Generic small time football blather thread FOREVER

[RobinRobin 9227]

28 minutes ago, Craig said:

 There's no need to be a condescending twat about it.

Now if there was an xCondescendingTwat chart Gemmill would be top by a mile 😉 

[Craig 6306]

11 minutes ago, Monkeys Fist said:

Clearly not you, shitdipper 

 

Stander-uppers are dirty fucking hounds. Your arse must look the world's worst tie-dye

[PaddockLad 14585]

19 minutes ago, The Fish said:

 

xG is objectively assessing the likelihood of a goal given thousands of data points across years of research. How often does a player score, with that body part, from that position, at that angle, after receiving the ball in that way from that place, when the defenders are there and the 'keeper is there. It doesn't account for who that player is, or who his teammates, or a subjective assessment of the difficulty of the chance. 

 

Wilson's goal was a 0.09 xG. So, if a player was in that position you'd only expect him to score 9/100. Not necessarily Wilson, or Haaland, or Botman or Dubravka, any player in that situation had a 9% chance of scoring.

 

 

They're not, it's not do I think that's a good chance, it's how many times did that situation result in a goal. 

 

There's so many intangibles added by a scout watching a player that you cannot find in statistics, that's why more onus was put on personality, character, bravery, etc.  But the margins are now so small, that statistics add objective value to a scouts subjective view. You will never see the likes of Ali Dia again. Or a Nacho Gonzalez. Looks mint on youtube, might look mint in the 2-3 games the scouts turns up for. But if his underlying numbers are bad, they're a better judge of the player's actual ability. 

 

Your eyes lie to you all the time.

I think there’s a bit of goalpost moving going on here  

 

The chances of Botman being in that position to execute that chance that Wilson scored from on Saturday are far less than Wilson being in there, let alone executing the finish. That being the case, am not sure having that 0.09 xG figure for all players in that scenario is if any use to man or bloody beast tbh . 
 

You’ve ignored the quality of the great football men who built their sides very largely without access to any technology. To me you seem to suggest stats are reinventing the wheel when it comes to recruitment. They’re an aid nothing more. For me the past proves it…am not sure what stats got Gary Birtles from Long Eaton Rovers to a league title in three years, and two European cup winners medals in the following two… If I’ve misrepresented what you’ve said sorry but  am of the opinion science can help but it’s not the be all and end all.. 

[The Fish 8989]

2 minutes ago, PaddockLad said:

I think there’s a bit of goalpost moving going on here  

 

The chances of Botman being in that position to execute that chance that Wilson scored from on Saturday are far less than Wilson being in there, let alone executing the finish. That being the case, am not sure having that 0.09 xG figure for all players in that scenario is if any use to man or bloody beast tbh . 
 

You’ve ignored the quality of the great football men who built their sides very largely without access to any technology. To me you seem to suggest stats are reinventing the wheel when it comes to recruitment. They’re an aid nothing more. For me the past proves it…am not sure what stats got Gary Birtles from Long Eaton Rovers to a league title in three years, and two European cup winners medals in the following two… If I’ve misrepresented what you’ve said sorry but  am of the opinion science can help but it’s not the be all and end all.. 

 

It's not about the likelihood of that chance occurring. It's simply given all the factors I mentioned above, how often does that situation result in a goal. Schar's shot is a good example. The chance of a player scoring from there is 1/100. that's averaged out over thousands of shots from thousands of players. A player has more chance of scoring that chance with his dominant foot, right? More chance if he's under no pressure, right? More chance if the 'keeper decided to go walkabout to the opposite corner flag, and so on. 

 

the xG records, not the predicted goals, but the expected goals. So someone takes that shot, it's 1 in 100 that it goes in. Because, say, 100 players took that type of shot across the historical data of 10,000 instances just like that, and 1 went in. You'd expect that when Schar strides forward against Brighton and finds himself in the exact same situation, you'd expect him to miss 99 times 100.

 

People traversed the globe without GPS, flight, or even a combustion engine. Doesn't mean those things are superfluous.

 

It's not the be all and end all, but it's a very useful tool. That's why I said scouting in person adds intangibles that stats alone can't. How does the player react to going behind, or missing a chance. How does he interact with the officials, his teammates, the opposition. You can never replace the value of an experienced eye, but these metrics aren't solely useful for assessing a teenage Bulgarian plying his trade in the Dutch second division. They're also really useful for working out the way to get the most out of the players you've got. Of course you need humans to apply subjectivity, and interpret the data. Do it wrong (over-exaggerating the value of POMO) and you get Allardyce. Do it right and you get the high quality, high energy, fluid football of Man City and Liverpool.

