This xG nonsense

[sunderpitt]

If the Premier league table was set according to this xG silliness, that pundits are always mentioning and we are supposed to take seriously, we would be 4th bottom, just ahead of Aston Villa who would be 16th.

Thankfully we play on grass and not inside a computer program.

To sum it up technically... it's a load of bollocks

 

[halfpast3]

I love how worked up and confused people on here get about it. It will normally give a rough idea of where you should but of course there will always be outliers. It’s just a bit of data that can be used. I think it’s even funnier how everyone mocks it when we are where we are basically by using data

 

[tunstallhill]

If the Premier league table was set according to this xG silliness, that pundits are always mentioning

literally not one person has suggested this

 

[eejit]

I love how worked up and confused people on here get about it. It will normally give a rough idea of where you should but of course there will always be outliers. It’s just a bit of data that can be used. I think it’s even funnier how everyone mocks it when we are where we are basically by using data

We should go back to old school scouting imho.

Sign people based on whether they look up for it and wear gloves etc.

 

[Bamber]

As has been said It’s obviously a very useful metric which is used my coaching teams to identify and hopefully remedy issues. But it’s one of many metrics used.

Sky Sports have always used alternate tables, “here’s the table of goals scored from 60 mins plus” “the VARless table” “the league table based on goals from set pieces” etc. So for them to present xG tables isn’t exactly a massive new idea.

 

[brains]

I don't think it's nonsense. Teams with a high Xg tend to be much better to watch and play 'more attractive' football.

Points on the board are more important though and wins can win you trophies. Creating the most chances isn't going to win you much on it's own though.

 

[Lucky1985]

It's got big flaws, doesn't legilsate for the state the game is in for example. You'd expect Leeds to have had more yesterday for example when they're chasing to get back into a game and got their tails up a bit after scoring.

But how many clear cut chances did they create?

There are flaws to it, but I don’t agree with your view on it; it a tea has their tail up then they should be taking more shots at goal, which increases the probability that they will score.

The biggest flaw to me is that opportunities like the Geertruida one yesterday don’t count because we didn’t shoot. I would have thought if he had pulled the trigger then it would have Carried a relatively high Xg, but because he bottled it and passed then it counts as an Xg of 0. If we create that chance for any other midfielder then it’s likely a shot on target at least.

 

[StayPuft]

Stats can be useful when recruiting players and scouting networks. Fulham's owner owns a statistics company that is used across their sporting teams

 

[AC Mack]

But how many clear cut chances did they create?

There are flaws to it, but I don’t agree with your view on it; it a tea has their tail up then they should be taking more shots at goal, which increases the probability that they will score.

The biggest flaw to me is that opportunities like the Geertruida one yesterday don’t count because we didn’t shoot. I would have thought if he had pulled the trigger then it would have Carried a relatively high Xg, but because he bottled it and passed then it counts as an Xg of 0. If we create that chance for any other midfielder then it’s likely a shot on target at least.

If Geertruida had shot the xG would have been low - he was outside the area and there were defenders between him and goal.

Stats can be useful when recruiting players and scouting networks. Fulham's owner owns a statistics company that is used across their sporting teams

Also see Brentford and Brighton. Forward thinking, successful clubs.

Some people still prefer to quote Brian Clough.

 

[Jyoosi]

It's a stat for people who love shite like generative AI

 

[Cuddles]

It's a useful stat to an extent but overused. Leeds fans banging on about Xg yesterday but they had 17 shots to our 8 in total and 6 to our 3 on target. Because they had more shots they had a higher Xg but it doesn't distinguish the quality of the chance. It also doesn't distinguish the players individually quality. Phil Foden is far more likely to ping one from range than most players.

I remember at least 3 of Leeds shots on target being tame efforts from outside the box that Roefs saved easily. When you break it down to good chances we were even both teams scored then they had 2 really good chances one which Hume blocked and one which Bogle mis controlled. Equally, we hit the bar and had Brobbey through on goal in the first half.

Theres been a lot of times this season I think we've been guilty of playing one pass too many rather than shoot and I think that's reflected in our poor Xg. Geertruijda is the prime example yesterday. Running onto a nicely weighted ball on the edge of the box and he tries to thread the needle with a pass rather than hit it.

Villa are the prime example of not trusting Xg. I watched an analysis on that very subject with them and its really eye opening. Basically it's clear Emery recognized they had quality ball strikers in his side and they were being given space and time outside the box, so he tells them to have a go as teams are trying to stay compact and block off gaps in the box.

The fact is better players require less high quality chances to score from and top players almost always outscore their Xg when playing well.

Look at Calvert Lewin, very good player yes, but when you look at where he scores from he only really scores when the ball comes to him between the posts from close range. Compare that with a top striker like Kane and they'll score far harder chances from more difficult angles and further out.

 

[John Byrne 92Crew]

If Geertruida had shot the xG would have been low - he was outside the area and there were defenders between him and goal.


Also see Brentford and Brighton. Forward thinking, successful clubs.

Some people still prefer to quote Brian Clough.

But if Geertruida's shot had taken a deflection or had been saved, that could have led to a chance with a higher xg? Does the fact that some teams may be playing for percentages a bit more these days (set plays/long throws etc)make a difference to how their xg looks? (I've no idea btw, genuine question).

