Statistical Domination

[king canary 7,607]

Just now, hogesar said:

Yes, good summary.

You have to say, for all the raw attributes Sara has, he has not once looked like playing the "match". I've not seen him adapt for different scenarios and he doesn't seem to have the instinct to do the right things when it needs to be simple. The quick free kick is an obvious one.

If anything I'd say this game really highlighted how much of a black hole we have in defensive midfield. 

Look at the first goal- yeah it is a counter attack but when Sorenson overcommits nobody in midfield is anywhere to be seen. It becomes an even bigger issue when you've got a relatively inexperienced pair of central defenders in the team.

[shefcanary 2,409]

12 minutes ago, king canary said:

1. Sorenson goes chasing after the ball, pulls everyone out of position and is absolutely powderpuff in the tackle. Concerning though that Gibbs is the only midfielder who is back in shot by the time the goal goes in. Aarons left totally exposed by Marquinos all game.

Ben Lee's tactical analysis explains Sorensen had to go for the ball because for some reason Onel left his player to have the freedom of the pitch (perhaps because we had made such a gung ho start?). You defend from the front and our attacking players seemed to forget this as a collective unit in that first half. This left our young defence hopelessly exposed and once that had happened all confidence was shot. 

[hogesar 9,676]

2 minutes ago, king canary said:

If anything I'd say this game really highlighted how much of a black hole we have in defensive midfield. 

Look at the first goal- yeah it is a counter attack but when Sorenson overcommits nobody in midfield is anywhere to be seen. It becomes an even bigger issue when you've got a relatively inexperienced pair of central defenders in the team.

At this level, not even a genuine defensive midfielder. Just one that senses danger.

You know where I'm going with this. And that's Kenny Mclean.

BUT

I think he's actually not very good at sensing danger, but under Wagner he has been pretty good at being disciplined and being so deep that he offers a level of protection that's decent enough at this level. Gibbs likes to chase and harry and but is then caught out of position....

MORE THAN TWO MIDFIELDERS WOULD BE NICE. We saw it vs Blackburn. Now Nunez is fit I'd really hope we don't have both Pukki and Sargent up top. Sargent can't play number 10 and when he comes deep his touch is poor and we give teams a chance to attack us on transition.

[Robert N. LiM 4,356]

58 minutes ago, hogesar said:

I think the forum would really warm to me starting a pre-game and post-game xG thread on each match day. Dont you think?

Personally I prefer the 20-page, 20,000-view behemoth sitting atop the homepage like Leviathan. On reflection, just continue this one next season. It might break the internet.

[Robert N. LiM 4,356]

51 minutes ago, hogesar said:

At this level, not even a genuine defensive midfielder. Just one that senses danger.

The fact that all three players we have who can vaguely play in that position - Kenny, Hayden and Lungi* - are unavailable really does suck.

*As in, Lungi is unavailable to play DM because we need him at CB.

 

[repman 635]

1 hour ago, hogesar said:

At this level, not even a genuine defensive midfielder. Just one that senses danger.

You know where I'm going with this. And that's Kenny Mclean.

BUT

I think he's actually not very good at sensing danger, but under Wagner he has been pretty good at being disciplined and being so deep that he offers a level of protection that's decent enough at this level. Gibbs likes to chase and harry and but is then caught out of position....

Obviously in the role that Kenny was playing there will always be a trade off between what he can do on and off the ball.

Just looking at the tackles + interception numbers for the players in that DM position. 

Sorensen: 4.18 p90
Hayden: 3.97 p90
Gibbs: 2.75 p90
McLean: 2.65 p90

Obviously it's not a perfect comparison as Kenny has the largest sample size, Sorensen and Hayden both have very small ones while Gibbs has a larger one but still half the 90s played by Kenny. Also of course Sorensen and Gibbs have spent time playing in other positions this year too. 

Kenny might be more disciplined than under Wagner, and he may be more so than Gibbs, but I do think he has that tendency to jump out of the midfield line often too. He's certainly not a screening player that's for sure.

