Jump to content

Recommended Posts

31 minutes ago, roversfan99 said:

Again, sceptical of the use of such a statistic. If a player takes on 2 players, gets to a third and loses it, does that come out as 66.67% successful, or 0%? How can you possibly define whether something is a dribble, does the close proximity of an opponent come into it? Impossible to objectify.

One dribble, regardless of the amount of players taken on, is a dribble. If they beat two but then lose it in the same play then it's unsuccessful. 

I think you're sceptical of the use of quite a lot of data, which is fine you do you, but clubs take these things extremely seriously these days, and increasingly so given current scouting climate with COVID.

Link to post
Share on other sites
  • Replies 3.3k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Honestly I'm at a point where I just can't get excited about new signings as long as Mowbray is in charge. Even if the signee appears impressive I already know his impact will be neutered by Mowbray's

Hi all. Would like to thank anyone and everyone who has given my work a read over the last few days. As you can probably tell it took a very long time to do and a lot of late nights around my full tim

Fast forwards 12 months.... 'We lost the quality of Holtby and the goals of Armstrong last summer along with the regular right back and the experience of Johnson.  It's a sad day for the club but

Posted Images

3 hours ago, roversfan99 said:

It just seems a bit of a swing from you thinking he is likely to go in this international break!

It seems more likely based on past events that if Armstrong goes,

No its not a swing as it was always a possibly he would be here next season since he is contract for next season. Despite my own views on the manager situation. 

Of course it more than likely based on Armstrong on him leaving

 

2 hours ago, JoeH said:

that data is worked into percentiles. He's in the 6th percentile out of 20 for Dribble Success %. It's a format enabling the data to be presented in the style of a Football Manager profile for the purpose of digestible content for Twitter users.

For clarity, Jerry Yates' Dribble Success % sits at around 46%, and for comparison, Sam Gallagher's sits at 45%

Blackpool manager has said he not for sale

2 hours ago, JoeH said:

Another name is Paul Mullin who has 24 goals this season in League Two.

Not incredibly young but still pre-peak age at 26. A big step up to Championship level but a cheap option if someone's willing to take a punt. A goalscorer is a goalscorer and he's just a natural one. Goals from all kinds of areas, right place right time. Unlike anything we really have since the departure of Danny Graham in terms of a real proper striker.

Available on a free at the end of his contract this summer and almost guaranteed a move up the EFL, could be a good squad option for us.

Is Paul Mullin more of a target man type of striker? Do you have any more info on him? 

If Mowbray did leave and if We appoint Alex Neil for an example I wonder if he target Norwich's striker Jordan Hugill. Good player and not playing at Norwich.  

1 hour ago, roversfan99 said:

Again, sceptical of the use of such a statistic. If a player takes on 2 players, gets to a third and loses it, does that come out as 66.67% successful, or 0%? How can you possibly define whether something is a dribble, does the close proximity of an opponent come into it? Impossible to objectify.

But more clubs in the current climate will be using data analysis and Wyscout. Do you accept? 

41 minutes ago, JoeH said:

 

I think you're sceptical of the use of quite a lot of data, which is fine you do you, but clubs take these things extremely seriously these days, and increasingly so given current scouting climate with COVID.

especially with Club scouts not allow to travel due to covid restrictions so much more focus on Data Analysis and Video analysis using Wyscout 

Link to post
Share on other sites
2 hours ago, JoeH said:

On Harry Chapman, he's featured in 14 Championship & 9 FA/League Cup games as a Blackburn Rovers player. I don't personally believe he has impressed for us at any level besides in League One. I think he's found his place there and whilst I find it frustrating that he's not under contract til 2024, and therefore saleable for a million pounds, he's had enough chance for me personally to have impressed at Rovers.

We'll have to agree to disagree on him, but I wouldn't be massively against him staying at the club either. 

Anyone dismissing the Charlie Kirk suggestion should really look into what he has going with Harry Pickering. Their partnership is mighty impressive and if they can connect with each other so well, and he's available so cheap, I think it would be foolish not to give it a go. 

