I think it's pretty useful for seeing who is potentially in a false position. For instance, Sunderland have been outperforming their xP (I think they were 6th before their last game against Coventry) so we might see them drop back down to earth a little.
Useful tools for managers to learn where they can improve their teams but I think if people try to use them as excuses or make claims and predictions about the future it falls short.
If we've got to the top without bossing possession or chances that's a positive for the team and an indicator of spirit. Strong teams lagging behind aren't in false positions they are more likely wasteful.
What does over or underperforming your xP mean? Is it a result of luck, or strikers who score more than you'd expect from the chances, or a goalkeeper saving more than you'd expect, or a combination of the above? Or something else altogether?
Does a team's xP tend to be pretty close to their actual points over the length of a whole season?
I have less truck with people talking about xP than they do xG, particularly in the sense of looking at a run of games and working out which games they expect to win, or draw or lose.
so in a run of 10 games, given the opposition coming up you can say you have an xP of 14 with 4 wins and 2 draws.
if they start saying “Our xP from the last game was 1.45” - they can get in the fucking bin. Same with “our xG for that game was 3.65 so we shouldn’t have lost 2-1”
Your aG (actual goals) were 1, and that’s all that matters.
Once you understand that xG rates a chance that falls to Erling Haaland as having the exact same likelihood of resulting in a goal than if that exact chance were to fall to Mark Rogers it helps you understand it a lot better
Actually, I've got a question about xg if someone will indulge me. Is it constantly updated with new data, and how far down the pyramid is data taken from, and from how many different countries?
Fair enough, but people say things like "We dominated and did not deserve to lose that" all the time, rather than "Here is the match report: we lost that 1-0". To my mind xP is just a way of turning that first comment into something resembling a statistic. I don't use it much myself, but the difference between "We dominated and did not deserve to lose that" and "We underperformed against our xG" or "We are underperforming this season against our xP" is basically the use of numbers, which some people like more than others.
It would not surprise me if part of our 'data driven' model which everyone is so excited about is 'Which players perform well against their own xG?'
It's such a blunt tool. If you come up against a side with a particularly good finisher then part of your game plan is going to be to organise yourselves in such a way that you prevent that player having chances to score.
That may mean that you're happy for another player to get in those positions because you know he's nowhere near as clinical. Because xg doesn't differentiate between who chances fall to it may well tell you that that team is "underperforming its xg"
The opposition manager may cite that as an example of being unlucky and deserving more, when in fact he's been out thought by the tactics of the opposition manager
True - and I would guess it is only one of many tools that are used. I think football is probably a much more difficult sport to analyze with data than something like baseball (which is obviously where Moneyball originated in the first place). To me, anything free flowing is going to present more of a challenge.
It's all about making things as objective and quantifiable as possible so you can make better-informed decisions in terms of recruitment etc. There's always going to be a level of subjectivity - that's the nature of football and it's a good thing - but clubs that get that approach right are generally going to succeed.
I’m more interested in how data can give an insight into the character of a player, given that has been a key component of the club for a decade or more.
Using either as a diagnostic tool so that those at a club can use it to enact change I have zero problem with. It’s information they can actually do something with.
Same as those that are inclined to bet on football and want to use it as a tool to try and predict what will happen in a game - where I believe the whole xG thing came from to start with. Those are the correct uses for it for me.
My issue, especially xG, is when it used by fans to describe what has happened as evidence to discuss how a game went. Football is more nuanced than that. XG does take into account, as far as I am aware, if a particular chance fell to haaland/zlatan/messi or indeed George wears cousin.
Stockport is interesting because they’ve actually had a good season apart from against us. Which makes our win there all the more special. Even their fans afterwards we’re complimentary about us and not too critical of their own side.
Come on chief, you should know there are certain gasroomers who literally live on twitter, Instagram, linkedin etc. Someone noticed within minutes and someone else queried why he was checking him on linked in etc
In our glorious Isthmian League days, as far as I can recall, anxiety about the imminent loss of key players was never a thing. It was more a case of looking forward to the next exciting signing.
With the interest in Richard Kone building rapidly among Championship clubs - see link below for the latest hint of interest - I’d like to think it will soon be a case of learning to live with the interest and stop worrying (a nod to the CND slogan re bombs).
As in the ‘seventies and ‘eighties, joining Wycombe Wanderers (if only on loan) should now be (and already seems to be) an attractive proposition for talented young players like Cameron Humphreys and Gideon Kodua.
Comments
Lampard is a perfectly ok Championship manager, but what a downgrade.
True i should have put Coventry fans.
Please tell me you’re joking and that ‘expected points’ is not a thing?
xP is a thing
Jesus.
It's based on xG, but I sense you're not keen on that either, so I shall 🤐
I think it's pretty useful for seeing who is potentially in a false position. For instance, Sunderland have been outperforming their xP (I think they were 6th before their last game against Coventry) so we might see them drop back down to earth a little.
