Norwich have Kim Hellberg on their radar

[cmajack]

It’s bollox. I always think of shots on target.
Player hits a thunderbolt and the ball is going in until the keeper makes a world class save.
Five minutes later player stubs his toe taking a shot and the ball trickles up to the keeper which he stops with his little finger.
Both go into the shots on target category.
It’s utter nonsense.

Love to have seen what Bill Shankly and Brian Clough would have made of this Xg bullshit.

 

[Skippyjack]

The dinosaurs are out tonight. It only takes a quick Google search to show you how xG stats are created and while I don't think they are as important as some do, they are a better yardstick of chances created than shots on target (as Daz ironically pointed out above).

As ever with statistics, xG is only as useful if the people analysing them know what they're doing. Brentford and Brighton are rightly praised for their stellar recruitment and you're deluding yourself if you think they don't extensively use analytics/statistics to make those signings.

Brighton and Brentford use a players position selection process, nothing to do with xG really.. but you obviously don't know what you're on about.. xG is f*cking stupid, it's solely based around someone's opinion on whether a chance is any good, you don't ask a sample of 100 people to come to the supposed correct conclusion within one minute.. the evidence is biased, because it's only one person's interpretation of events unfolding in front of their eyes in their opinion.. it's an arrogant way to look at a football match

 

[Andrew - North Hill]

It's really not hard to understand xG, you don't have to be a "nerd".

How many times have you come away from the Vetch/Liberty saying "how the **** did we win that, they battered us" or "how the **** did we not win that, we battered them". xG or xGA is basically a way of making sense of that.

It's a way of trying to measure how lucky or unlucky you were in a match, or even a season. If you are consistently overperforming on either count, you've got a problem on your hands (see Cardiff).

Again, I'm sympathetic to the argument that for one off games ultimately it doesn't mean all that much. But that still doesn't render it completely useless as a metric.

 

[Skippyjack]

It's really not hard to understand xG, you don't have to be a "nerd".

How many times have you come away from the Vetch/Liberty saying "how the **** did we win that, they battered us" or "how the **** did we not win that, we battered them". xG or xGA is basically a way of making sense of that.

It's a way of trying to measure how lucky or unlucky you were in a match, or even a season. If you are consistently overperforming on either count, you've got a problem on your hands (see Cardiff).

Again, I'm sympathetic to the argument that for one off games ultimately it doesn't mean all that much. But that still doesn't render it completely useless as a metric.

xG is created by one person pressing buttons, you don't use a 100 people sample within one minute, they make assumptions that if you're in a specific area of the pitch and you miss a chance, it's written Into a computer system that churns out a percentage of the chance, on data that is collected solely by a person having an opinion on whether it's a good chance or not

 

[Andrew - North Hill]

the evidence is biased, because it's only one person's interpretation of events unfolding in front of their eyes in their opinion

It's not opinion, though.

It's based on analysis of thousands of shots taken from different positions in and around the penalty area, from 30 yards, from 40 yards, wide positions, from the D, etc etc.

There is a known % chance you will score from each of those areas of the pitch, based on actual real outcomes.

So if you shoot from the D there's a 40% chance xG value that you'll score, inside the 6 yard box it's 90% (I've made those exact figures up, but you get the idea)

There's no opinion involved.

There's a reason bookies and betting syndicates use this data, and you don't find them wrong too often.

 

[Darran]

Love to have seen what Bill Shankly and Brian Clough would have made of this Xg bullshit.

I couldn’t agree more especially when a team is playing the Christmas tree formation with three at the back,a pyramid with one upfront,one in the hole and a false 9.

 

[Neath_Jack]

I couldn’t agree more especially when a team is playing the Christmas tree formation with three at the back,a pyramid with one upfront,one in the hole and a false 9.

But is a midfielder a 6 or an 8? And how lucky is he?

Fucking nonsense made up by nerds for other nerds to pull themselves off to.

 

[Skippyjack]

It's not opinion, though.