As I mentioned though, there's always a trade off in that role and when you look at the progressive passes stat it reads in near enough reverse order.

McLean: 6.56 p90
Gibbs: 5.29 p90
Sorensen: 2.97 p90
Hayden: 2.31 p90

The big positive out of these numbers is Gibbs imo. For a guy in his first season, he's probably the last player you'd expect to be the jack of all trades. The fact he can do both sides decently well is a bonus early on, and if he grows both tactically and physically at a steady rate over the next few years you could have an excellent player there. The key now is getting him plenty of minutes on the pitch, due to the injuries we suffered early on in the year he skipped the loan move like most of our academy players get. He more than met the challenges he faced this year so sending him out to a weaker side in an attempt to guarantee minutes seems counterproductive to me. 

1 hour ago, hogesar said:

MORE THAN TWO MIDFIELDERS WOULD BE NICE. 

Ultimately though, between now and the end of the year if we want to do well this is the best solution.

[king canary 7,607]

50 minutes ago, shefcanary said:

Ben Lee's tactical analysis explains Sorensen had to go for the ball because for some reason Onel left his player to have the freedom of the pitch (perhaps because we had made such a gung ho start?). You defend from the front and our attacking players seemed to forget this as a collective unit in that first half. This left our young defence hopelessly exposed and once that had happened all confidence was shot. 

I can see that but I'd also argue this is where a good quality dm comes in. We've lacked for years a player who plays horizontally rather than vertically. The type with the awareness to know that if Onel goes forward he might get caught out and either looks to cover or to drop deep and let the Central defenders split. Neither Gibbs or Sara have that sort of awareness.

[Jim Smith 2,319]

15 hours ago, king canary said:

If anything I'd say this game really highlighted how much of a black hole we have in defensive midfield. 

Look at the first goal- yeah it is a counter attack but when Sorenson overcommits nobody in midfield is anywhere to be seen. It becomes an even bigger issue when you've got a relatively inexperienced pair of central defenders in the team.

I thought Gibbs (who I rate) got done for at least a couple of the goals not tracking back/getting goal side of the runner which again does highlight the cdm issue.

I’d also throw McCullums throw in in the build up to one of their goals (can’t recall if  3rd or 4th) into the mix. He took a ridiculous throw to Sara who miscontrolled it and fouled his man which led directly to their goal. 

 

[Jim Smith 2,319]

13 hours ago, repman said:

Obviously in the role that Kenny was playing there will always be a trade off between what he can do on and off the ball.

Just looking at the tackles + interception numbers for the players in that DM position. 

Sorensen: 4.18 p90
Hayden: 3.97 p90
Gibbs: 2.75 p90
McLean: 2.65 p90

Obviously it's not a perfect comparison as Kenny has the largest sample size, Sorensen and Hayden both have very small ones while Gibbs has a larger one but still half the 90s played by Kenny. Also of course Sorensen and Gibbs have spent time playing in other positions this year too. 

Kenny might be more disciplined than under Wagner, and he may be more so than Gibbs, but I do think he has that tendency to jump out of the midfield line often too. He's certainly not a screening player that's for sure.

As I mentioned though, there's always a trade off in that role and when you look at the progressive passes stat it reads in near enough reverse order.

McLean: 6.56 p90
Gibbs: 5.29 p90
Sorensen: 2.97 p90
Hayden: 2.31 p90

The big positive out of these numbers is Gibbs imo. For a guy in his first season, he's probably the last player you'd expect to be the jack of all trades. The fact he can do both sides decently well is a bonus early on, and if he grows both tactically and physically at a steady rate over the next few years you could have an excellent player there. The key now is getting him plenty of minutes on the pitch, due to the injuries we suffered early on in the year he skipped the loan move like most of our academy players get. He more than met the challenges he faced this year so sending him out to a weaker side in an attempt to guarantee minutes seems counterproductive to me. 

Ultimately though, between now and the end of the year if we want to do well this is the best solution.

