If the table was sorted by XG..

[cambridgeshire canary 6,865]

[Pyro Pete 1,940]

Would you like to add the source for this table?

Edit: Ah, I see it's from Reddit. Hmm.

Edited December 21, 2022 by Pyro Pete

[hogesar 9,806]

Probably about right looking at the numbers because particularly earlier on in the season and shortly after the unbeaten run we dropped points somewhat undeserved, where we'd clearly created the better chances.

We also, for example created better chances than Boro at home, where we shouldn't have lost.

What that table doesn't really do is explain or identify the recent downturn in xG where we are creating less but conceding more in general.

[GJL Mid-Norfolk Canary 1,818]

You're Dean Smith and I claim my £5...

[canarydan23 4,118]

36 minutes ago, hogesar said:

We also, for example created better chances than Boro at home, where we shouldn't have lost.

Does that mean you accept we shouldn't have won versus Swansea?

[hogesar 9,806]

1 minute ago, canarydan23 said:

Does that mean you accept we shouldn't have won versus Swansea?

Yeah, not that I find it difficult because I said as such already in more than one thread. I also thought we were pretty fortunate vs Rotherham.

[Norfolk Dan 308]

30 minutes ago, GJL Mid-Norfolk Canary said:

You're Dean Smith and I claim my £5...

£5 for DS... Be lucky to get 20p

[canarydan23 4,118]

Just now, hogesar said:

Yeah, not that I find it difficult because I said as such already in more than one thread. I also thought we were pretty fortunate vs Rotherham.

Well at least your consistent. A lot of the people who were belittling those whose concerns were not allayed by Swansea, despite the result, were the same people quoting xG at us when we weren't getting results at the start of the season.

As long as you're being consistent, I don't mind people being a disciple of this malarkey. But it is malarkey; given that after 23 games, half a season, teams are generally were they deserve to be, this table shows as much. Middlesbrough deserve to be in the bottom half. Burnley deserve to be top, not 5th. Blackburn are genuine playoff contenders and shouldn't be sitting one place above the drop zone.

That there are so many outliers shows it's not really an accurate measure of anything.

[hogesar 9,806]

38 minutes ago, canarydan23 said:

Well at least your consistent. A lot of the people who were belittling those whose concerns were not allayed by Swansea, despite the result, were the same people quoting xG at us when we weren't getting results at the start of the season.

As long as you're being consistent, I don't mind people being a disciple of this malarkey. But it is malarkey; given that after 23 games, half a season, teams are generally were they deserve to be, this table shows as much. Middlesbrough deserve to be in the bottom half. Burnley deserve to be top, not 5th. Blackburn are genuine playoff contenders and shouldn't be sitting one place above the drop zone.

That there are so many outliers shows it's not really an accurate measure of anything.

Well, it depends how much you're willing to utilise and perhaps how much weight you put towards it.

For example, xG suggested to us that in our last promotion season, we should have conceded more goals and were a bit "lucky" not to do so. In fact, xG suggested Brentford were a better team than us. The following season proved it (with all the caveats that go with it obviously).

[canarydan23 4,118]

3 minutes ago, hogesar said:

The following season proved it

The previous season's xG did nothing to prove Brentford were better than us in that season. That season's performances proved that.

And the previous season's league table proved that we were better than Brentford.

If the xG didn't show that, the xG was wrong.

[nutty nigel 7,656]

XG will always be a view of the past but I understand it can be used to affect results in the future. I don't think it was ever meant to be 'a we are better than our results' thing.

[hogesar 9,806]

50 minutes ago, canarydan23 said:

The previous season's xG did nothing to prove Brentford were better than us in that season. That season's performances proved that.

And the previous season's league table proved that we were better than Brentford.

If the xG didn't show that, the xG was wrong.

 

4 minutes ago, nutty nigel said:

XG will always be a view of the past but I understand it can be used to affect results in the future. I don't think it was ever meant to be 'a we are better than our results' thing.

Correct. It can be used as an alternative analysis to simply, the results. We know football has so many factors involved and luck is a huge slice of it. Its part of why we love it. Its unpredictable and luck plays a huge role. Fans arguably play a huge role.

So it can be used as a retrospective view but yes, it's intention was to predict and make money in betting markets, which before bookies adopted the stats themselves, it did reap huge rewards.

[Canaries north 151]

You hear some fans talking about how many shots we've had as a reason we should have won a game and this is crazy without knowing how good the chances were. XG is more accurate for this but you still have so many other factors to consider. What if Matt LeTissier was playing today. So many of his goals were from what XG would count as very small chances and very few were from very high scoring chances. I'm sure the XG from a team with him in it might be next to nothing but he could still walk away with a match ball. 