I've decided I reject the concept of xP
Useful tools for managers to learn where they can improve their teams but I think if people try to use them as excuses or make claims and predictions about the future it falls short.
If we've got to the top without bossing possession or chances that's a positive for the team and an indicator of spirit. Strong teams lagging behind aren't in false positions they are more likely wasteful.
Whereas we've been absolutely clinical. Incredible stuff.
What does over or underperforming your xP mean? Is it a result of luck, or strikers who score more than you'd expect from the chances, or a goalkeeper saving more than you'd expect, or a combination of the above? Or something else altogether?
Does a team's xP tend to be pretty close to their actual points over the length of a whole season?
I have less truck with people talking about xP than they do xG, particularly in the sense of looking at a run of games and working out which games they expect to win, or draw or lose.
so in a run of 10 games, given the opposition coming up you can say you have an xP of 14 with 4 wins and 2 draws.
if they start saying “Our xP from the last game was 1.45” - they can get in the fucking bin. Same with “our xG for that game was 3.65 so we shouldn’t have lost 2-1”
Your aG (actual goals) were 1, and that’s all that matters.
Once you understand that xG rates a chance that falls to Erling Haaland as having the exact same likelihood of resulting in a goal than if that exact chance were to fall to Mark Rogers it helps you understand it a lot better
Actually, I've got a question about xg if someone will indulge me. Is it constantly updated with new data, and how far down the pyramid is data taken from, and from how many different countries?
Fair enough, but people say things like "We dominated and did not deserve to lose that" all the time, rather than "Here is the match report: we lost that 1-0". To my mind xP is just a way of turning that first comment into something resembling a statistic. I don't use it much myself, but the difference between "We dominated and did not deserve to lose that" and "We underperformed against our xG" or "We are underperforming this season against our xP" is basically the use of numbers, which some people like more than others.
It would not surprise me if part of our 'data driven' model which everyone is so excited about is 'Which players perform well against their own xG?'
It's such a blunt tool. If you come up against a side with a particularly good finisher then part of your game plan is going to be to organise yourselves in such a way that you prevent that player having chances to score.
That may mean that you're happy for another player to get in those positions because you know he's nowhere near as clinical. Because xg doesn't differentiate between who chances fall to it may well tell you that that team is "underperforming its xg"
The opposition manager may cite that as an example of being unlucky and deserving more, when in fact he's been out thought by the tactics of the opposition manager
True - and I would guess it is only one of many tools that are used. I think football is probably a much more difficult sport to analyze with data than something like baseball (which is obviously where Moneyball originated in the first place). To me, anything free flowing is going to present more of a challenge.
It's all about making things as objective and quantifiable as possible so you can make better-informed decisions in terms of recruitment etc. There's always going to be a level of subjectivity - that's the nature of football and it's a good thing - but clubs that get that approach right are generally going to succeed.
I’m more interested in how data can give an insight into the character of a player, given that has been a key component of the club for a decade or more.
Using either as a diagnostic tool so that those at a club can use it to enact change I have zero problem with. It’s information they can actually do something with.
Same as those that are inclined to bet on football and want to use it as a tool to try and predict what will happen in a game - where I believe the whole xG thing came from to start with. Those are the correct uses for it for me.
My issue, especially xG, is when it used by fans to describe what has happened as evidence to discuss how a game went. Football is more nuanced than that. XG does take into account, as far as I am aware, if a particular chance fell to haaland/zlatan/messi or indeed George wears cousin.
Madness. Stockport looked like they'd never stepped foot on a football pitch before when we played them
Stockport is interesting because they’ve actually had a good season apart from against us. Which makes our win there all the more special. Even their fans afterwards we’re complimentary about us and not too critical of their own side.
By the way, did anyone already note that Scott Mitchell rocked up at Oxford?
Come on chief, you should know there are certain gasroomers who literally live on twitter, Instagram, linkedin etc. Someone noticed within minutes and someone else queried why he was checking him on linked in etc
Ah, late to the party as always! 😂
@Boosh?
Thought I'd get in before anyone else.
Blooms 12/1 and 6th favourite for the Hull City job
Lampard confirmed
\o/
In our glorious Isthmian League days, as far as I can recall, anxiety about the imminent loss of key players was never a thing. It was more a case of looking forward to the next exciting signing.
With the interest in Richard Kone building rapidly among Championship clubs - see link below for the latest hint of interest - I’d like to think it will soon be a case of learning to live with the interest and stop worrying (a nod to the CND slogan re bombs).
As in the ‘seventies and ‘eighties, joining Wycombe Wanderers (if only on loan) should now be (and already seems to be) an attractive proposition for talented young players like Cameron Humphreys and Gideon Kodua.
https://footballleagueworld.co.uk/norwich-city-must-send-scouts-to-watch-tyrique-george-richard-kone/