 

[viccarlton]

I love how worked up and confused people on here get about it. It will normally give a rough idea of where you should but of course there will always be outliers. It’s just a bit of data that can be used. I think it’s even funnier how everyone mocks it when we are where we are basically by using data

We are where we are by doing the basics of football, try and prevent the opposition scoring and scoring more goals than them, not an X or a G in sight.

 

[AC Mack]

But if Geertruida's shot had taken a deflection or had been saved, that could have led to a chance with a higher xg? Does the fact that some teams may be playing for percentages a bit more these days (set plays/long throws etc)make a difference to how their xg looks? (I've no idea btw, genuine question).

I agree, it's one of the areas I'm quite sceptical over xG. The fact that you can have a shot, then another one and even another and the xG registered will be different to if that first shot went in (or wide or whatever).

We are where we are by doing the basics of football

We wouldn't have the players we have if it wasn't for data.

 

[Teejay]

Good Xg is like averaging 105 at darts but only 10% on the doubles . Or hitting 18 greens in regulation but taking 54 putts.

 

[fireballs]

It's a useful stat to an extent but overused. Leeds fans banging on about Xg yesterday but they had 17 shots to our 8 in total and 6 to our 3 on target. Because they had more shots they had a higher Xg but it doesn't distinguish the quality of the chance. It also doesn't distinguish the players individually quality. Phil Foden is far more likely to ping one from range than most players.

I remember at least 3 of Leeds shots on target being tame efforts from outside the box that Roefs saved easily. When you break it down to good chances we were even both teams scored then they had 2 really good chances one which Hume blocked and one which Bogle mis controlled. Equally, we hit the bar and had Brobbey through on goal in the first half.

Theres been a lot of times this season I think we've been guilty of playing one pass too many rather than shoot and I think that's reflected in our poor Xg. Geertruijda is the prime example yesterday. Running onto a nicely weighted ball on the edge of the box and he tries to thread the needle with a pass rather than hit it.

Villa are the prime example of not trusting Xg. I watched an analysis on that very subject with them and its really eye opening. Basically it's clear Emery recognized they had quality ball strikers in his side and they were being given space and time outside the box, so he tells them to have a go as teams are trying to stay compact and block off gaps in the box.

The fact is better players require less high quality chances to score from and top players almost always outscore their Xg when playing well.

Look at Calvert Lewin, very good player yes, but when you look at where he scores from he only really scores when the ball comes to him between the posts from close range. Compare that with a top striker like Kane and they'll score far harder chances from more difficult angles and further out.

Agree with a lot of that but xG does distinguish the quality of the chance doesn't it?


Regardless, I think it's fairly obvious from watching us we're never going to top an xG table. Only Everton and the bottom 4 have scored less than us but we have the third best defensive record up to now behind city and arsenal.

That link someone put up earlier suggests we're over performing both xG for and against. Whilst it's not "luck" the whole point of statistics is to look at longer term trends. All these do is suggest we'll drop down the table eventually.

 

[Lucky1985]

I agree, it's one of the areas I'm quite sceptical over xG. The fact that you can have a shot, then another one and even another and the xG registered will be different to if that first shot went in (or wide or whatever).


We wouldn't have the players we have if it wasn't for data.

Aye that’s another point. If a striker has his penalty saved and the keeper parries it straight to his feet, but he hits it wide, then those two chances would add up to an Xg of more than 1, but you can only score one goal from the situation, he can’t score the pen and the rebound.

 

[John Byrne 92Crew]

I agree, it's one of the areas I'm quite sceptical over xG. The fact that you can have a shot, then another one and even another and the xG registered will be different to if that first shot went in (or wide or whatever).


We wouldn't have the players we have if it wasn't for data.

Yeah, it doesn't sound like the be all and end all tbh does it, I think a Boro fan on here last year said that points on the board were "just noise" and their underlying xg stats would see them ahead of us at the end of the season. Obviously still very useful as part of a wider range of metrics though rather than taken in isolation.

 

[Cuddles]

Agree with a lot of that but xG does distinguish the quality of the chance doesn't it?


Regardless, I think it's fairly obvious from watching us we're never going to top an xG table. Only Everton and the bottom 4 have scored less than us but we have the third best defensive record up to now behind city and arsenal.

That link someone put up earlier suggests we're over performing both xG for and against. Whilst it's not "luck" the whole point of statistics is to look at longer term trends. All these do is suggest we'll drop down the table eventually.

Sorry should have clarified. Yes Xg is based on chance quality but the overall Xg number you see is just every chances Xg added together. A team could have have an Xg of 2 because they had 30 shots on goal but all of them from distance. Another team could have an Xg of 0.8 but create 2 really decent chances. Some people would say they team with the Xg of 2 deserved to win but it doesn't tell the whole story.

Typically yes better teams have higher Xg but you could look at us and say maybe we just have better finishers than the likes of Leeds and they don't need as many or as high quality a chance to score.

Equally, you could say we have a brilliant keeper who can make more stops and defenders who read the game better and make good blocks.

 

[halfpast3]

We are where we are by doing the basics of football, try and prevent the opposition scoring and scoring more goals than them, not an X or a G in sight.

Yeah that’s how all our recruitment is done mate. The whole set up they have built over the last 4 years , all the players they have brought in. By just saying do the basics of football