Jump to content
IGNORED

Player Ratings


bcfc01

Recommended Posts

2 hours ago, And Its Smith said:

I hate xG but some people love it.  xG says we should have drawn today.  
 

Would like to hear the xG lovers justify that!! 

Eh? But doesn't that exactly back up what loads of people are saying - we also had loads of good chances but didn't take them?

I'd also add if you're looking at xG in one off scenarios it's just an indication you either haven't listened to what people have repeatedly said about it, or you just haven't understood. Probably the latter I'm guessing!

Link to comment
Share on other sites

36 minutes ago, IAmNick said:

Eh? But doesn't that exactly back up what loads of people are saying - we also had loads of good chances but didn't take them?

I'd also add if you're looking at xG in one off scenarios it's just an indication you either haven't listened to what people have repeatedly said about it, or you just haven't understood. Probably the latter I'm guessing!

It just proves it’s hugely flawed. We didn’t deserve to draw today.  The model doesn’t work.  I’ve listened to what people have said and am still convinced it doesn’t work as a model.  

Link to comment
Share on other sites

4 minutes ago, And Its Smith said:

It just proves it’s hugely flawed. We didn’t deserve to draw today.  The model doesn’t work.  I’ve listened to what people have said and am still convinced it doesn’t work as a model.  

It's not saying we deserved to draw. It's saying that on balance our chances to score were as good as theirs which I'd say is probably fair. There are of course many, many other factors such as who's taking the shot - which is why it doesn't claim (and nobody I've ever seen claim it) to be perfect.

For example it's in fact hugely warped in this single instance due to the penalty, which it rightly estimates has a high chance of being scored - it's an exceptionally good chance to score a goal all things considered. Unfortunately we didn't, but if you're going to build a model on EXPECTED goals, then a penalty would fall in the "expected" category - or at least, more in that one than any other. And that's what happened.

For that reason alone, using it as a one-off indication of anything is inherently flawed.

Link to comment
Share on other sites

2 minutes ago, IAmNick said:

It's not saying we deserved to draw. It's saying that on balance our chances to score were as good as theirs which I'd say is probably fair. There are of course many, many other factors such as who's taking the shot - which is why it doesn't claim (and nobody I've ever seen claim it) to be perfect.

For example it's in fact hugely warped in this single instance due to the penalty, which it rightly estimates has a high chance of being scored - it's an exceptionally good chance to score a goal all things considered. Unfortunately we didn't, but if you're going to build a model on EXPECTED goals, then a penalty would fall in the "expected" category - or at least, more in that one than any other. And that's what happened.

For that reason alone, using it as a one-off indication of anything is inherently flawed.

And yet I see xG league tables!  People comment on xG as if it’s the true picture. What a waste of time. 

Link to comment
Share on other sites

5 minutes ago, And Its Smith said:

And yet I see xG league tables!

ignore them, scroll on by

People comment on xG as if it’s the true picture.

ignore them, scroll on by

What a waste of time.

ignore them, do something else with your time

⬆️⬆️⬆️ The model isn’t flawed in its concept (it won’t ever be perfect), the model is being misused.

Link to comment
Share on other sites

26 minutes ago, And Its Smith said:

And yet I see xG league tables!  People comment on xG as if it’s the true picture. What a waste of time. 

Sure. But people are also surprised when Leicester win the league, we beat Man U in the cup, a 25yr old triathlete dies of a heart attack, or that they flip a coin and it comes up heads 5 times in a row. Models would predict against all those things.

Does that mean the underlying models are nonsense though? Or that perhaps they're being used wrong - or are just that, models?

If your thoughts are binary - perfect, or flawed, then a hell of a lot of models including a load you rely on for your day to day life are "flawed".

Link to comment
Share on other sites

@Davefevs 8 for Bentley?

Thought he should have done better for the first, flapped at at least one cross, and spilled the ball 3 mins later as well.

Seeing as he is very much a player who has demons in his head - could be a rough time ahead for bents.

Great stopper 90% of the time - very very good, but he doesn't command his box, and is quite the flapper at times.

 

2p

Link to comment
Share on other sites

30 minutes ago, The Constant Rabbit said:

@Davefevs 8 for Bentley?

Thought he should have done better for the first, flapped at at least one cross, and spilled the ball 3 mins later as well.

Seeing as he is very much a player who has demons in his head - could be a rough time ahead for bents.

Great stopper 90% of the time - very very good, but he doesn't command his box, and is quite the flapper at times.

 

2p

2p’s are absolutely fine.