Sorensen should be playing there but because the idiot went with a cebtre back short again for the second half of the season he’s having to play CB. Personally I would risk an U21 or even Byram at CB for the remaining games and get Sorensen back in midfield. 

[hogesar 9,676]

2 hours ago, Jim Smith said:

Sorensen should be playing there but because the idiot went with a cebtre back short again for the second half of the season he’s having to play CB. Personally I would risk an U21 or even Byram at CB for the remaining games and get Sorensen back in midfield. 

Yes, Byram is an option. Hayden is even an option. Don't know how close they are to 90 minutes but with both out of contract we might as well play them till the end of the season - if they get injured again, in a very ruthless world, it's not really our problem.

[shefcanary 2,409]

1 hour ago, hogesar said:

Yes, Byram is an option. Hayden is even an option. Don't know how close they are to 90 minutes but with both out of contract we might as well play them till the end of the season - if they get injured again, in a very ruthless world, it's not really our problem.

Play Byram for as long as he can take it, play Sorenson in the CDM role, hopefully Sarge converts an early chance and we coast the rest of the match. Then bring Byram off, Sorenson to drop back to CB, Hayden on as the CDM. Simples?

Or is that a pig flying over there.

[repman 635]

Thought I'd revive this thread with the 5 game xG rolling average now the season is over. Doesn't take long to see that things are not good at all. I'm sure straight away people will point to injuries as the reason have things gone so wrong. I tend to think most defending is dependent on the system and shouldn't vary that much per player unless you play like Man City or Liverpool where you're asking very specific things of your defenders.

Wagner's inability to change things/solve problems after his first few games is clear here. The time around February on the chart is probably the ideal of Wagner's football, low scoring and effectively nullifying games. Not pretty but maybe it can work. 

To end the year we have a 5 game average of 1.28xG/1.87xGA, some of the worst marks on both ends all year.

[It's Character Forming 1,161]

47 minutes ago, repman said:

Thought I'd revive this thread with the 5 game xG rolling average now the season is over. Doesn't take long to see that things are not good at all. I'm sure straight away people will point to injuries as the reason have things gone so wrong. I tend to think most defending is dependent on the system and shouldn't vary that much per player unless you play like Man City or Liverpool where you're asking very specific things of your defenders.

Wagner's inability to change things/solve problems after his first few games is clear here. The time around February on the chart is probably the ideal of Wagner's football, low scoring and effectively nullifying games. Not pretty but maybe it can work. 

To end the year we have a 5 game average of 1.28xG/1.87xGA, some of the worst marks on both ends all year.

I'd point to injuries as the reason have things gone so wrong.  I think the idea defending is just "about the system" and to think you can lose both your first choice CBs at the same time as losing your only DM (or the nearest to that role we have) without your defence suddenly looking extremely porous is, to be honest, a bit bonkers.  There you go.

 

Also I'd be interested in the breakdown here, last 5 games have been 2 very bad results and 2 narrow defeats plus a draw, it would be interesting to see what xG says about those 3 results and how far the average xGA is skewed down by the 2 bad results.

 

Based on his historic successful period with Huddersfield, is there any actual reason to say Wagner's ideal is probably low scoring? If you can point to anything at all that actually supports that I'd be interested, but if it's just your opinion based on our results in Feb, then you're entitled to your opinion (but don't expect me to agree).

 

For me the last part of the season the big issue, aside from the defence (see above) is that we've been creating plenty of chances (mostly) but failing to score, and we're very fragile and tend to crumple when we concede a goal.  Is that down to Wagner ?

[hogesar 9,676]

Thanks @repman

I think it's very hard to tell as an average fan how a manager will perform when given his own time to train, tune, and more importantly buy his own players. Certainly, if you go back to Farke's first season (yeah, that old chestnut) - read the threads at the time.

- Concede soft goals

- Pass the ball around the back but give it away needlessly

- Can't score

Were all accusations thrown at Farke. We went from midtable also-rans to title winners despite selling our best player(s).