[Dr Gloom 19824]

50 minutes ago, Craig said:

 

Who fucking stands up for christ's sake

I’m mentioning no names. 
 

Cough

 

 

Cough

 

 

 

 

Fist.

 

…the filthy fucker

[Dr Gloom 19824]

4 minutes ago, The Fish said:

 

It's not about the likelihood of that chance occurring. It's simply given all the factors I mentioned above, how often does that situation result in a goal. Schar's shot is a good example. The chance of a player scoring from there is 1/100. that's averaged out over thousands of shots from thousands of players. A player has more chance of scoring that chance with his dominant foot, right? More chance if he's under no pressure, right? More chance if the 'keeper decided to go walkabout to the opposite corner flag, and so on. 

 

the xG records, not the predicted goals, but the expected goals. So someone takes that shot, it's 1 in 100 that it goes in. Because, say, 100 players took that type of shot across the historical data of 10,000 instances just like that, and 1 went in. You'd expect that when Schar strides forward against Brighton and finds himself in the exact same situation, you'd expect him to miss 99 times 100.

 

People traversed the globe without GPS, flight, or even a combustion engine. Doesn't mean those things are superfluous.

 

It's not the be all and end all, but it's a very useful tool. That's why I said scouting in person adds intangibles that stats alone can't. How does the player react to going behind, or missing a chance. How does he interact with the officials, his teammates, the opposition. You can never replace the value of an experienced eye, but these metrics aren't solely useful for assessing a teenage Bulgarian plying his trade in the Dutch second division. They're also really useful for working out the way to get the most out of the players you've got. Of course you need humans to apply subjectivity, and interpret the data. Do it wrong (over-exaggerating the value of POMO) and you get Allardyce. Do it right and you get the high quality, high energy, fluid football of Man City and Liverpool.

[The Fish 8989]

1 minute ago, Dr Gloom said:

[Monkeys Fist 38761]

All you manky spelunkers, swabbing around Poop cavern blindly hoping to connect, clearly have something in common. 
Since your loudest and proudest proponent, Wi Pi Pi Gemmill is famously deficient in the forward trouser area, it’s just logical to conclude that you all have nothing stopping you either, as you blindly dab away from the front. 

I have to dry the tip after sitting, so my squat to wipe from the back is not only more hygienic in every sense, but a physical necessity due to my gi-fucking-normous man eggs and bairn’ses’ arm. 
 

 

[Craig 6306]

34 minutes ago, Monkeys Fist said:

All you manky spelunkers, swabbing around Poop cavern blindly hoping to connect, clearly have something in common. 
Since your loudest and proudest proponent, Wi Pi Pi Gemmill is famously deficient in the forward trouser area, it’s just logical to conclude that you all have nothing stopping you either, as you blindly dab away from the front. 

I have to dry the tip after sitting, so my squat to wipe from the back is not only more hygienic in every sense, but a physical necessity due to my gi-fucking-normous man eggs and bairn’ses’ arm. 
 

 

 

Standing up to wipe FFS. You must absolutely stink you rotten hound

[Gemmill 34461]

The Brown Butterfly. Fist's nickname at the launderette. 

 

[Alex 30409]

Wi Pi Pi  

[Craig 6306]

19 minutes ago, Gemmill said:

The Brown Butterfly. Fist's nickname at the launderette. 

 

 

 

Don't forget he's been tarting around in an artic truck. The stench in that cab must be horrific 

[PaddockLad 14585]

I get it @The Fish but an average of all players doesn’t tell me much when a chance drops to a specific player because of different individual skill sets. Am not sure what xG tells us that we can find useful.  The likelihood of any given player to score a goal given a set of variables at the point at which he takes a shot is always altered by the specific player the chance goes to. 
 

Strikers are likely to score more than defenders so the overall average is largely irrelevant. That’s just the way I think. Fab’s goal marks him out as having that in his skill set, which makes him look good versus the average, but we can tell that he’s a skilful attacking centre back capable of scoring goals just by watching him on the pitch…. the other thing about that is he’s scored similar before so that’s just counting…. 
 

Maybe I’ve got hold of the wrong end of the stick but I don’t see many applications for xG. It’s too general:

 

Dave “x player missed a chance that had an xG of y “ 

 

Me “Well who was the player?” 
 

 

 

 

 

[spongebob toonpants 3603]

That stat the other day about 6 passes before a challenge now as opposed to 24 passes before a challenge under Bruce. Interesting Stat, brings a bit of information to support the evidence that anybody with a brain could see happening.

xG. Boring wank adds nothing to the conversation

[Gemmill 34461]

This place is starting to feel like Jurassic Park. Dave, I'm Chris Pratt, you can be Dicky Attenborough. 