Thought Pukki’s touch and early shot was top class for his first goal.  Made several saves.  That’s my view of today.

I didn’t see any demons today....just exposed by crap defending.

Link to comment
Share on other sites

The big issue was our decision making at the back. When there is no pressure on the opposing midfield (which was the case 90% of the game) you have to drop deep, otherwise at this level an opponent with time on the ball will find the attacker too often. See goal one and two. The third was the opposite, pressure on the ball and Kalas dropped on his own allowing there lad to run in. If Kalas stepped up it was offside/over hit pass. Naivety that will be punished at this level by all the top sides. 

Link to comment
Share on other sites

1 hour ago, IAmNick said:

Sure. But people are also surprised when Leicester win the league, we beat Man U in the cup, a 25yr old triathlete dies of a heart attack, or that they flip a coin and it comes up heads 5 times in a row. Models would predict against all those things.

Does that mean the underlying models are nonsense though? Or that perhaps they're being used wrong - or are just that, models?

If your thoughts are binary - perfect, or flawed, then a hell of a lot of models including a load you rely on for your day to day life are "flawed".

I know nothing about this XG thing or about any of the modern day statistics. 

Does it take in to consideration on who has the chances to score such as Junior Bent through on goal with a near zero chance of converting, compared to Bob Taylor in the same circumstance? 

Link to comment
Share on other sites

11 minutes ago, stephenkibby. said:

I know nothing about this XG thing or about any of the modern day statistics. 

Does it take in to consideration on who has the chances to score such as Junior Bent through on goal with a near zero chance of converting, compared to Bob Taylor in the same circumstance? 

Nope, hence why it's just one part of a whole range of things you should look at.

I believe what it does is total up all shots of each type from each position, and work out how often they are scored. From that it tries to predict how many goals a team should score each game.

For example, it will see where a team had a shot from, look at the hundreds of shots taken from that position, angle etc. and see how often they get scored. That's not just that team, but all teams (obviously not the Downs league, but you get the general idea). Based off that it will give it a number, where 1 is a certain goal, and 0 is no chance. A penalty is about a .7 I think (so 7/10 are scored).

It does that for all the shots in a game, and gives you a number of how many "expected" goals your team would score. I think the game today was about 1.6 for us, 1.8 for Norwich.

Obviously if you have Bas Savage or Harry Kane up front, who's in goal, where the sun is, what everyone had for breakfast and so on that's going to make a hell of a difference- hence why it's just a model, and should be used either over a large number of games or in conjunction with other stuff (including your eyes!!)

Link to comment
Share on other sites

2 minutes ago, IAmNick said:

Nope, hence why it's just one part of a whole range of things you should look at.

I believe what it does is total up all shots of each type from each position, and work out how often they are scored. From that it tries to predict how many goals a team should score each game.

For example, it will see where a team had a shot from, look at the hundreds of shots taken from that position, angle etc. and see how often they get scored. Based off that it will give it a number, where 1 is a certain goal, and 0 is no chance. A penalty is about a .7 I think (so 7/10 are scored).

It does that for all the shots in a game, and gives you a number of how many "expected" goals your team would score. I think the game today was about 1.6 for us, 1.8 for Norwich.

Obviously if you have Bas Savage or Harry Kane up front, who's in goal, where the sun is, what everyone had for breakfast and so on that's going to make a hell of a difference- hence why it's just a model, and should be used either over a large number of games or in conjunction with other stuff (including your eyes!!)

Some of the stats companies are building more sophistication into it, e.g. height of ball, position of keeper etc, etc.  But as you increase the parameters you decrease the number in each sample.  For example if you had 1000 shots from the penalty spot (not a penalty though) it might be 500 goals, xG 0.5.  But that might be 300 with feet, 200 with head.  But those might’ve come from 400 shots and 600 headers....meaning a shot is 0.75 and a header 0.33.

You can then keep going and going to lower levels of granularity.

Link to comment
Share on other sites

1 minute ago, Davefevs said:

Some of the stats companies are building more sophistication into it, e.g. height of ball, position of keeper etc, etc.  But as you increase the parameters you decrease the number in each sample.  For example if you had 1000 shots from the penalty spot (not a penalty though) it might be 500 goals, xG 0.5.  But that might be 300 with feet, 200 with head.  But those might’ve come from 400 shots and 600 headers....meaning a shot is 0.75 and a header 0.33.

You can then keep going and going to lower levels of granularity.

Ah that's interesting, I didn't know how far it had been taken currently.

