Topic: Andy Robertson (Andrew Henry Robertson MBE)

[jooneyisdagod]

I really like his crossing ability. Not sure the rest of his play is quite up there yet but all of that is almost masked by how good his crossing was in the game. I still think he needs to adapt to our game more, which means fewer early crosses and more interplay, through passes etc. Very promising debut though and it'll hopefully give Moreno the kind of competition that he needs to kick up his game a couple of notches too.

[Oside Red]

<a href="https://www.youtube.com/v/fFeg_hl0Se4" target="_blank" rel="noopener noreferrer" class="bbc_link bbc_flash_disabled new_win">https://www.youtube.com/v/fFeg_hl0Se4</a>

  still laugh every time.

PUSKAS AWARD
all the years

[BabuYagu]

Isn't that bad? That means his attacking output is lesser than Moreno and Milner rite?

It means that playing for a shite Hull side, he rarely produced anything of note. People who watched him for Dundee waxed lyrical about him in an attacking setup. Not that he is playing for an attacking side, we are seeing the sort of things Royhendo saw and spoke of that many others doubted he had. There is no way he will keep those numbers up of course, most good Premier League wingers output is something like 0.2 xA per game. But it is a very good indicator we have the right full back for our system.

I've done a roundtable going into more details which should go live tomorrow/Monday.

[BEAST]

Baba - I'm curious which fullback had the highest xA last year of say the bottom 5-7 sides?  Anyone do anything out of the ordinary?

But you're right ..... "attacking fullback" doesn't really exist if your team is being dominated week in week out, very few opportunities to been in an attacking position.

[joekim87]

Looked good. Great crossing and wasn't noticeable on defense, which is a good thing because our defenders usually make the highlight reel making howlers.

[BabuYagu]

Baba - I'm curious which fullback had the highest xA last year of say the bottom 5-7 sides?  Anyone do anything out of the ordinary?

But you're right ..... "attacking fullback" doesn't really exist if your team is being dominated week in week out, very few opportunities to been in an attacking position.

It's on my to-do list for the full backs thread when I get back to continuing that. I believe Trippier & Bellarin were the two highest per 90 when I looked at the end of the season. I believe Bertrand and Soares both came out quite well. Masuaku from West Ham also - although he played less than 1000 minutes of football.

The thing is though - for a full back an xA per 90 of 0,1 is considered pretty good for Premier League levels. Robertson could do literally nothing now for the next 10 games and still come out near the top of the list

[BEAST]

Can't wait till the next time a team comes to Anfield and parks 10 men behind the ball and Robertson just spends 90 minutes firing balls into the 6 yard box and punishes them for it ..... have some of that 

[harleydanger]

Out of the three we've signed so far I get the feeling Robertson will be the most important, purely for what he'll do to low block sides

[oakton]

Our new No.1 left back.

[RedMichelFerri]

As stated he has great cross on him, decent pace and very comfortable on ball. But what impressed me most is the way he plays 1-2 and the positions he takes which shows his football intelligence. He also knows when to hug baseline and when you move inside. So many times he ghosted in the box.... I am really excited about this player.

[JovaJova]

That was a very impressive attacking performance from him. It's amazing how strong our full back options now are (in an attacking sense). On any normal day his crosses result in a minimum of two goals. I'm a big Moreno fan as well (somebody has to be) so looking forward to the competition this season between them, and also Clyne and TAA on the other side. Another major positive for a squad that is way stronger than people outside the club think.

[The Final Third]

I really like his crossing ability. Not sure the rest of his play is quite up there yet but all of that is almost masked by how good his crossing was in the game. I still think he needs to adapt to our game more, which means fewer early crosses and more interplay, through passes etc. Very promising debut though and it'll hopefully give Moreno the kind of competition that he needs to kick up his game a couple of notches too.