We've seen how important successful partnerships can be with Dack & Graham, Rhodes & Gestede. You can't teach/train that understanding between players very quickly.

How many of them are starts, if any? I am not saying that Chapman definitely should be here next season, although if he was under contract, there is no way that anyone would pay a million pounds for him anyway. But I dont understand why you are so dismissive of a player who has never started a Championship game that I can recollect, yet are championing similar players with similar goal tallies at the same level albeit across many more games this season, on the basis that they havent had a chance when Chapman equally has not.

One thing that I am still unclear on too is whether you think it is necessary to sign a striker to replace Armstrong, as if you are saying that we do need a new striker but also a couple of wide men/attacking midfielders because replacing Armstrongs goals with 1 player may be nigh on impossible, I would tend to agree, if it is just to sign wide men/attacking midfielders and not bring in a striker, then it is not something I understand.

I do get your point on partnerships although for me they were never really a pairing that benefitted from one another, they didnt link up etc, it was more that we had 2 excellent strikers at that time. I am not sure that we should sign more Crewe players based on how they link up within Crewe's specific set up, what if Pickering gets injured or fails to perform?

57 minutes ago, JoeH said:

One dribble, regardless of the amount of players taken on, is a dribble. If they beat two but then lose it in the same play then it's unsuccessful. 

I think you're sceptical of the use of quite a lot of data, which is fine you do you, but clubs take these things extremely seriously these days, and increasingly so given current scouting climate with COVID.

I am sceptical because many of the stats are meaningless, especially out of context, at best they may display tendencies of a players style which you would know anyway. Many are flawed, for example a dribble is very difficult to tangibly ascertain, and also it means that taking on 4 men and passing, or turning around near 1 man and passing, both are equally going to count as a successful dribble. Stats like xG, so many issues, how do you objectify a chance, if a player misses a tap in by a cm, it doesnt count towards, if a team scores an early goal, the other team will then likely and naturally have a higher xG due to chasing a goal for so long, does that prove anything, etc. For defenders in particular, how do you decifer what stats are relevant, if a defender makes more interceptions/tackles, is he better, or is he out of position more often. You have to factor in how teams play, how good a team is, so many limitations that I can see.

Data usage will have a use but surely that would be to narrow down the search for a player, if we are signing players solely off data or selected clips then that is a massive worry. We need to be watching numerous full games, even if that is only allowed via TV due to the circumstances, not ideal I concede, before deciding whether to sign a player. Data has a use but it has to be taken with a huge dollop of salt.

I think the problem is when people go down a road whereby they feel the data is objective and proves a point, which it doesnt. Data can be manipulated to prove anything, and due to the limitations and varying relevance of data, it proves nothing more than a persons opinion IMO.

Link to post
Share on other sites
30 minutes ago, roversfan99 said:

One thing that I am still unclear on too is whether you think it is necessary to sign a striker to replace Armstrong, as if you are saying that we do need a new striker but also a couple of wide men/attacking midfielders because replacing Armstrongs goals with 1 player may be nigh on impossible, I would tend to agree, if it is just to sign wide men/attacking midfielders and not bring in a striker, then it is not something I understand.

 

The point from me is that yes in a perfect world you'd want a striker to replace Armstrongs goals, but on the budget we're likely to have I don't think it's at all viable to look to do that. Instead we need to invest in more financially sensible areas of the pitch, in an attempt to get goals out of the expensive forwards we already have.

Whilst you disagree, I think it's more than possible to get 15 goals out of Sam Gallagher, if fed properly and played in the right place. 

Link to post
Share on other sites
35 minutes ago, roversfan99 said:

Stats like xG, so many issues, how do you objectify a chanceif a player misses a tap in by a cm, it doesnt count towards, if a team scores an early goal, the other team will then likely and naturally have a higher xG due to chasing a goal for so long, does that prove anything, etc

1. How do you objectify a chance? Well that's what an xG model is. It takes into account all the variables such as shot type, pass quality, position of defenders, distance from goal, goalkeeper position, placement of shot etc etc etc.. 