Realistically each shot will only ever happen exactly once, ever. There will only be one perfect occurrence of events which led to it to ever happen, so the "true" sample for each is exactly 1. That's obviously not so useful so we have to generalise, and to start off I assume we just generalised to the extreme  - to just take a "shot" from that position with zero other influences. Once you've got that working, you can start to build the rest in.

As you've said though, I would imagine even introducing some very simple parameters (head/foot, incoming ball speed or direction, keeper position) will HUGELY reduce the sample size - and probably quite often reduce it to such a point that the result is no longer statistically significant at all! Head/foot might cut your entire sample size in half, and that's just a binary parameter. Once you add in something with a range of values like a player position or ball speed you're running the risk of making predictions based off almost nothing, so the generalised stuff ends up being more accurate!

Maybe we need to just quantify each players attributes more effectively, and run every situation a billion times in a simulation and plug that in - I expect someone (and not just Sports Interactive who make Football Manager) is working on it already! Then we might not need the real data at all. And then we might not even need to play the games at all ;) 

Link to comment
Share on other sites

3 minutes ago, IAmNick said:

Ah that's interesting, I didn't know how far it had been taken currently.

Realistically each shot will only ever happen exactly once, ever. There will only be one perfect occurrence of events which led to it to ever happen, so the "true" sample for each is exactly 1. That's obviously not so useful so we have to generalise, and to start off I assume we just generalised to the extreme  - to just take a "shot" from that position with zero other influences. Once you've got that working, you can start to build the rest in.

As you've said though, I would imagine even introducing some very simple parameters (head/foot, incoming ball speed or direction, keeper position) will HUGELY reduce the sample size - and probably quite often reduce it to such a point that the result is no longer statistically significant at all! Head/foot might cut your entire sample size in half, and that's just a binary parameter. Once you add in something with a range of values like a player position or ball speed you're running the risk of making predictions based off almost nothing, so the generalised stuff ends up being more accurate!

Maybe we need to just quantify each players attributes more effectively, and run every situation a billion times in a simulation and plug that in - I expect someone (and not just Sports Interactive who make Football Manager) is working on it already! Then we might not need the real data at all. And then we might not even need to play the games at all ;) 

Yep, exactly that.

There was a classic xG chance the other night in a European game, where the keeper had come charging outside his box missed the ball, but as the player slotted home into an open goal from about 15 yards on a slight angle it was something like 0.08 xG, because THAT model was very basic, e.g. shot position only.

Someone jokingly said:

need to add a 1 or 0 into my dataset for whether the keeper is still in goal.

Even that sounds trivial, but every parameter starts to increase the complexity, and as you say the sample size.

Wyscout measure a penalty at 0.76.

Link to comment
Share on other sites

1 minute ago, Davefevs said:

Yep, exactly that.

There was a classic xG chance the other night in a European game, where the keeper had come charging outside his box missed the ball, but as the player slotted home into an open goal from about 15 yards on a slight angle it was something like 0.08 xG, because THAT model was very basic, e.g. shot position only.

Someone jokingly said:

need to add a 1 or 0 into my dataset for whether the keeper is still in goal.

Even that sounds trivial, but every parameter starts to increase the complexity, and as you say the sample size.

Wyscout measure a penalty at 0.76.

Hah, yeah that's a perfect example.

It's why stats are great though right? It's not the stats themselves - anyone can look them up and copy/paste some table of numbers. There's no skill or interest in that for me. That's pretty much just what a low level commentator without any actual knowledge does to fill air time.

The skill is in picking out which ones are relevant, and then how you interpret them to tell a story.

Link to comment
Share on other sites

17 minutes ago, IAmNick said:

Hah, yeah that's a perfect example.

It's why stats are great though right? It's not the stats themselves - anyone can look them up and copy/paste some table of numbers. There's no skill or interest in that for me. That's pretty much just what a low level commentator without any actual knowledge does to fill air time.

The skill is in picking out which ones are relevant, and then how you interpret them to tell a story.

I guess you’ve read Moneyball?  Don’t bother with the film, but the book is great. On Base Percentage!!!

Link to comment
Share on other sites

15 hours ago, Davefevs said:

Of course it’s harsh, nobody is really a 0, but as a collective three I thought they were as bad as I’ve ever seen from a city defence 

I didn't think Kalas was that bad, probably deserved a point or two more, Moore out of position a lot and Vyner tends to switch off, at times decisions are poor. Just watching some of the highlights and it kind of makes a point I made elsewhere. The first goal, not sure what Vyner is doing, he's not marking or covering Puki, but there is no pressure on the ball. Not only that, Stiepermann is given so much time he can study what pass he wants to make, very good pass just the same.