Before I continue, let me say now that I've seen very little in Wagner that suggests he's a definite improvement on Smith. I've seen nothing if anything to suggest he is as good as Farke. I wouldn't be sad to see us make a change if we felt it necessary. However:

• I've never seen a central defensive partnership as inept as Omo and Sorensen in a very long time. They both lose basic individual battles and both make rash decisions with challenges.
• With no Mclean in midfield we had no real strength and grit in front of the back four. Mclean isn't great at this, but he is head and shoulders above Gibbs trying to do that role, who unfortunately is too weak physically and in the air to be a sole option for us at this level
• Sara makes too many mistakes to come and sit deep as well - he's a talented player but only when high up the pitch. 
• We had no other midfield options to try and solidify the team. The injuries and squad left for Wagner didn't really give him lots of options to mix things up.
• No other manager has had to deal with this squad missing the leadership and experience of both Hanley and Mclean. Much maligned by some sets of fans but it was pretty obvious to me at least that this current crop of players would really, really miss them. Even I didn't think it would be to this level, though.
• If we use xG as a measure (which is what this thread is about) then before the injuries we had a Wagner xG of 17.7 xGF vs 12.4 xGA. Under Wagner, with Mclean and Hanley, we generally created more chances than we conceded.

 

[king canary 7,607]

11 minutes ago, It's Character Forming said:

Based on his historic successful period with Huddersfield, is there any actual reason to say Wagner's ideal is probably low scoring? If you can point to anything at all that actually supports that I'd be interested, but if it's just your opinion based on our results in Feb, then you're entitled to your opinion (but don't expect me to agree).

Well...the stats from that successful period kind of point to that.

After he took over from Chris Powell they scored 43 goals in 30 games.

In their promotion season Huddersfield scored 56 goals, putting them in the bottom half of the league for goals scored.

After promotion they stayed up with the joint lowest total of goals scored in the division.

Wagner's Huddersfield were never free scorers.

[repman 635]

6 minutes ago, It's Character Forming said:

I'd point to injuries as the reason have things gone so wrong.  I think the idea defending is just "about the system" and to think you can lose both your first choice CBs at the same time as losing your only DM (or the nearest to that role we have) without your defence suddenly looking extremely porous is, to be honest, a bit bonkers.  There you go.

 

Also I'd be interested in the breakdown here, last 5 games have been 2 very bad results and 2 narrow defeats plus a draw, it would be interesting to see what xG says about those 3 results and how far the average xGA is skewed down by the 2 bad results.

 

Based on his historic successful period with Huddersfield, is there any actual reason to say Wagner's ideal is probably low scoring? If you can point to anything at all that actually supports that I'd be interested, but if it's just your opinion based on our results in Feb, then you're entitled to your opinion (but don't expect me to agree).

 

For me the last part of the season the big issue, aside from the defence (see above) is that we've been creating plenty of chances (mostly) but failing to score, and we're very fragile and tend to crumple when we concede a goal.  Is that down to Wagner ?

No I do think injuries have played some part in the quite dramatic slide but there were signs before the injury that it wasn't necessarily great either. If you combine that with the lack of chances created then you have an average side. Under Farke we weren't good defensively but we made up for it by creating much more.

To your question about Wagner's Huddersfield, his promoted side scored 56 goals and finished with a negative goal difference. There's no xG data available that I can find for that season so I can't tell you whether their underlying numbers said they should've had more/less. What is available is their 17/18 PL numbers, where they registered the lowest xG in the league but their xGA was ranked 11th. Obviously the task in the championship vs the PL is very different, in one you need to win 25+ games whereas in the PL you'd be happy with 10 wins. I think a lot of Norwich fans remember his Huddersfield side beating us 3-0 and think that was representative of their side, it was one of only 4 times that season they scored 3 goals. When you combine this with the fact his coaching expertise if you will is pressing, I think it's fair to say his first aim is to prevent goals and then hope to generate goals second (Not to say this is a bad thing btw).