I see it differently. His early crossing is what we've lacked, especially against low block teams where it catches the defence out before it sets, and it is our forwards' role now to take advantage of it by making anticipatory early runs into the opposition's box. We did this in the second half yesterday and looked much more of a threat doing it.

[Ravishing Rick Dude]

Let's not make a cliché about his crosses, because in the future, everyone will try to stop them.

[FLRed67]

Let's not make a cliché about his crosses, because in the future, everyone will try to stop them.

I very much doubt what he did yesterday has gone unnoticed by opposition EPL managers and scouts.

If they try to stop his crosses, that simply creates more space for Firmino, Lallana, and Mane to run through.

Robertson looks to be the weapon we needed.

[Ravishing Rick Dude]

I very much doubt what he did yesterday has gone unnoticed by opposition EPL managers and scouts.

If they try to stop his crosses, that simply creates more space for Firmino, Lallana, and Mane to run through.

Robertson looks to be the weapon we needed.

Ah fuck it, i don't even think that they can stop him - especially early crosses it's very hard to stop.

[Doc Red]

Very impressive.
Once he got over the initial butterflies, showed some neat , precise, touches and passes.
Was impressed by his movement and awareness, he seemed on the same wavelength as the likes of Mane ane Firmino, which bodes well for the future. I was pleasantly surprised,  thought it might take him awhile to get into the flow of things offensively.
Should be our starting left back by the time we've played our 10th game of the season.

[clinical]

Was brilliant yesterday. Will be tested more defensively but going forward, wow.

[Pistolero]

Good interview with Robbo on the LFC site

“It was great. When I got told I was getting the nod and then you start thinking about playing at Anfield and in front of the fans - you can’t quite imagine it until you’re out there but it was different class,” the Scotland international said.

“At times they were frustrated but so were we and it was important we stuck together, which we did.

“This man [Mane] popped up with a goal when we needed him most and then I thought we were quite comfortable after that with the clean sheet and keeping the ball. That’s what we can do and luckily we managed to see it out and get the three points and a clean sheet.”

“As soon as we saw their result last weekend, we knew we were going to get a reaction with the new manager,” Robertson explained, in reference to Palace’s 3-0 opening-day defeat to Huddersfield Town.

“They’d have been working at it all week, maybe sitting in deep and going a bit more direct when you’ve got a threat like Benteke, but I thought Raggy and Joel dealt with it brilliantly and defensively I thought we were sound. That’s what we needed.”

“It’s brilliant when you get the ball and you look up and you’ve got Sadio, Roberto, Studge in the first half and Mo came on with big Dom - it’s great to have the options,” Robertson noted.

“You know they’re going to make good runs and you know when they get a chance they should put it away and Sadio’s in great form just now, everything he hits seems to go in, so hopefully that continues and everyone else can start pitching in with goals.

“We’re not fully up to speed yet, but we’re getting there and when we are we’ll be a threat.”

[Zee_26]

Wonder if we could see him and TAA on corner duties when the two of them do play together. If so, we might finally have some kind of consistent threat at corners.

[GIPPO77]

Great debut, looks like a very good addition.

[Tony19:6]

Very impressive debut going forward, wasnt tested much defensively but that will come.

His crossing was the highlight for me with some lovely early ones, something weve lacked.

Keep up the good work.

[Flinstone]

It means that playing for a shite Hull side, he rarely produced anything of note. People who watched him for Dundee waxed lyrical about him in an attacking setup. Not that he is playing for an attacking side, we are seeing the sort of things Royhendo saw and spoke of that many others doubted he had. There is no way he will keep those numbers up of course, most good Premier League wingers output is something like 0.2 xA per game. But it is a very good indicator we have the right full back for our system.

I've done a roundtable going into more details which should go live tomorrow/Monday.

How can Morenos xA be 4.23 with a sample size of a whole season and a good premier league winger be 0.2  ?

[hesbighesred]

Interesting stat on MOTD about Robertson having more touches than anyone else - fits in with what Royhendo was saying about his ability to run games from LB when he was at Dundee.