2. It certainly does in most models

3. Yes it shows you the evidence as to how much further forward a team perhaps looked to play once behind. If you look at the xG diagrams over 90 minutes for your upcoming oppositions last five games and see that they always spike once trailing, you'd perhaps tailor your approach to the next game with that knowledge in mind.


If a team has a high xG value but a low G value you can determine whether they are wasting big chances, or perhaps often looking to shoot from range. I think personally you can see when Rovers have played well when you look at our xG. The win over Millwall, which is one of our most convincing wins since the early push against Wycombe/Derby etc.. had us with an xG value right around the 2-2.5 mark, where we got 2 goals. We've scraped wins over sides like Rotherham and our xG was lower than our actual G value, which shows we didn't actually play too well at all.

It's not the defining data point, we could survive without it but I think it tells us a lot of information that would otherwise take a long time to decipher manually.

Link to post
Share on other sites
1 minute ago, Sparks Rover said:

Its amazing how people forget.

What I remember is Sam Gallagher being completely pointless as a winger. When played as a central striker on his own he looks really average, but in my mind with a striker partner and balls into the box for 90 minutes, 15 goals for Sam Gallagher is a perfectly reasonable expectation. 

Link to post
Share on other sites
4 minutes ago, JoeH said:

What I remember is Sam Gallagher being completely pointless as a winger. When played as a central striker on his own he looks really average, but in my mind with a striker partner and balls into the box for 90 minutes, 15 goals for Sam Gallagher is a perfectly reasonable expectation. 

Aye mate...dream on.

Link to post
Share on other sites
48 minutes ago, JoeH said:

The point from me is that yes in a perfect world you'd want a striker to replace Armstrongs goals, but on the budget we're likely to have I don't think it's at all viable to look to do that. Instead we need to invest in more financially sensible areas of the pitch, in an attempt to get goals out of the expensive forwards we already have.

Whilst you disagree, I think it's more than possible to get 15 goals out of Sam Gallagher, if fed properly and played in the right place. 

Until that happens, its all opinions on conjecture, but my point is more that it would leave us short to only go into the season with Gallagher as a solitary striker. You cant put off recruiting in obvious areas simply because it isnt "financially sensible" and we cant have one striker all season. If Gallagher gets injured, we have no striker.

41 minutes ago, JoeH said:

1. How do you objectify a chance? Well that's what an xG model is. It takes into account all the variables such as shot type, pass quality, position of defenders, distance from goal, goalkeeper position, placement of shot etc etc etc.. 

2. It certainly does in most models

3. Yes it shows you the evidence as to how much further forward a team perhaps looked to play once behind. If you look at the xG diagrams over 90 minutes for your upcoming oppositions last five games and see that they always spike once trailing, you'd perhaps tailor your approach to the next game with that knowledge in mind.


If a team has a high xG value but a low G value you can determine whether they are wasting big chances, or perhaps often looking to shoot from range. I think personally you can see when Rovers have played well when you look at our xG. The win over Millwall, which is one of our most convincing wins since the early push against Wycombe/Derby etc.. had us with an xG value right around the 2-2.5 mark, where we got 2 goals. We've scraped wins over sides like Rotherham and our xG was lower than our actual G value, which shows we didn't actually play too well at all.

It's not the defining data point, we could survive without it but I think it tells us a lot of information that would otherwise take a long time to decipher manually.

My point was that it is inherently flawed though, I appreciate its intention but im saying that it is impossible to do to anything like pure authenticity what it is trying to do. For example, Gazza missed the ball by millimetres at Euro 96, that would go down as 0 xG when in fact it was as close as you could come. According to xG, a player missing the ball in front of an open goal, or say being really close and choosing the wrong option, is not a chance at all, whereas a long range pot shot would register, even obviously a very small xG. Even if you take into account all variables, it is still very difficult to make it objective too.