Below Stiepermann has just got the ball, lines aren't great.

 

1663489290_Screenshot2020-11-01at10_22_51.png.b0a6f796fd1fd0358c1b75d03154c59b.png

3 CB's and neither striker or runner covered , well played pass though. Hunt no one to mark maybe should have gone with Puki?

1906568607_Screenshot2020-11-01at10_23_14.png.439de78756e6ab8d3fef1415fa8696b2.png

3 CB's and only one back defending and it's 14 minutes in !! Another good ball , but no-one within 10 yards of the ball.

Kalas deepest defender, the other 2 too far up giving him no support. If TM has gone forward Vyner has to stay surely.

1418197409_Screenshot2020-11-01at10_28_04.png.e5e2b010f98c4ab95a72af335d0244f5.png

We didn't learn, although Norwich did. This time some blame for Kalas, too deep? Plays Buendia on side no other cover within 15-20 yards.
Simple straight ball , Vyner actually looks he's checking the line here. No one near the ball player again though.

710079111_Screenshot2020-11-01at10_33_53.png.ff8ed9ab92d963dd4e874343c09a4925.png

I wonder if we played 2 CB's they would be closer together and there would be a better line held? 

I always think if I notice something it has to be fairly obvious. I wouldn't be surprised to see us switch to 2 CB's for Tuesday, I also have a horrible feeling that Holden will be stubborn and stick with the shape, system AND Pato. 

Hey Ho, bloody nose, move on, go again , blah, blah blah 

 

Link to comment
Share on other sites

5 minutes ago, 1960maaan said:

I didn't think Kalas was that bad, probably deserved a point or two more, Moore out of position a lot and Vyner tends to switch off, at times decisions are poor. Just watching some of the highlights and it kind of makes a point I made elsewhere. The first goal, not sure what Vyner is doing, he's not marking or covering Puki, but there is no pressure on the ball. Not only that, Stiepermann is given so much time he can study what pass he wants to make, very good pass just the same.

Below Stiepermann has just got the ball, lines aren't great.

 

1663489290_Screenshot2020-11-01at10_22_51.png.b0a6f796fd1fd0358c1b75d03154c59b.png

3 CB's and neither striker or runner covered , well played pass though. Hunt no one to mark maybe should have gone with Puki?

1906568607_Screenshot2020-11-01at10_23_14.png.439de78756e6ab8d3fef1415fa8696b2.png

3 CB's and only one back defending and it's 14 minutes in !! Another good ball , but no-one within 10 yards of the ball.

Kalas deepest defender, the other 2 too far up giving him no support. If TM has gone forward Vyner has to stay surely.

1418197409_Screenshot2020-11-01at10_28_04.png.e5e2b010f98c4ab95a72af335d0244f5.png

We didn't learn, although Norwich did. This time some blame for Kalas, too deep? Plays Buendia on side no other cover within 15-20 yards.
Simple straight ball , Vyner actually looks he's checking the line here. No one near the ball player again though.

710079111_Screenshot2020-11-01at10_33_53.png.ff8ed9ab92d963dd4e874343c09a4925.png

I wonder if we played 2 CB's they would be closer together and there would be a better line held? 

I always think if I notice something it has to be fairly obvious. I wouldn't be surprised to see us switch to 2 CB's for Tuesday, I also have a horrible feeling that Holden will be stubborn and stick with the shape, system AND Pato. 

Hey Ho, bloody nose, move on, go again , blah, blah blah 

 

Those stills are horrendous defending at any level

The third one is about as bad as it gets

Where the ***** is Taylor Moore going or been ?

and as a back line of any combination being so high up the pitch and so close to the ball without pressure on the ball and acres behind is simply begging to be cut open - Truly Sunday Morning basics

The size of the channels for the run and pass mean that most of us could have picked a through ball

Our only hope was that he was spoilt for choice and hesitated too long !

Thats before you take into consideration the three centre backs shape and playing to three different ideas

 

Link to comment
Share on other sites

19 minutes ago, Sheltons Army said:

Those stills are horrendous defending at any level

The third one is about as bad as it gets

Where the ***** is Taylor Moore going or been ?

and as a back line of any combination being so high up the pitch and so close to the ball without pressure on the ball and acres behind is simply begging to be cut open - Truly Sunday Morning basics

The size of the channels for the run and pass mean that most of us could have picked a through ball

Our only hope was that he was spoilt for choice and hesitated too long !