Just to the bit about the last 5 games

Middlesbrough 0.84 - 2.72
QPR 2.22 - 1.57
Swansea 0.75 - 2.22
WBA 0.61 - 2.01
Blackpool 1.99 - 0.85

If anything I would say it's 2 positive games skewing the numbers up rather than the other way.

[king canary 7,607]

12 minutes ago, repman said:

To your question about Wagner's Huddersfield, his promoted side scored 56 goals and finished with a negative goal difference. There's no xG data available that I can find for that season so I can't tell you whether their underlying numbers said they should've had more/less. What is available is their 17/18 PL numbers, where they registered the lowest xG in the league but their xGA was ranked 11th. Obviously the task in the championship vs the PL is very different, in one you need to win 25+ games whereas in the PL you'd be happy with 10 wins. I think a lot of Norwich fans remember his Huddersfield side beating us 3-0 and think that was representative of their side, it was one of only 4 times that season they scored 3 goals. When you combine this with the fact his coaching expertise if you will is pressing, I think it's fair to say his first aim is to prevent goals and then hope to generate goals second (Not to say this is a bad thing btw).

 

I think they played some good football but I also think, looking back, this was a team that really rode their luck those two seasons. An incredible 22 of their 25 wins that promotion season were won by a single goal and they didn't actually win a single game in the playoffs- the drew all 3 and won on penalties in the semi and the final.

[repman 635]

20 minutes ago, hogesar said:

 

• If we use xG as a measure (which is what this thread is about) then before the injuries we had a Wagner xG of 17.7 xGF vs 12.4 xGA. Under Wagner, with Mclean and Hanley, we generally created more chances than we conceded.

 

I agree with a lot of your points in your post. Just looking at this bit though, I know we have different sources and for the 13 games that both McLean and Hanley played I have it as 14.9 - 12.55. That works out at an xGD per game of 0.18, which would be joint 7th in the league (our overall one this season was 0.08 p90).

[It's Character Forming 1,161]

35 minutes ago, king canary said:

Well...the stats from that successful period kind of point to that.

After he took over from Chris Powell they scored 43 goals in 30 games.

In their promotion season Huddersfield scored 56 goals, putting them in the bottom half of the league for goals scored.

After promotion they stayed up with the joint lowest total of goals scored in the division.

Wagner's Huddersfield were never free scorers.

Is that Wagner's "ideal" ?  Or was it the way he got that Huddersfield team promoted (and kept them up) ?

[king canary 7,607]

1 minute ago, It's Character Forming said:

Is that Wagner's "ideal" ?  Or was it the way he got that Huddersfield team promoted (and kept them up) ?

Well we can only go on what he's done- the above for Huddersfield, 38 goals in 34 games his full season at Schalke.

If he wants his team to score more goals he's not very good at getting them to do it.

[Ulfotto 634]

I would throw in that at the start of the season we seemed to have too many players and many decent players were not even making the match day squad.

Fast forward to the end of the season and it seems we now have a total lack of options all over the pitch. 

[king canary 7,607]

13 hours ago, Ulfotto said:

I would throw in that at the start of the season we seemed to have too many players and many decent players were not even making the match day squad.

Fast forward to the end of the season and it seems we now have a total lack of options all over the pitch. 

I think this speaks to a lack of progress from our younger options combined with some poor squad building.

At the start of the season we had Cantwell, Sinani, Rashica, Hernandez, Ramsey, Dowell and Nunez all competing for the same 3 spots, all with the same issues of being inconsistent and largely lightweight. Then we also had Rower and Springett on the edges of the first team who were injured or loaned out.

Similarly in central midfield we've had McLean, Sara and Gibbs all suited to being the kind of midfielder who breaks from deep but only an unfit Hayden to address DM. 

3 central defenders on the books but 4 strikers up until we let Hugill go.

Just a bit of a mess. I think clearing out a few in January was the right way to go. We need quality not quantity- a couple of players who are a no brainer to be starting week in week out and can deliver a bit of consistency. 