[Purple Red]

How can Morenos xA be 4.23 with a sample size of a whole season and a good premier league winger be 0.2  ?

Because stats.

[Szemerényi]

Probably the cumulated values of his expected assists per game over a premier league season? If Moreno's expected number of assists over a season is 4.23 then we get approximately an xa per game of 0.1 which is what Babuyagu said was a decent rate for a fullback.

[[new username under construction]]

Didn't think he was tested a large amount defensively so will have to wait and see, looks good though and seems to be a lot more/better communication with whomever is next to him

[Smellytrabs]

Great debut from the lad. Very promising.

[BabuYagu]

It means that playing for a shite Hull side, he rarely produced anything of note. People who watched him for Dundee waxed lyrical about him in an attacking setup. Not that he is playing for an attacking side, we are seeing the sort of things Royhendo saw and spoke of that many others doubted he had. There is no way he will keep those numbers up of course, most good Premier League wingers output is something like 0.2 xA per game. But it is a very good indicator we have the right full back for our system.

I've done a roundtable going into more details which should go live tomorrow/Monday.

How can Morenos xA be 4.23 with a sample size of a whole season and a good premier league winger be 0.2  ?

Moreno's is a total. The other is per game. The per/90 amounts would be as follows:-

Player Name	Season	xA per 90
Robertson	2017/18	1,21
Robertson	2016/17	0,05
Robertson	2014/15	0,06
Moreno	2015/16	0,15
Milner (all)	2016/17	0,11
Milner (open play only)	2016/17	0,06
Coutinho	2016/17	0,26
Mahrez	2016/17	0,22
Sterling	2016/17	0,18
Bellarin*	2016/17	0,13
D.Gray	2016/17	0,13
Zaha	2016/17	0,13
Adama Traore	2016/17	0,10

* Best full back with over 1000 minutes played.

[Rush 82]

I very much doubt what he did yesterday has gone unnoticed by opposition EPL managers and scouts.

If they try to stop his crosses, that simply creates more space for Firmino, Lallana, and Mane to run through.

Robertson looks to be the weapon we needed.

Yep.

They had us worked out down the left and right. Nothing to worry about.

Now we've suddenly got genuine threats in Robertson and TAA and the mid ain't getting easier with Salah having been added.

Happy days - back to the drawing board you fuckers

[Yanwoo]

But it is a very good indicator we have the right full back for our system.

You mean the right left full back... 

[Flinstone]

Because stats.

And yet the answer makes perfect sense

Moreno's is a total. The other is per game. The per/90 amounts would be as follows:-

Player Name	Season	xA per 90
Robertson	2017/18	1,21
Robertson	2016/17	0,05
Robertson	2014/15	0,06
Moreno	2015/16	0,15
Milner (all)	2016/17	0,11
Milner (open play only)	2016/17	0,06
Coutinho	2016/17	0,26
Mahrez	2016/17	0,22
Sterling	2016/17	0,18
Bellarin*	2016/17	0,13
D.Gray	2016/17	0,13
Zaha	2016/17	0,13
Adama Traore	2016/17	0,10

* Best full back with over 1000 minutes played.

Gotcha, interesting stuff. Is there an xA per team collated? I supposed you could apply the difference b/w Hull and Liverpool and add that to Robertsons numbers to see how it might look at the end of the season.

Bit elementary but still.

[BabuYagu]

I see it differently. His early crossing is what we've lacked, especially against low block teams where it catches the defence out before it sets, and it is our forwards' role now to take advantage of it by making anticipatory early runs into the opposition's box. We did this in the second half yesterday and looked much more of a threat doing it.

I agree. The important thing is to mix it up.

Talking purely about low-block defences, tactically, if a player tends towards something then you can have a tactical plan to nullify it. Take Milner - at left back he tends to cut onto his right. Therefore you can sit narrow. Therefore compressing the center of the pitch more. The risk is that he could not cut onto his right and burst past you. But then he is 31 and no long has that burst of pace he once had, so it's probably not going to happen. Or he could hit a first time ball with his left. But those tend to lack power, direction and be floaty in nature so heavily favour the defence.