Thats before you take into consideration the three centre backs shape and playing to three different ideas

 

Going 1-0 so early against a good side, regroup stay solid/tight for a while. Nope gunging ho ! I guess you could say Kalas was too deep, but do you play a high line against such good movement and a pacy striker? One word, shambolic !

Link to comment
Share on other sites

1 hour ago, 1960maaan said:

I didn't think Kalas was that bad, probably deserved a point or two more, Moore out of position a lot and Vyner tends to switch off, at times decisions are poor. Just watching some of the highlights and it kind of makes a point I made elsewhere. The first goal, not sure what Vyner is doing, he's not marking or covering Puki, but there is no pressure on the ball. Not only that, Stiepermann is given so much time he can study what pass he wants to make, very good pass just the same.

Below Stiepermann has just got the ball, lines aren't great.

 

1663489290_Screenshot2020-11-01at10_22_51.png.b0a6f796fd1fd0358c1b75d03154c59b.png

3 CB's and neither striker or runner covered , well played pass though. Hunt no one to mark maybe should have gone with Puki?

1906568607_Screenshot2020-11-01at10_23_14.png.439de78756e6ab8d3fef1415fa8696b2.png

3 CB's and only one back defending and it's 14 minutes in !! Another good ball , but no-one within 10 yards of the ball.

Kalas deepest defender, the other 2 too far up giving him no support. If TM has gone forward Vyner has to stay surely.

1418197409_Screenshot2020-11-01at10_28_04.png.e5e2b010f98c4ab95a72af335d0244f5.png

We didn't learn, although Norwich did. This time some blame for Kalas, too deep? Plays Buendia on side no other cover within 15-20 yards.
Simple straight ball , Vyner actually looks he's checking the line here. No one near the ball player again though.

710079111_Screenshot2020-11-01at10_33_53.png.ff8ed9ab92d963dd4e874343c09a4925.png

I wonder if we played 2 CB's they would be closer together and there would be a better line held? 

I always think if I notice something it has to be fairly obvious. I wouldn't be surprised to see us switch to 2 CB's for Tuesday, I also have a horrible feeling that Holden will be stubborn and stick with the shape, system AND Pato. 

Hey Ho, bloody nose, move on, go again , blah, blah blah 

 

Great pics....essentially the 3CBs between them aren’t doing their basic jobs correctly.  They don’t mark, they are marking space properly.

In some respects I feel a tad sorry for Vyner because Pukki’s first touch and early shot was top drawer, but he was too far away and got caught watching the passer not Pukki.  In that position he’s got to really keep quick-checking man and passer....he can’t focus on passer for anything more than a split second.

10D673B0-33EF-4916-9BAA-9C48E9D38968.thumb.jpeg.8f1ecfc6af7db7804b12bb2495a52e72.jpeg

Too far away ⬆️⬆️⬆️

There was an occasion where Norwich attacked down their LHS, I watched Moore....he continually focussed on the ball.  He never checked where his man was, who just drifted off the back of him.  I was quite shocked that he never twisted his head to check his man.

In one of the training videos I recall Downing saying something to Mawson about how he was using his head to observe where his man was going whilst also tracking the crosser.  Think he stopped Martin getting across him.  That’s the difference at this level.

I also want to know why Moore was so far up the pitch of the second goal?

Heres another, the Placheta chance that Bents saved with his head.

4189F67B-042B-44C7-AA04-9369AF02EF41.thumb.jpeg.99e64f57aacb8faad39c7dda4e645bf9.jpeg

Awful line and shape / marking

AEFEE713-0B00-48F9-B3D6-BD344B2CC89B.thumb.jpeg.e08b22e9577e5e237d40b58f5425d54b.jpeg

Still shocking! ⬆️⬆️⬆️
C0B69836-12F8-4FF5-8369-0A2FAB521D6E.thumb.jpeg.5e4edace822b3be64e8a7d0b91317c76.jpeg

7 onto 4 but only Kalas reacts ⬆️⬆️⬆️.

4F4E6FBC-EC28-4A6A-A74E-C242C3EB8BB6.thumb.jpeg.b7ca0e4df04418d2867895ea0d39cdd6.jpeg

Kalas and Vyner out of picture, Brunt doing nothing, O’Dowda doing nothing, Pato walking.

828B4B3E-46C4-4B78-81EF-83358433ACD0.thumb.jpeg.ba718c191d50b1eff95459e2f0c58f8e.jpeg

Vyner now in shot, Kalas deep...but none of the three marking anywhere near close enough.