[Dr Greenthumb 749]

On 09/05/2023 at 15:49, hogesar said:

Thanks @repman

I think it's very hard to tell as an average fan how a manager will perform when given his own time to train, tune, and more importantly buy his own players. Certainly, if you go back to Farke's first season (yeah, that old chestnut) - read the threads at the time.

- Concede soft goals

- Pass the ball around the back but give it away needlessly

- Can't score

Were all accusations thrown at Farke. We went from midtable also-rans to title winners despite selling our best player(s).

Before I continue, let me say now that I've seen very little in Wagner that suggests he's a definite improvement on Smith. I've seen nothing if anything to suggest he is as good as Farke. I wouldn't be sad to see us make a change if we felt it necessary. However:

• I've never seen a central defensive partnership as inept as Omo and Sorensen in a very long time. They both lose basic individual battles and both make rash decisions with challenges.
• With no Mclean in midfield we had no real strength and grit in front of the back four. Mclean isn't great at this, but he is head and shoulders above Gibbs trying to do that role, who unfortunately is too weak physically and in the air to be a sole option for us at this level
• Sara makes too many mistakes to come and sit deep as well - he's a talented player but only when high up the pitch. 
• We had no other midfield options to try and solidify the team. The injuries and squad left for Wagner didn't really give him lots of options to mix things up.
• No other manager has had to deal with this squad missing the leadership and experience of both Hanley and Mclean. Much maligned by some sets of fans but it was pretty obvious to me at least that this current crop of players would really, really miss them. Even I didn't think it would be to this level, though.
• If we use xG as a measure (which is what this thread is about) then before the injuries we had a Wagner xG of 17.7 xGF vs 12.4 xGA. Under Wagner, with Mclean and Hanley, we generally created more chances than we conceded.

 

Good post.

It’s no coincidence we went shockingly bad after the Blackburn injuries and McLean prior. 
It’s worrying we haven’t been able to upgrade in the central midfield position in 3 years!

Omo has to be one of the most overrated centre backs I’ve seen in a Norwich shirt, but the hype is great for his price tag.

Wagner doesn’t seem a great manager at all, but he really has had a rough time with injuries and a severe lack of wingers to utilise. 
 

Just hope the fire is reignited in the recruitment team and the signings are ready and waiting to come through the door.

[Robert N. LiM 4,356]

@hogesar

 

 

[hogesar 9,676]

11 minutes ago, Robert N. LiM said:

@hogesar

 

 

Statistical Domination v2 en route 😅

[Robert N. LiM 4,356]

10 hours ago, hogesar said:

Statistical Domination v2 en route 😅

YES HOGGY

[repman 635]

Don't know if someone will start a new topic as it's a new season but I'll post it here anyway.

I know last year there were a few different sources for these charts and they weren't always posted, I got the data from fbref like in the tweet above.

Looks pretty good to me (perhaps with some caveats)

[It's Character Forming 1,161]

24 minutes ago, repman said:

Don't know if someone will start a new topic as it's a new season but I'll post it here anyway.

I know last year there were a few different sources for these charts and they weren't always posted, I got the data from fbref like in the tweet above.

Looks pretty good to me (perhaps with some caveats)

That's useful to see, thanks for posting, and I'd say it's a fair reflection of the match - the 2-1 scoreline flattered them and we were unfortunate that we needed to get a goal right at the death to get the win.  But at least we're comparing a 2-1 actual score with an xG differential of a couple of goals in our favour (not trying to justify a draw/defeat on the basis of xG showing we should have done better).

[littleyellowbirdie 2,550]

For my tastes, a fresh thread for the new season would be easier to refer to in the future, but I think it's more up to you and Hoggy to decide how you want to do it, seeing as it's mostly you two who put the information up.

[Fuzzar 1,702]

44 minutes ago, littleyellowbirdie said:

For my tastes, a fresh thread for the new season would be easier to refer to in the future, but I think it's more up to you and Hoggy to decide how you want to do it, seeing as it's mostly you two who put the information up.

Statistically the older thread would get more xPected views.