It's based on analysis of thousands of shots taken from different positions in and around the penalty area, from 30 yards, from 40 yards, wide positions, from the D, etc etc.

There is a known % chance you will score from each of those areas of the pitch, based on actual real outcomes.

So if you shoot from the D there's a 40% chance xG value that you'll score, inside the 6 yard box it's 90% (I've made those up, but you get the idea)

There's no opinion involved.

There's a reason bookies and betting syndicates use this data, and you don't find them wrong too often.

Who decides whether it's a good chance though? It's not a 100% fact, if you put it in that area, he may or might score based on a percentage, because Pele did it 100 years ago in that game in a similar position.. plus percentages differ the more games you play.. it's stupid, this 5 game winning streak, must be hurting our new assembled data statisticians, using the good old defend and nick a goal, I bet it's bruising egos..

 

[bytholwyn]

"Expected Goals (xG) is a metric that assesses the quality of a shot and predicts its likelihood of resulting in a goal. It uses historical data to estimate the probability of a goal based on factors like the shot's location, angle, assist type, and whether it was a header or a big chance. xG values range from 0 to 1, with 1 representing a near-certain goal.

How xG is used:

• • Evaluating Player and Team Performance:
    By tracking a player or team's xG, you can get a more nuanced understanding of their attacking and defensive capabilities than simply looking at goals scored or shots taken. For example, a team might lose 1-0 but have a high xG, indicating they created numerous high-quality chances. "
  • 
    
  • 
    xG comes into its own when you try and gauge how clinical a player is in front of goal. For instance Macaulay Langstaff scored a shed load of goals for Notts County. But his expected goals were even greater. This reflected the fact he was given fantastic service, especially by Jodi Jones. So, any recruitment team would sensibly take that into account. Macaulay's failure at Millwall suggests their recruitment team didn't look closely enough at the context of his goalscoring feats.
  • 
    
  • Whittaker, in contrast, outscored his expected goals by some distance last season, with a lot of his clinical finishes coming from outside the box. That over performance is highly rated, although of course that clinical finishing ability is offset by his low work rate - which again can be measured in a whole host of ways.
    
    
  • Will Evans is apparently attracting interest from Preston and other unnamed Championship clubs. He's consistently outscored his expected goals, so he should be of interest, especially in light of his excellent defensive stats, which should marry well with our high pressing ethos.

 

[Andrew - North Hill]

Who decides whether it's a good chance though?

Data decides.

Thousands and thousands and thousands of instances where someone has taken a shot from a certain position, and scored, or not scored.

Surely you understand that you can attach a value to that? That if someone takes a shot from the penalty spot, they're more likely to score than if they take a shot from their own half?

That's not hard to understand?

 

[Skippyjack]

Data decides.

Thousands and thousands and thousands of instances where someone has taken a shot from a certain position, and scored, or not scored.

Surely you understand that you can attach a value to that? That if someone takes a shot from the penalty spot, they're more likely to score than if they take a shot from their own half?

That's not hard to understand?

Who interprets the data? I could go on all day

 

[bytholwyn]

One of the possibly bad consequences of an understanding of xG in the game is the decline in pot shots from distance. Because statistically the chance of scoring goals from distance is very small. So, there's been a drop off in efforts from outside the box. Trouble is, from a spectacle point of view long range efforts are very much desirable, such as Brannagan's effort in Monday, and not just because of the opponent...

 

[Andrew - North Hill]

Who interprets the data? I could go on all day

You have completely lost me now. I can only conclude it's beyond your understanding, which I guess is fair enough.

 

[Neath_Jack]

I don’t know why we bother employing a manager, may as well just put a laptop in the fucking dugout

And people moan about ‘process’ managers when we’ve got this nonsense flying about. Get in the bin with it.

 

[Skippyjack]

You have completely lost me now. I can only conclude it's beyond your understanding, which I guess is fair enough.

Yeah, OK mate, you might develop a brain one day, but I doubt it