I also do question what use it has, in that like I said, a game will pan out dependant on goals, so say when we played Stoke away, we would have been miles up on their xG. They scored early and sat deep, always had us at arms length, but if you looked at xG, I bet it looked like we was the far superior team. I read a recent article on Brighton and how they are doing so well in the xG league, but questioning what that actually means. Can they conclude that they should carry on and things will fall their way, or is it poor finishing that will remain? And if its over a long period of time, does it perhaps show an example of how results and xG have far from a 100% correlation?

In reality, I personally think that it is easy without it to determine how you think it has played, based on the key chances which as mentioned may fall through the cracks of xG. A low xG doesnt prove poor performance, and the same for high xG.

One thing I would say that data does however that is of benefit you touch upon at the end, it can be very time saving. You shouldnt make key decisions including transfer recruitment solely on data though, and it doesnt prove anything on its own, it can be used in tandem with opinions but it wouldnt make one superior or indeed factual.

Link to post
Share on other sites
2 hours ago, roversfan99 said:

You shouldnt make key decisions including transfer recruitment solely on data though, and it doesnt prove anything on its own, it can be used in tandem with opinions but it wouldnt make one superior or indeed factual.

Nothing beats a scouts eye, and then the stats will  back his or her decision to a limited extent.  I agree with you, all this percentile gubbins is just people trying to justify their jobs.  

Football is an old boys club anyway, until that changes nothing changes 

Link to post
Share on other sites
6 hours ago, Sparks Rover said:

I think you're more likely to get 15 out of Brereton and I'd look to sign a smaller player to play with him, a nippy lad like AA. What happened to Zac Clough?

 

Attacking midfielder at Wigan. Funnily enough he was at Forest alongside Brereton...

Link to post
Share on other sites
21 minutes ago, Mike E said:

Attacking midfielder at Wigan. Funnily enough he was at Forest alongside Brereton...

Notlob, he's from Westhoughton or near there...He had bags of potential, small, low centre of gravity, bit of pace....seemed to just disappear 

Edited by Sparks Rover
Link to post
Share on other sites
7 hours ago, JoeH said:

The point from me is that yes in a perfect world you'd want a striker to replace Armstrongs goals, but on the budget we're likely to have I don't think it's at all viable to look to do that. Instead we need to invest in more financially sensible areas of the pitch, in an attempt to get goals out of the expensive forwards we already have.

Whilst you disagree, I think it's more than possible to get 15 goals out of Sam Gallagher, if fed properly and played in the right place. 

Do you not think we would have decent budget to improve the team overall and good replacement for Armstrong if we sold him? 

I do understand your point about improve team and getting more goals from our wide players and midfield. 

3 hours ago, Tormund said:

Clarke-Harris? Scores for fun, only 26. Bound to be clubs looking at him.

How much would Peterborough sell him for? 

Link to post
Share on other sites
13 hours ago, roversfan99 said:

Until that happens, its all opinions on conjecture, but my point is more that it would leave us short to only go into the season with Gallagher as a solitary striker. You cant put off recruiting in obvious areas simply because it isnt "financially sensible" and we cant have one striker all season. If Gallagher gets injured, we have no striker.

My point was that it is inherently flawed though, I appreciate its intention but im saying that it is impossible to do to anything like pure authenticity what it is trying to do. For example, Gazza missed the ball by millimetres at Euro 96, that would go down as 0 xG when in fact it was as close as you could come. According to xG, a player missing the ball in front of an open goal, or say being really close and choosing the wrong option, is not a chance at all, whereas a long range pot shot would register, even obviously a very small xG. Even if you take into account all variables, it is still very difficult to make it objective too.

I also do question what use it has, in that like I said, a game will pan out dependant on goals, so say when we played Stoke away, we would have been miles up on their xG. They scored early and sat deep, always had us at arms length, but if you looked at xG, I bet it looked like we was the far superior team. I read a recent article on Brighton and how they are doing so well in the xG league, but questioning what that actually means. Can they conclude that they should carry on and things will fall their way, or is it poor finishing that will remain? And if its over a long period of time, does it perhaps show an example of how results and xG have far from a 100% correlation?