[repman 640]

I think the big thing to note from that table regarding the 2 big outliers at the top (Boro and WBA) is that both those sides started off very poorly and have now started to pick up under new managers. Both of their previous managers were sacked due to this underperformance, and the new managers seem to be getting their teams up the league which is where you would say their squads belong. Their big problems early on was the simple fact they kept on conceding the first goal.

Baseball, the most data efficient sport, has 162 games per season and each player will come to the plate maybe 500 times a year. The data set for that can therefore equal out over time. Basketball, a more similar game to football in the sense that it's an 'invasion' sport, will see at least 80+ shots per game in an 82 game season. In Norwich's games this season there has been on average 25 shots per game. The average goals per game this season in the championship is 2.44, conceding 1 stacks the odds against you winning that game right away. It's why looking at xG in a vacuum can be bad. If a team started every game 1-0 down then they would probably have a very good xG, owing to the fact their opponents would have less reason to attack when they have a lead to sit on.

Over the course of the season the xG table and league table will converge to a strong degree but there will always be a few outliers. xG basically posits the theory that if you have more shots, in better positions, than your opponents then you will have a better chance of winning the game. Unfortunately (and fortunately!) football really isn't that easy, in fact the beauty of football is probably how unfair it is. If we wanted the game to be truly fair we would have long seasons like baseball, or we would play for longer than 90 minutes, or even play until a side has a 3 goal lead. All these would be 'fairer' barometers for ranking the best teams but that's not the way the sport works.

 

2 hours ago, Canaries north said:

You hear some fans talking about how many shots we've had as a reason we should have won a game and this is crazy without knowing how good the chances were. XG is more accurate for this but you still have so many other factors to consider. What if Matt LeTissier was playing today. So many of his goals were from what XG would count as very small chances and very few were from very high scoring chances. I'm sure the XG from a team with him in it might be next to nothing but he could still walk away with a match ball. 

There's 2 points I'd raise against this.

1. xG is generally a good way of assessing team performance, if you were a manager putting together your side then you might have Le Tissier in the team scoring from distance, but what if he gets injured? You'd much rather a side which created 5 shots inside the box than outside of it.

2. While you can remember the great goals Le Tissier scored, what you probably won't remember is the number of shots that went into row Z. Over time I'm sure it would've balanced out in terms of xG. 

[AJ 1,235]

This is all good and well but is there also an xG for goals conceded? I'd love to see how many we SHOULD have conceded (statistically) compared to how many we actually have.

[TeemuVanBasten 3,327]

Its out of date.

Here you go...

 

Source:

https://footballxg.com/xg-league-tables/

[TeemuVanBasten 3,327]

4 minutes ago, AJ said:

This is all good and well but is there also an xG for goals conceded? I'd love to see how many we SHOULD have conceded (statistically) compared to how many we actually have.

The table I posted above has xG (for), xGA (conceded) and xG points (number of points our xG suggest we should have). Table is sorted by xG points. 

[Mello Yello 2,314]

xG is about as much use as a bee with hay-fever....

[aBee 152]

12 hours ago, canarydan23 said:

The previous season's xG did nothing to prove Brentford were better than us in that season. That season's performances proved that.

And the previous season's league table proved that we were better than Brentford.

If the xG didn't show that, the xG was wrong.

You’re right. 
What the xG did suggest was that you’d substantially overperformed and we’d somewhat underperformed. Your overperformance was likely to be in part from having a couple of players who were too good for the Championship (eg Buendia scoring chances that were low xG because he was just that much better than the typical level, Skipp giving defensive cover to make opponents less able to take advantage of good attacks). Weakening in both those areas on promotion predictably was unhelpful in the higher division. 
 

Whereas our underperformance suggested there were tweaks which could address the reasons for it- eg by becoming a bit more direct at times and tightening defensively. 
 

You could arrive at similar conclusions without xG but xG certainly was a high level validation for us of what needed doing next time round. In more traditional ways we translated it to “we will not let anyone beat us without working for it and we will work to provide a threat to anyone”. It contrasted with “well, obviously we can’t expect anything from our opening fixtures other than against Watford”. 

[hogesar 9,806]

1 hour ago, Mello Yello said:

xG is about as much use as a bee with hay-fever....

Try telling that to those who made millions from it.