There in terms of risk v reward - starting narrow against him means you give the central players less space to operate in. Starting closer to him gives the central players more space to operate in. The trade off in risk from Milner means you can just sit narrow is he is likely to need several touches and will also likely come inside on his right anyway.

Robertson (& Moreno) though threatens first time crosses, getting in behind, attacking the space between full back and center back and looks for combination plays. Moreno also has a powerful shot and therefore defenders want to close that down when he comes inside also.

Milner's attacking threats involve going in front or down the side of the block. Nothing destabilises it, moves the central players across to him, commits players, etc. Whereas if you get in behind a full back, or commit him, the nearest CB or DM will come across. This creates space centrally. If you get between the CB & FB then again central players will shift towards you, especially if you can take on and commit/beat that CB. If you get into the half space and have a dangerous shot, the CB or DM needs to press out to you, this creates space again. Having full backs that can destabilise the block is one of the important tools to breaking it down. Glen Johnson was very good at this for Rafa's Liverpool as he had so many different threats going on. Milner doesn't really do this at LB at all which is why I doubt we will see him there for games against such sides this season.

[BabuYagu]

And yet the answer makes perfect sense


Gotcha, interesting stuff. Is there an xA per team collated? I supposed you could apply the difference b/w Hull and Liverpool and add that to Robertsons numbers to see how it might look at the end of the season.

Bit elementary but still.

Those xA numbers though would include the fact that far more creative players like Coutinho & Lallana are here earning them so it wouldn't really tell you much. You could adjust them for possession or touches in the final 1/6th maybe. e.g. If Liverpool have 200 touches in the final 6th (20-25 yards from goal) and Hull only have 40, then Liverpool players have 5 times more chance of playing an attacking pass. I don't think I have that data though. Something for me to speak to the tech heads that help me with this and see if we could come up with something.

Can you see from the number btw why I am not too keen on Zaha & Gray. Gray's underlying xA was really low, but with a high conversation rate. e.g. - if Milner cuts the ball back to Coutinho who hammers it in from 30 yards, that is 1 assists, but xA of 0.01. Whereas Robertson yesterday got 0 assists but xA of 1.21. A very overly simplistic example of course but you get the idea why I like to look at xA numbers rather than assists or chances created. It tells you how good those chances were and ignores if you were simply lucky or unlucky with someone converting them.

Of course Zaha's pace and dribbling give him some useful tools to have but his production numbers last season could be something of a lie. His underlying numbers have been the same for 3 years in terms of xA and xG - it's just suddenly the conversation jumped up to unsustainable levels which is a red flag when looking at potential transfer targets.

[The Final Third]

I agree. The important thing is to mix it up.
Snip..

Right. Just changing the point of attack really loses it's effectiveness once a defence is set. We've tried this probing for a while, moving the ball around in front of a low block and often without any penetration, finally resorting to low percentage shots on goal from distance.

First-time delivery of the ball or passing it at speed with great movement that utilises the full width of the pitch combined with changing the point of attack will really test a defensive shape.

We have the pace in the front line now combined with wide players on their natural side along with inside forwards that are cutting in to comprehensively test even the most unyielding of defences.

[Flinstone]

Those xA numbers though would include the fact that far more creative players like Coutinho & Lallana are here earning them so it wouldn't really tell you much.

A workaround would be summing up the xA of the individual players adjusted per minute to give the xA for the entire team. Another option would be to isolate Liverpools defensive players and compare their numbers to the numbers they were posting when they were playing elsewhere to see if there has been a tangible jump due to system and if it could be applied to incoming players.

