Wacky Waving Inflatable Arm Flailing Tube Man 4,620]

1 minute ago, sonyc said:

Reading look very good. Only 1 conceded I believe in 7 games.

With a goals against xG of 5.6, it just shows that their opponents have been very wasteful.

[Canaries north 155]

I have watched a couple of Watford games and they do look clinical up front and quite tight at the back. They also have Deeney, Gray, Hughes and Capoe to come back. Reading are tight at the back but I believe their goals for will dry up. Also their squad will suffer with 2 games per week while we will be ok as long as defenders don't get injured or suspended. This season will be about depth of squad so the three relegated teams should do well in my opinion. 

[Canary Wundaboy 1,423]

We need to be quicker to pull the trigger. We give the ball up far too easily by trying to play that cute little extra pass which other teams are totally wise to and just crowd the box to intercept. One of the reasons I'm not thrilled that Idah is out, we always look likely to either score through one of his efforts or on the rebound from it.

[hogesar 10,766]

1 hour ago, Wacky Waving Inflatable Arm Flailing Tube Man said:

Good table. I know a lot of people are anti-xG but it often raises a lot of different points.

I'm surprised to see us that high for xG personally, as we don't seem to have created a huge amount of chances, certainly before the international break. 

Interesting to see that Reading have the lowest xG in the league. Obviously it shows they've been very clinical, but it's often a sign that they've been a bit fortunate and their luck will run out sooner rather than later. But then again, they've got the lowest xG against, so that balances it out I suppose...

 

There was an interesting xG table during one of Ipswich's recent seasons in the championship which they started really strongly despite having one of the lowest xG - obviously at the time there was a lot of "Oh, they've just got good finishers" but it's never that simple over an entire season and surely enough they ended up lower midtable or worse (I can't remember which season, but it included Waghorn starting well with Garner I believe).

[hogesar 10,766]

Equally seeing Blackburn top disproves some of the comments on social media that they're not viable promotion candidates.

The caveat of course is weve really not played enough matches so one high scoring or chance creating fixture can skew the results.

[Yellow and Green 352]

16 hours ago, sonyc said:

Reading look very good. Only 1 conceded I believe in 7 games.

They've also had quite easy games, apart from Watford.

You never know, of course, but they've had Cardiff, Barnsley, Derby, Rotherham, Wycome, Borough and Watford.

[Peanuts 154]

Thanks for posting this, really interesting to see.

On 25/10/2020 at 09:43, Nuff Said said:

Didn’t we outperform our xG in the last promotion season?

I believe so and it looks like the opposite is happening so far this season.

 

[Midlands Yellow 4,682]

I’m going with official league table and it’s looking better. 

[lake district canary 4,830]

You'd be better off just guessing at how many goals a team might score in any match. Talking about "expected goals" just puts an extra daft level of analysis that can add pressure on to a team that they could do without, thinking they aren't up to the level they should be - or worse, that they may be over achieving, which in itself can affect them adversely.  Just play the game, create chances and convert them. 

[NFN FC 1,142]

1 hour ago, lake district canary said:

You'd be better off just guessing at how many goals a team might score in any match. Talking about "expected goals" just puts an extra daft level of analysis that can add pressure on to a team that they could do without, thinking they aren't up to the level they should be - or worse, that they may be over achieving, which in itself can affect them adversely.  Just play the game, create chances and convert them. 

⬆️ xP (expected post) 😉

[lake district canary 4,830]

6 minutes ago, NFN FC said:

⬆️ xP (expected post) 😉

Yeah, I know, too predictable 🤦‍♂️

 

[hogesar 10,766]

2 hours ago, lake district canary said:

You'd be better off just guessing at how many goals a team might score in any match. Talking about "expected goals" just puts an extra daft level of analysis that can add pressure on to a team that they could do without, thinking they aren't up to the level they should be - or worse, that they may be over achieving, which in itself can affect them adversely.  Just play the game, create chances and convert them. 

Its only a statistic on its own and as always requires context but its one of the more 'interesting' stats that clearly Farke sees as important.