In reality, I personally think that it is easy without it to determine how you think it has played, based on the key chances which as mentioned may fall through the cracks of xG. A low xG doesnt prove poor performance, and the same for high xG.

One thing I would say that data does however that is of benefit you touch upon at the end, it can be very time saving. You shouldnt make key decisions including transfer recruitment solely on data though, and it doesnt prove anything on its own, it can be used in tandem with opinions but it wouldnt make one superior or indeed factual.

McBride?

Link to post
Share on other sites

Just looking at the midfielders who's contracts and loans are up is eye watering.

What's the betting this manager tries to keep Bennett, Evans and Downing then lets the rest go ?

I fear next season will be a real struggle or another mid table plod at best, unless there is some proper ambition or nous from somewhere.

Link to post
Share on other sites
12 hours ago, Sparks Rover said:

Nothing beats a scouts eye, and then the stats will  back his or her decision to a limited extent.  I agree with you, all this percentile gubbins is just people trying to justify their jobs.  

Football is an old boys club anyway, until that changes nothing changes 

Level 2 qualified Scout here and I don’t disagree with you. Both traditional scouting and modern data scouting have their place.

Link to post
Share on other sites
15 hours ago, Sparks Rover said:

I think you're more likely to get 15 out of Brereton and I'd look to sign a smaller player to play with him, a nippy lad like AA. What happened to Zac Clough?

 

So my solution is a load of tosh and I’m dreaming, and your solution is to sign Zach Clough?

I think I’d win the public vote off personally, get myself through to judges houses.

Link to post
Share on other sites
15 hours ago, roversfan99 said:

I also do question what use it has

I feel like I’ve stated many uses but maybe they’ve been missed

15 hours ago, roversfan99 said:

You shouldnt make key decisions including transfer recruitment solely on data though, and it doesnt prove anything on its own

Of course you shouldn’t. But it’s relevance is becoming increasingly pertinent within the wider football world. 

 

15 hours ago, roversfan99 said:

it can be used in tandem with opinions but it wouldnt make one superior or indeed factual.

I think personally if one person says “this player is really poor with his passing” and then shows you that on average his pass success % data falls significantly below the leagues mean average for his position, then their opinion is definitely backed up by a little bit more evidence. Maybe not pure fact, but not far off given the mathematical proof that says the player is a poor passer by positional & league comparison.

Link to post
Share on other sites
4 minutes ago, JoeH said:

So my solution is a load of tosh and I’m dreaming, and your solution is to sign Zach Clough?

I think I’d win the public vote off personally, get myself through to judges houses.

I just said, what happened to Zac Clough....don't remember saying we should sign him....

Yes, I have many years scouting behind me, all the best players have been discovered by a scout, not an algorithm.

Link to post
Share on other sites
1 minute ago, Sparks Rover said:

I just said, what happened to Zac Clough....don't remember saying we should sign him....

Yes, I have many years scouting behind me, all the best players have been discovered by a scout, not an algorithm.

And the gap between modern and traditional scouting is closing every single year.

Every single team analyses video data for every single game. Every single recruitment department spends thousands on data softwares & databases.

There isn’t a single scout at a football game watching a player at the top level, who’s club didn’t originally find or analyse that player using data.

Link to post
Share on other sites
44 minutes ago, tomphil said:

Just looking at the midfielders who's contracts and loans are up is eye watering.

What's the betting this manager tries to keep Bennett, Evans and Downing then lets the rest go ?

I fear next season will be a real struggle or another mid table plod at best, unless there is some proper ambition or nous from somewhere.

Unless there is a  change of manager  , bottom 5 next year ...

Relegation will be a realistic option .

Link to post
Share on other sites
21 minutes ago, JoeH said:

 

There isn’t a single scout at a football game watching a player at the top level, who’s club didn’t originally find or analyse that player using data.

Thats rubbish, sorry.  But let's agree to disagree.  

Link to post
Share on other sites
  • K-Hod pinned this topic
  • K-Hod locked this topic
  • K-Hod unlocked this topic

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.