[Rayvin 4964]

57 minutes ago, PaddockLad said:

I get it @The Fish but an average of all players doesn’t tell me much when a chance drops to a specific player because of different individual skill sets. Am not sure what xG tells us that we can find useful.  The likelihood of any given player to score a goal given a set of variables at the point at which he takes a shot is always altered by the specific player the chance goes to. 
 

Strikers are likely to score more than defenders so the overall average is largely irrelevant. That’s just the way I think. Fab’s goal marks him out as having that in his skill set, which makes him look good versus the average, but we can tell that he’s a skilful attacking centre back capable of scoring goals just by watching him on the pitch…. the other thing about that is he’s scored similar before so that’s just counting…. 
 

Maybe I’ve got hold of the wrong end of the stick but I don’t see many applications for xG. It’s too general:

 

Dave “x player missed a chance that had an xG of y “ 

 

Me “Well who was the player?” 
 

 

 

 

 

 

I wasn't going to comment on this whole topic because I'm squarely on the fence and want to avoid being a "fanny"

 

That said though, as a data man at heart, i feel that xG does have merit even applied across the full spectrum of players. Yes forwards might have a higher xG than players in other positions absolutely, but they're also far more likely to actually be in an xG position to start with, since that is their job. So the value potentially becomes in analysing the player as an individual irrespective of their team.

 

If the best striker in the world is in a shit team, he may have very few chances to score. On paper, this may mean that he appears to be a poor player. "Only 10 goals per season? We need a 20 goal man ffs".

 

xG can demonstrate a nuance that we otherwise wouldn't be able to see, which is that if the player had a high xG held alongside a low overall return on goals, we might conclude that in a better team he would score far more. This potentially reveals "diamonds in the rough" as it were. If the player has a low xG but a high return on goals, maybe his teammates are doing most of the work for him and giving him plenty of chances to look good.

 

I think the latter formulation of this might be more obvious to casual viewing than the former.

 

Anyway that's my two cents. I want to point out that I actually know nothing about it, I've just vaguely assumed that this must be the premise. Watch it turn out that I'm wrong

[Monkeys Fist 38761]

27 minutes ago, Gemmill said:

This place is starting to feel like Jurassic Park. Dave, I'm Chris Pratt, you can be Dicky Attenborough. 

Indeed, life…ahh…finds a way

[The Fish 8989]

11 minutes ago, PaddockLad said:

I get it @The Fish but an average of all players doesn’t tell me much when a chance drops to a specific player because of different individual skill sets. Am not sure what xG tells us that we can find useful.  The likelihood of any given player to score a goal given a set of variables at the point at which he takes a shot is always altered by the specific player the chance goes to. 
 

Strikers are likely to score more than defenders so the overall average is largely irrelevant. That’s just the way I think. Fab’s goal marks him out as having that in his skill set, which makes him look good versus the average, but we can tell that he’s a skilful attacking centre back capable of scoring goals just by watching him on the pitch…. the other thing about that is he’s scored similar before so that’s just counting…. 
 

Maybe I’ve got hold of the wrong end of the stick but I don’t see many applications for xG. It’s too general:

 

Dave “x player missed a chance that had an xG of y “ 

 

Me “Well who was the player?” 

 

xG is better at gauging if someone is consistently scoring or missing good chances. If Wilson missed that chance he scored from, you can't criticise him for it really, it was a really difficult chance. Schar's goal, it doesn't matter who's out there, someone taking a shot from there has a small chance of scoring. 

 

a high xG tells us that the player gets into a situation to take a shot that you'd expect to lead to a goal. Combine that with their goal return and you get a mark of how good/bad a player is at putting away those chances. e.g. if Jonny Scoresalot get 20 league goals in a campaign, you may think, "Jonny's the one for me", but when you look and see his accumulated xG is 40 you realise that given the chances he's getting, he's only putting away half of what you'd expect. Likely because he's in a team that makes a load of good chances. 

 

If Shearer said "WIlson missed an easy chance today", you'd likely believe him, because Shearer knows his stuff. What you might not account for is this; Shearer may think of that as an easy chance, but that's because he was fucking mint and would regularly do better than expected given the chances that fell to him. You come on here and see some stat-nerd has said that it was a hard chance and Shearer's being harsh. What's more reliable, thousands of objective data points that say what the expected outcome of that situation was, or a subjective account from a biased witness?

 

 

Also, to take your example, I wouldn't say x player, I'd be specific.

 

Dave " Adama Traore has a PL career non-penalty xG of 0.11 per 90 minutes, he has scored 0.1 goals per 90.

Allan Saint-Maximin is npxG 0.16 and has scored 0.16 per 90.