[sonyc 6,145]

https://www.infogol.net/en/leagues/english-football-league-championship-table-2020-21/264

Thought this alternative table might be of interest @hogesar

Shows expected and forecast positions at the end of the season too. I liked the layout. Interesting to see that they forecast a falling off of Reading and Brentford and Watford (at this early stage) to be 2nd and 1st respectively. We are forecast at 4th at this time.

[hogesar 10,766]

13 minutes ago, sonyc said:

https://www.infogol.net/en/leagues/english-football-league-championship-table-2020-21/264

Thought this alternative table might be of interest @hogesar

Shows expected and forecast positions at the end of the season too. I liked the layout. Interesting to see that they forecast a falling off of Reading and Brentford and Watford (at this early stage) to be 2nd and 1st respectively. We are forecast at 4th at this time.

Cheers, Its not displaying on my phone properly but will have a look tomorrow!

[sonyc 6,145]

So...after last night's games the forecast position remains 4th for us, whereas Watford have actually moved up a place, likewise Bournemouth with a game in hand have improved, Brentford fallen back. To see the explanation of expected and forecast positions see the notes given on the site. 

[Hank shoots Skyler 2,094]

On ‎26‎/‎10‎/‎2020 at 19:50, lake district canary said:

You'd be better off just guessing at how many goals a team might score in any match.

Of course the xG can be widely inaccurate in the course of a match, or even a run of fixtures, but the point is in the long term the current outliers are likely to edge their ways towards the mean - or otherwise keep pulling off unlikely results based on their efforts for an entire season.

There is a direct trend between xG and actual goals scored / conceded, so there is clearly some degree of accuracy to it. I think xG can also be a good measure of how fortunate you were to pick up a result in a game - i.e. Derby beating us.

Our current position suggests we should be scoring more goals - and IMO we will start to see this soon. I also think we will see Reading drop off somewhat when they start missing chances.

PS this thread feels empty without Bill..

[It's Character Forming 1,204]

In 18/19 when we scored a lot more than xG predicted (also conceded more) I thought it was a load of rubbish.  Now that it shows we "should" be doing better than we are, I'm definitely liking it...

 

Seriously, I think xG is useful (certainly a lot more usual that the stats we normally hear, like possession which tells you literally nothing about how well the teams were playing) as long as people realise it's just a start point for the analysis and it's not a case of "xG tells you what should have happened."

 

If you compare football with other team sports like say rugby, American football, ice hockey which are the main team sports where the teams are trying to score at the opposition end of the playing field (basketball different in that scoring is so frequent).  In all of them, for any given scoring chance, there's a possibility of scoring but never a certainty - it could go either way.

 

& If you look at those other sports, it's normal for both teams to score several times and the game comes down to who scores more often (obviously ignoring the different points you get for a try/penalty etc).  Not to score at all is rare (and normally means one side is completely dominant). Whereas football is a low scoring game, and it's not unusual for one or even both sides to score nil.  Can you imagine a game of rugby or American football ending 0-0 ?

 

This means football comes down to very fine margins.  Often just a handful of chances which could go either way.  Whereas for other team sports, normally it will come out in the wash, if you miss one chance, mostly you'll put away another one.  I'm not saying rugby or American football don't have close outcomes, of course they do (and American football is very much designed to cram excitement into the last 2 minutes as possible). But the difference is that football at professional level, most games come down to a handful of moments which could go either way.

 

This is why coaches find xG useful.  They know that all they can do is setup their team to create as many chances as they can and stop the opposition creating too many, and of course coach their team to be the best they can at finishing, but I'm sure DF knows he can set out his team, they can play better than the opposition, create more chances, but just not put any away, then Rooney takes a free kick right at the end....  xG simply shows you "on average" how many goals the sides would have scored from the chances created and therefore gives a good indication of how well your team has performed - often, from the coach's perspective, a better indication than the scoreline itself.

 

But when you look at xG, the question always must be - why is it different from the actual scoreline (if it is)?  Looking back at 18/19, there was a poster called @westcoastcanary who loved the xG stats but seemed to take the view that because Norwich were scoring more than our xG, before too long we'd simply score less (although the fact that we conceded more than xG seemed to him to mean we were just defending badly).  (I also remember him saying just after Christmas that season "no sign of Leeds imploding yet", shortly before the Leeds implosion started).