Who defines what these metrics are? Is the data simply coming from video feeds where there are people sitting and collating it after the match or is the whole thing automated? I wonder if Liverpool is collecting their own external data or simply buying it from the third party who is creating all this

[dramared]

Superb crossing, best we have seen from the left in a long time, always looks to pass forward and I like his football link up with the other players in the final third.
He was talking throughout the game, telling others and asking for the ball. His run where he pointed and held his run to stay onside was really nice.

I think this lad given time will be special against park the bus teams.

[Qston]

Made up for the lad. Very very solid debut and it will be tough for others to take his spot now. Some lovely balls put into the box, decent first touch and defensively pretty solid from what I saw. I sincerely hope he is a long term answer to what has been a problem position. With him and TAA at full back we could have those positions sewn up for many years to come.

[Chakan]

Good debut, funny how much better the crosses look when they don't have to shifted into a right foot.

[BabuYagu]

A workaround would be summing up the xA of the individual players adjusted per minute to give the xA for the entire team. Another option would be to isolate Liverpools defensive players and compare their numbers to the numbers they were posting when they were playing elsewhere to see if there has been a tangible jump due to system and if it could be applied to incoming players.

Who defines what these metrics are? Is the data simply coming from video feeds where there are people sitting and collating it after the match or is the whole thing automated? I wonder if Liverpool is collecting their own external data or simply buying it from the third party who is creating all this

One of the problem with any adjustments you make is that they are more likely to tell you nothing at all more than help shine a light on anything. For example, Liverpool under Klopp and Liverpool under Rodgers, statistically the teams look nothing alike. Likewise United under Mr Ferguson, Moyes, van Gaal, Mourinho. 1 team, 4 coaches, 4 different approaches. A defensive system from one coach can focus on tackles and interceptions - another can focus on blocking passes and recoveries. Likewise, a winger/full back can either be playing through sides or around sides, overlapping or holding position depending on his tactical instruction.

For example - Ted Knutsen, who knows far more about this than I ever will, adjusts players tackles and interception numbers for team possession. However, Spurs rank near the bottom in terms of both tackles and interceptions. Liverpool rank near the bottom in interceptions but midtable in terms of tackles. United and Arsenal rank near the top in both. The possession of the sides is pretty much the same. Which kinda indicates there is no correlation there at all - and that tackle & interception numbers are both a poor measurement in isolation in terms of how good a defender is, and also is more dependant on tactical instructions of the team than anything else. Whereas you look at something like blocked passes - something nobody really pays much attention to for rating players - and the highest ranking sides are the likes of Chelsea, Liverpool & Spurs (the sides who conceded the least shots in the league) and at the other end were the porous defences of Bournemouth, Stoke, Swansea and Sunderland.

Of course it's easier to look at attacking numbers than defensive ones but even then, for a full back, if they are a defensive full back who needs to maintain shape then they will never get into the positions Robertson was yesterday. If you watch the game back, watch how much freedom Robertson has to play his game with Milner filling in every time he went forward. A different game with say Can in front of him, you would see him go forward and Can just leaving a space there. The result would be he would either hurt us defensively or he would need to reign in those attacking instincts. So you have to look at Robertson & Milner as a combo yesterday - especially when you considered the top 2 biggest passing combinations in the game was Robertson to Milner and vice versa.

As for your other question, it's coming from Opta. No idea on the rest.

[Yanwoo]

A workaround would be summing up the xA of the individual players adjusted per minute to give the xA for the entire team. Another option would be to isolate Liverpools defensive players and compare their numbers to the numbers they were posting when they were playing elsewhere to see if there has been a tangible jump due to system and if it could be applied to incoming players.

Who defines what these metrics are? Is the data simply coming from video feeds where there are people sitting and collating it after the match or is the whole thing automated? I wonder if Liverpool is collecting their own external data or simply buying it from the third party who is creating all this

Here's a good article on how the data is collected:
https://fivethirtyeight.com/features/the-people-tracking-every-touch-pass-and-tackle-in-the-world-cup/