Or to football clubs who utilise the stat along with hundred of others for match analysis, and consistently invest hundreds of thousands of pounds trying to improve stats like xG 🙂

 

[lake district canary 4,562]

1 hour ago, Mello Yello said:

xG is about as much use as a bee with hay-fever....

This is my impression too.

What over reliance on stats does - and xgs seems the most nebulous - is stultify players into thinking too much about stats rather than actual football.  The human race is gradually being led down the path of almost robotic responses - an extension of computers - and football is on that path too. 

Just let the players play with freedom, give them a few pointers to help of course, but let them just go out and play!  The best managers do this - they make it easy by giving simple clear instructions and then let the players get on with it.

[hogesar 9,806]

Just now, lake district canary said:

This is my impression too.

What over reliance on stats does - and xgs seems the most nebulous - is stultify players into thinking too much about stats rather than actual football.  The human race is gradually being led down the path of almost robotic responses - an extension of computers - and football is on that path too. 

Just let the players play with freedom, give them a few pointers to help of course, but let them just go out and play!  The best managers do this - they make it easy by giving simple clear instructions and then let the players get on with it.

Yet Farke spoke about xG on more than one occasion, and was known to be very meticulous and give detailed instructions to players. 

And actually, the best managers in the world, like Pep, like Klopp, talk regularly about xG and their investment into stats.

 

[Midlands Yellow 4,118]

11 minutes ago, hogesar said:

Yet Farke spoke about xG on more than one occasion, and was known to be very meticulous and give detailed instructions to players. 

And actually, the best managers in the world, like Pep, like Klopp, talk regularly about xG and their investment into stats.

 

Can you blame Pep and Klopp? It’s a riveting conversational topic. Starting to catch on on radio phone ins and the lads in the pubs pre match and post match talk about nothing else. 

[lake district canary 4,562]

10 minutes ago, hogesar said:

Yet Farke spoke about xG on more than one occasion, and was known to be very meticulous and give detailed instructions to players. 

And actually, the best managers in the world, like Pep, like Klopp, talk regularly about xG and their investment into stats.

The best managers, including Pep, will not fill players heads too much. Freedom within a framework is what players need to express themselves well. I get the impression that lesser managers try too hard to instill good things and that trying too hard is often counter productive. 

I believe our players at the moment are being given too much information and too much to think about - and can't process it quickly enough in matches. Stats can be helpful (there I said it) but only if used in the right way.

[TIL 1010 4,824]

Hindsight is an exact science.

[Capt. Pants 4,325]

One or two outliers but correlates quite well with league position.

It will be interesting to see where Middlesbrough and WBA end up.

[Nuff Said 5,239]

32 minutes ago, lake district canary said:

The best managers, including Pep, will not fill players heads too much. Freedom within a framework is what players need to express themselves well. I get the impression that lesser managers try too hard to instill good things and that trying too hard is often counter productive. 

I believe our players at the moment are being given too much information and too much to think about - and can't process it quickly enough in matches. Stats can be helpful (there I said it) but only if used in the right way.

You “believe” and “get the impression”. Basically you’re telling us what you think without any evidence along with a few self-fulfilling prophecies (“the best managers will…”).

Plus Smith actually said that the coaching team were trying to get the players to think more for themselves which directly contradicts what you say is going on at the club.

 

Sorry to be blunt but there’s so much nonsense on this board and this thread in particular, your post tipped me over the edge.

[hogesar 9,806]

48 minutes ago, lake district canary said:

The best managers, including Pep, will not fill players heads too much. Freedom within a framework is what players need to express themselves well. I get the impression that lesser managers try too hard to instill good things and that trying too hard is often counter productive. 

I believe our players at the moment are being given too much information and too much to think about - and can't process it quickly enough in matches. Stats can be helpful (there I said it) but only if used in the right way.

Pep is well known for being incredibly detailed and arguably too complex in his player instructions. The complete opposite. In fact one criticism is his absolute attention to detail in tactical analysis and detailed instructions impact on a players natural flair. 

I think you've just made a lot up without any sort of evidence. Please do some simple online reading about Pep first.

[repman 640]

3 hours ago, TeemuVanBasten said:

Its out of date.

Here you go...

 

Source:

https://footballxg.com/xg-league-tables/

This one is using the xG results from each game to award points, the one posted at the top is net xG.

[Mello Yello 2,314]

1 hour ago, hogesar said:

Try telling that to those who made millions from it.

Or to football clubs who utilise the stat along with hundred of others for match analysis, and consistently invest hundreds of thousands of pounds trying to improve stats like xG 🙂

 

Lies, Damned Lies.....and Statistics.....