 

What actually happened in 18/19 IMO is that our finishing for the first half of the season was just outstanding, hence we scored a lot more than xG would indicate.  Towards the end of the season, our xG improved because we were creating more chances, and the gap between that and actual goals narrowed and I do think our finishing because slightly less clinical.

[TeemuVanBasten 3,328]

On 25/10/2020 at 10:51, Wacky Waving Inflatable Arm Flailing Tube Man said:

With a goals against xG of 5.6, it just shows that their opponents have been very wasteful.

No idea what this really mean, but could it not mean that they have a great keeper? 

Edited October 29, 2020 by TeemuVanBasten

[king canary 8,758]

13 hours ago, It's Character Forming said:

In 18/19 when we scored a lot more than xG predicted (also conceded more) I thought it was a load of rubbish.  Now that it shows we "should" be doing better than we are, I'm definitely liking it...

 

Seriously, I think xG is useful (certainly a lot more usual that the stats we normally hear, like possession which tells you literally nothing about how well the teams were playing) as long as people realise it's just a start point for the analysis and it's not a case of "xG tells you what should have happened."

 

If you compare football with other team sports like say rugby, American football, ice hockey which are the main team sports where the teams are trying to score at the opposition end of the playing field (basketball different in that scoring is so frequent).  In all of them, for any given scoring chance, there's a possibility of scoring but never a certainty - it could go either way.

 

& If you look at those other sports, it's normal for both teams to score several times and the game comes down to who scores more often (obviously ignoring the different points you get for a try/penalty etc).  Not to score at all is rare (and normally means one side is completely dominant). Whereas football is a low scoring game, and it's not unusual for one or even both sides to score nil.  Can you imagine a game of rugby or American football ending 0-0 ?

 

This means football comes down to very fine margins.  Often just a handful of chances which could go either way.  Whereas for other team sports, normally it will come out in the wash, if you miss one chance, mostly you'll put away another one.  I'm not saying rugby or American football don't have close outcomes, of course they do (and American football is very much designed to cram excitement into the last 2 minutes as possible). But the difference is that football at professional level, most games come down to a handful of moments which could go either way.

 

This is why coaches find xG useful.  They know that all they can do is setup their team to create as many chances as they can and stop the opposition creating too many, and of course coach their team to be the best they can at finishing, but I'm sure DF knows he can set out his team, they can play better than the opposition, create more chances, but just not put any away, then Rooney takes a free kick right at the end....  xG simply shows you "on average" how many goals the sides would have scored from the chances created and therefore gives a good indication of how well your team has performed - often, from the coach's perspective, a better indication than the scoreline itself.

 

But when you look at xG, the question always must be - why is it different from the actual scoreline (if it is)?  Looking back at 18/19, there was a poster called @westcoastcanary who loved the xG stats but seemed to take the view that because Norwich were scoring more than our xG, before too long we'd simply score less (although the fact that we conceded more than xG seemed to him to mean we were just defending badly).  (I also remember him saying just after Christmas that season "no sign of Leeds imploding yet", shortly before the Leeds implosion started).

 

What actually happened in 18/19 IMO is that our finishing for the first half of the season was just outstanding, hence we scored a lot more than xG would indicate.  Towards the end of the season, our xG improved because we were creating more chances, and the gap between that and actual goals narrowed and I do think our finishing because slightly less clinical.

Pretty spot on summary really.

Under performing on goals scored doesn't mean we'll inevitably start scoring more- it might just mean we've got too many below average finishers.

I wonder how much of this hangs on Pukki's form- I believe in 18/19 he hugely overperformed from an xG perspective, he's come back down to earth somewhat of late so that brings the overall team down.

Also a higher xG doesn't mean you deserved to win- it just means you didn't take your chances.

[Wacky Waving Inflatable Arm Flailing Tube Man 4,620]

7 hours ago, TeemuVanBasten said:

No idea what this really mean, but could it not mean that they have a great keeper? 

I suppose it could actually, but I guess it depends if the shots were on target.