Xg....Question

Yeah its definitely for the more nerdy types. The term 'Expected goals' itself is stupid especially if used post match, as it implies what should have happened rather than what actually happened which is useless. Chance quality or value would make more sense

 

Was listening to talksport this morning, thats when the Coventry stat was mentioned, I have to agree with Ally McCoist over you sorry mate

 

There’s a reason he’s not had a job in 11 years i reckon.

 

You being serious ha, works nearly every day on tv/radio so must av smmat about him?

 

It was tongue in cheek, but i’ll take the opinions of the people that work with these statistics daily over someone working in the media.

If they weren’t useful they wouldn’t exist, simple as that. It costs money for clubs to get access to this data and every club signs up for it. If it wasn’t useful they wouldn’t do it would they.

 

Kind of get what you saying mate, in my eyes it would not be the stat I would be judging players by tho.

 

It’s not used solely though, it’s used in combination with other stats, with traditional scouting etc.

Personally i think it’s useful. If a player has an xG much lower than their actual goals, that’s a decent indicator that they’re not getting chances but are overperforming with the chances they do get.

Doesn’t tell the whole story of course, that could be because they’re not getting themselves into the right positions etc. or could be that the rest of their team aren’t creating. That’s where other factors would play into it such as heatmaps, scouting etc.

Similarly if a player has a much higher xG compared to their goals it could indicate that the team is playing well and the player is just not finishing for example. (as was the case with Watters in my opinion)

 

Finally some sense

Want proof of this? Check out the promotion season under Stendel. We were miles ahead in both defence and attack and that ultimately reflected in the campaign

 

I like using xG as a stat to generally see how a game has gone. Obviously it’s not a 100% science but it’s better than using possession and just total shots.

 

Not trying to be pedantic/argumentative but how as the xG(1.1) given anyone an idea how the Coventry game as gone on when they won 7-1?....that’s my all point really mate.

 

It wouldn't be used in isolation to give an idea of how the game went. That's the point.

 

I will leave it at that then as I honestly don’t understand it’s value pal.

 

I can't stand it either mate.

 

I get the point of it, but it doesn’t work.

I think it is

A 50/50 chance is 0.5 XG , whereas statistically say 9 pens from 10 are scored it’s 0.9

I think …..

so it could give data on how clinical finishing is ….i guess ? Or whether a team ‘deserved it’ based on chances.

I might have it all wrong though as the figures are clearly nonsense as per QPR match

 

Yeah it’s basically how often each shot is statistically likely to be scored, based on things like ball position, defender positions etc. In the case of the Coventry game all their shots were in and around the 18 yard line or further out so they’re not going to be high xG. It just means they beat the odds, or probably more likely that the QPR defenders/keeper was having an off day.

 

That game is a bit of an outlier but using that game as an example. On the scores app I use it has Coventry’s xG as 1.27 and QPR’s as 0.61 so given Coventry’s is double I would say in general terms, they deserved to win the game.

As I said before, it’s not an exact science and not something I would argue with against someone who has actually watched the game but can be used as a general indicator.

 

They should create a stat for xPen. Ours is of course 0.