Therefore ASM gets more chances per game, and puts more of them away than Traore did.

Couple that with ASM getting 41% of his shots on target to AT's 29.7%, 74% pass completion to 67%, 0.43 to 0.32 Goal creating actions per 90. It shows that, not only does ASM have the skills that get you out of your seat, he's also got decent end product that will likely increase when he's playing alongside better players, in a better system."

PL "Oh, so stats can help you back up your opinion that Traore is ASM's little muggy bitch boy?"

Dave "Yes PL, that's spot on, he's ASM's muggy little bitch boy"

 

 

 

 

 

[Gemmill 34461]

15 minutes ago, Rayvin said:

 

I wasn't going to comment on this whole topic because I'm squarely on the fence and want to avoid being a "fanny"

 

That said though, as a data man at heart, i feel that xG does have merit even applied across the full spectrum of players. Yes forwards might have a higher xG than players in other positions absolutely, but they're also far more likely to actually be in an xG position to start with, since that is their job. So the value potentially becomes in analysing the player as an individual irrespective of their team.

 

If the best striker in the world is in a shit team, he may have very few chances to score. On paper, this may mean that he appears to be a poor player. "Only 10 goals per season? We need a 20 goal man ffs".

 

xG can demonstrate a nuance that we otherwise wouldn't be able to see, which is that if the player had a high xG held alongside a low overall return on goals, we might conclude that in a better team he would score far more. This potentially reveals "diamonds in the rough" as it were. If the player has a low xG but a high return on goals, maybe his teammates are doing most of the work for him and giving him plenty of chances to look good.

 

I think the latter formulation of this might be more obvious to casual viewing than the former.

 

Anyway that's my two cents. I want to point out that I actually know nothing about it, I've just vaguely assumed that this must be the premise. Watch it turn out that I'm wrong

 

If a player has a low xG but a high actual goals, it means they're better at converting chances than average. 

 

So as an example, Shearer scored 5 goals at Euro 96 for which the xG was only 3.45. Tiny sample size obviously but I'm sure if you looked at his career stats they would tell a similar story. 

[Gemmill 34461]

4 minutes ago, Monkeys Fist said:

Indeed, life…ahh…finds a way

 

If you were a dinosaur you'd be stood over that with at least 30% of it spread all over your filthy Dino-cheeks. 

[Monkeys Fist 38761]

1 minute ago, Gemmill said:

 

If you were a dinosaur you'd be stood over that with at least 30% of it spread all over your filthy Dino-cheeks. 

 I bet Pickford is a sitter, the T-Rex armed filth monger. 

[Rayvin 4964]

3 minutes ago, Gemmill said:

 

If a player has a low xG but a high actual goals, it means they're better at converting chances than average. 

 

So as an example, Shearer scored 5 goals at Euro 96 for which the xG was only 3.45. Tiny sample size obviously but I'm sure if you looked at his career stats they would tell a similar story. 

 

So do I have it the wrong way round then? My understanding of xG is that it tracks goals you would expect to see from the position and circumstance that a player is in. So if someone has a low XG with high goals, to me that would suggest that this player has many opportunities in which you would expect them to score, and is only able to convert a limited number of them.

 

That's clearly the opposite of what you're saying, so I'm wondering what I'm misunderstanding in my premise about what xG is?

[Gemmill 34461]

3 minutes ago, Monkeys Fist said:

 I bet Pickford is a sitter, the T-Rex armed filth monger. 

 

Whatever option he chooses, he won't be able to reach. 

[Monkeys Fist 38761]

Just now, Gemmill said:

 

Whatever option he chooses, he won't be able to reach. 

This explains a lot. 

[Gemmill 34461]

3 minutes ago, Rayvin said:

 

So do I have it the wrong way round then? My understanding of xG is that it tracks goals you would expect to see from the position and circumstance that a player is in. So if someone has a low XG with high goals, to me that would suggest that this player has many opportunities in which you would expect them to score, and is only able to convert a limited number of them.

 

That's clearly the opposite of what you're saying, so I'm wondering what I'm misunderstanding in my premise about what xG is?

 

I don't know if you're just getting your low and high mixed around cos going from this:

 

Quote

My understanding of xG is that it tracks goals you would expect to see from the position and circumstance that a player is in.

 

To this application of it:

 

Quote

So if someone has a low XG with high goals, to me that would suggest that this player has many opportunities in which you would expect them to score, and is only able to convert a limited number of them.

 

... Doesn't make sense. 

 

If xG basically tells you how many goals a player's given chances should convert to, and the player is converting at a higher rate, then they're exceeding the expected conversion - so they're either lucky or they're good.