7D6D508A-0AA1-4DD9-B530-8F24B39D9824.thumb.jpeg.3cd58f20963e8305ba42157d9695ae73.jpeg

It just gets worse.

Link to comment
Share on other sites

3 minutes ago, And Its Smith said:

Slightly jealous of how much time people have to dissect a football match!  

I will be too from next week.  Should’ve been starting new job tomorrow (7 months without work for mainly shielding reasons), but references haven’t been processed.  The fact that I worked for this company for almost 28 years and told them in their forms upon application seems to have slipped through their net.  The fact that the hiring manager is also my referee seems to have been ignored too.  So might have an extra day or two at home....but unlikely to have time to do a Huddersfield thread this week.

Link to comment
Share on other sites

11 minutes ago, And Its Smith said:

Slightly jealous of how much time people have to dissect a football match!  

Made redundant end of August, isolating as my Wife is waiting on an operation (minor) so I'm not looking for work until she's had it, shouting at City being shit, and dissecting their shitness is keeping me sane, that and the housework ?

Link to comment
Share on other sites

26 minutes ago, Davefevs said:

Just to add I’ve now watched the second goal again.  Moore tracks Buendia into Norwich’s half who lays it wide to Sorensen.  Moore does try to get back in fairness.

 

Now I never played at any level, and only played CB a couple of times ( 5'8" explains that ? ) and never in a 3 but, surely if TM goes, Vyner should drop in alongside Kalas. I remember an Arsenal video when the defence talked of George Graham . They said they were made to act as though they were tied together in a chain. LB goes and that pulls LCB, RCB and RFB over to cover , and vice versa.  I know they're only stills but I hope Holden has a good hard look because positionally we look awful.

Link to comment
Share on other sites

24 minutes ago, 1960maaan said:

Now I never played at any level, and only played CB a couple of times ( 5'8" explains that ? ) and never in a 3 but, surely if TM goes, Vyner should drop in alongside Kalas. I remember an Arsenal video when the defence talked of George Graham . They said they were made to act as though they were tied together in a chain. LB goes and that pulls LCB, RCB and RFB over to cover , and vice versa.  I know they're only stills but I hope Holden has a good hard look because positionally we look awful.

I remember that, or at least another version.  Adams and Bould / Keown were also told, never be deeper than your full-backs.  The full-backs set the offside line as they could see across the pitch.  Dixon and Winterburn would often be slightly deeper and in the cover....as they believed a straight ball down the middle was harder to achieve, and therefore if they had to defend an angled ball, they could cover, as they’d given themselves a yard.

I played a lot as the RCB in a three....and that was at the highest level I played.  We had simple rules.  Mark your man.  Usually two strikers so easy.  The “Sweeper” always stayed spare.  We only ever passed our markers over if they crossed over the man we were marking.  Saved us having go totally man-to-man.  Requires good comms, but that seemed a natural thing to do.  We talked each other through our opposite man’s opponents movement, with simple shouts of “left shoulder”, “get touch tight” etc.  Both our jobs were to attack the ball in the air, if that meant leaving our marker or trying to get in front, that’s when the spare CB’s job would be to cover us if we lost it.

How would we have coped with one up front?  One of us would’ve pushed into midfield or full-back.  If midfield then just made another body in there.  If I went to full-back, then the RWB pushed on.

If I stayed as the CB, then I’d still predominantly attack the ball in the air, but the previously spare CB would play left and I’d play right of each other....but I’d still mark as much as possible unless it genuinely went into my partners channel.

That was the theory.  I always preferred marking than dropping off.  Kalas should’ve sorted the 3v1 on the pitch, the coaching staff should’ve seen that too.

Link to comment
Share on other sites

13 minutes ago, Davefevs said:

Kalas should’ve sorted the 3v1 on the pitch

That's the thing that was disappointing in his game. He is the experience in that 3, if say you forget the WB's as we want them high up the pitch the other two should take their position from Kalas, but if one goes the other drops to always have 2 CB's.  Unbelievable to have both L & R CB pushing on and leaving Kalas so deep.  It looked like they'ed never played a 3 before. I give credit to Norwich , nice crisp passing good movement throughout the game , but being exposed by a simple on ball again is schoolboy.  It is incredible that 3 CB's , with one player to mark lost him so regularly.  

Looking at the highlights , so many times players walking around. All that improved movement and showing for the ball was obviously the exception rather than the rule. Sadly looking more like Johnson's City all over again..

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...