Whats this xG stat all about?
- Thread starter Coopertrueblue
- Start date
ClockworkOrange
Well-Known Member
Two years in a row now they've been spot on.
I think it's also been fairly clear from watching us this season - we beat Hibs, Livvy and Aberdeen 1-0 this season but created so few chances and had to rely on scrappy goals to get us out of jail. That kind of performance isn't tenable over the course of the season, and so it proved.
Mount_Bear
Well-Known Member
It really isn't. It's just that people who don't know much about modern football don't understand it.
It's a good stat, one among many, that all teams will be using in their performance analysis. Like all stats though it's not meant to be looked at on the basis of a single game where variance can play a part, it's over the longer term that it is useful.
There were folk who were raging on the thread about their report earlier in the season. Turned out to be correct again.
spartacus
Well-Known Member
Also a good indicator of a team’s form. Handy stat for punters
CrunchBear
Well-Known Member
I don’t get why people are so against it. Basically trying to give a stat that is better to have one good chance than loads of speculative 30 yard shots
It’s not perfect but it’s useful.
It’s not perfect but it’s useful.
Leo_Ger
Well-Known Member
Exactly, the quality of the individual taking the shot argument doesn't hold up to invalidate xG.
The analysts have looked to see if any individual consistently outperforms their individual xG. Over a decade of stats, there was only one player who did, Messi.
It's quite funny that some people think that's something professional analysts wouldn't investigate before progressing with the metric.
BobbyShearer55-65
Well-Known Member
New age statto pish.
I don't need a spreadsheet to tell me when someone should have scored a chance.
I don't need a spreadsheet to tell me when someone should have scored a chance.
WinkieWATP
Well-Known Member
It’s a lot of bollocks.
jscotty1
Well-Known Member
Whether people think it's pish or not all the top clubs operate using data analysts to drive signings and influence tactics (including ours). It works.
That doesn't mean you have to like it, it's just that this is the direction football is going in and it's only going to become more ingrained in the sport at every level over time.
Watch the film 'moneyball' to see how data influenced another sport.
That doesn't mean you have to like it, it's just that this is the direction football is going in and it's only going to become more ingrained in the sport at every level over time.
Watch the film 'moneyball' to see how data influenced another sport.
GovanFront4
Well-Known Member
I’m doing a course in analytics and data at the moment and xG is covered. It really does help get a better picture of the game but it obviously shouldn’t be the be all and end all. Like all stats is just a guide and isn’t to replace your eyes.
You can also use xA as a guide for players like Kent, Hagi and Aribo who aren’t out and out strikers but can be judged on the quality of chances they create rather than simply assists.
One bit of data I seen recently was Messi’s sprint data. Followed him through a game and showed at what points he would break in to a sprint, jog etc. One interesting thing is that most of the time he received the ball he was standing completely still because his movement was already good enough to get him in the correct position unmarked.
You can also use xA as a guide for players like Kent, Hagi and Aribo who aren’t out and out strikers but can be judged on the quality of chances they create rather than simply assists.
One bit of data I seen recently was Messi’s sprint data. Followed him through a game and showed at what points he would break in to a sprint, jog etc. One interesting thing is that most of the time he received the ball he was standing completely still because his movement was already good enough to get him in the correct position unmarked.
rfc_chris_rfc
Well-Known Member
It’s a good useful stat for measuring out how teams will do for the season. Rangers Review guys have been putting out the Xg league table and shows that Celtic should’ve been top. As the season has played out that has came to fruition. Their tables have been correct for every season that I can remember following.
There can be outlier individual games were teams will have a lower Xg and will win the game, that’s why it’s a good measure of a block of games at a time.
There can be outlier individual games were teams will have a lower Xg and will win the game, that’s why it’s a good measure of a block of games at a time.
You mean those who pay attention to the important stuff, and ignore irrelevant trivia?
Dinosaur Loyal
Coda
Well-Known Member
That prick doesn't seem to have taken xFL into the equation (expected 19th Century Terrorist Linesman).
Coda
Well-Known Member
That's not the point of it. It's as much for the other players around them. I remember Daniel Candeias' expected assists was a fair bit higher than his actual assists. It's not his fault that the strikers didn't finish the chances that he created.
If a winger has 1 assist, but 20 expected assists, then you know it's the striker, not the winger that's the problem. Likewise if the expected goals was 5, and you ended up drawing 0-0, you probably want to sign the opposition keeper.
Marlo Stanfield
Well-Known Member
Video is a few years old now but still explains it well enough.
Last edited:
