• This is the new server

xG

xG

OG's are not counted in xG at all. So 0.0.

Weird game though! I can't see Oldham getting 5 xG in a game ? I do think they'll be a National League side next year!
 

xG

I never really see the point of this. Ok we should be higher in the table but in reality we are not it’s actual goals that count rather than expect to score, is it just me?
 

xG

Cbee said:
I never really see the point of this. Ok we should be higher in the table but in reality we are not it’s actual goals that count rather than expect to score, is it just me?
Click to expand...
There are 3 pages of discussion and debate on this thread which would probably address your question...
 

xG

I love it, takes away the emotion from how you see a game. I’d say that XG is about right, we’ve thrown away quite a few points it feels from poor finishing.
 

xG

Cbee said:
I never really see the point of this. Ok we should be higher in the table but in reality we are not it’s actual goals that count rather than expect to score, is it just me?
Click to expand...
It gives an indication of performance. Over the short term you may see a bigger difference between xg and actual, over the longer run they will likely match. Perform the same xg as that over say 1000 games and we will do well in the league.
 

xG

Cbee said:
I never really see the point of this. Ok we should be higher in the table but in reality we are not it’s actual goals that count rather than expect to score, is it just me?
Click to expand...
Depends what you use it for. XG is a reasonable indicator of underlying quality. The league table obviously tells you actual results. But if you’re planning long term you can knee jerk into unnecessary change if results are worse than performance. Or you can ignore real issues with the players and managed because the results have been good.

It’s the sort of thing that stops TF getting fired after losing 8 of his first 10 games in charge and if ignored makes other teams bin their manager and buy a load of new players when they would probably have turned things round soon anyway (eg Smith at Villa- they’re spending a load to support Gerrard but he’s hardly got them zooming up the table and the biggest change has been players like Buendia starting to gel).
 

xG

xG

RAF_Patto said:
Brighton were very similar last season, finished 16th but their xG had them in the top 10.

If things click into place then we'll do very well, we just need to get that consistency and better conversion rates!
Click to expand...
Yep, Brighton historically similar to us re underperforming against xG.

I may have already made this point somewhere in this thread, but I really think one of the reasons we've scored fewer than expected is because our big chances keep falling to the wrong players! The United game was another example where, if Toney or Mbeumo had the chances that Jensen, Norgaard and Sorensen had, we'd have won the game by half time!
 

xG

maddogging said:
I love it, takes away the emotion from how you see a game. I’d say that XG is about right, we’ve thrown away quite a few points it feels from poor finishing.
Click to expand...
And losing to worldies (e.g. Tielemans goal, Chilwell’s was pretty good too etc)
 

xG

I remember the season before we sold Maupay he pretty much matched his xg. So going by that he was a bang average finisher.
 

xG

Does sound right! As much as they controlled the second half, Raya wasn't tested until the end. Good signs though just need to capitalise.
 

xG

Rob-BEE said:
Interesting! I wasn't able to go today, but reading through the comments I assumed we created next to nothing?

1.91 xG is pretty good, we must have had at least a couple of very good chances and probably should've won?
Click to expand...
I have no real knowledge of how they ‘score’ chances vs the XG but from the top of my head we had a handful of chances of varying levels of difficulty from which we could/should have scored from:
- Mbuemo 1 on 1 from Canos through ball
- Wissa 2 separate chances inside the six yard box which somehow went over the bar and wide
- Canos header from a cross from the left
- Janelt half volley following a bit of head tennis in the box
- Rico blocked shot from the left hand corner of the box

From a Palace perspective they had the 1 on 1 near the end of the first half, following Norgaards misplaced pass, that they blasted over.

Cant really think of any clear cut chances for either team second half (had to leave on 92 mins so might have missed some in the last 90 seconds).
 

xG

That xG Philosophy account is the fraud that worked as a football observer at Smartodds for £10/day.
 

xG

I think we managed to break xG today!

Our top 3 chances:

- 0.67 - Wissa on 28 minutes. I'll have to look at the replays of this but from where I was sitting it didn't "feel" like a 67% chance (I know xG doesn't care about my feelings!).
- 0.56 - Canós on 41 minutes. The header from Mbeumo's cross that was behind him and ended up closer to the corner flag than the goal. 56% chance of going in? Do me a favour!
- 0.40 - Mbeumo on 26 minutes. Canós played a delightful through-ball but Mbeumo was only able to stretch and toke-poke it right at the goalkeeper who had the goal completely covered. I had a perfect view of this and the only chance of Mbeumo scoring was if it managed to squirm through the goalkeeper's legs.

(source: Brentford 0 - 0 Crystal Palace (February 12 2022) | EPL | 2021/2022 | xG | Understat.com)
 

xG

BuzzBeeSi said:
I think we managed to break xG today!

Our top 3 chances:

- 0.67 - Wissa on 28 minutes. I'll have to look at the replays of this but from where I was sitting it didn't "feel" like a 67% chance (I know xG doesn't care about my feelings!).
- 0.56 - Canós on 41 minutes. The header from Mbeumo's cross that was behind him and ended up closer to the corner flag than the goal. 56% chance of going in? Do me a favour!
- 0.40 - Mbeumo on 26 minutes. Canós played a delightful through-ball but Mbeumo was only able to stretch and toke-poke it right at the goalkeeper who had the goal completely covered. I had a perfect view of this and the only chance of Mbeumo scoring was if it managed to squirm through the goalkeeper's legs.

(source: Brentford 0 - 0 Crystal Palace (February 12 2022) | EPL | 2021/2022 | xG | Understat.com)
Click to expand...
?Spot on.
Wissa "chance" didn't even make Sky's 3 minute highlights passage.......and I thought it should have been a corner anway......
 

xG

I was a little surprised by today's xG but did feel we created 2-3 excellent first half chances despite not having the ball. I also felt Palace created very little despite their possession. If we can get Toney back and a midfield three of Norgaard behind Eriksen and JDS hopefully we'll start seeing xG converted into real goals and points.

Reading the comment above about Brighton last season finishing 16th despite being 10th on xG... hopefully that's us come the end of this season!
 
  • Like
Reactions: AB

xG

In terms of location of chance it may rate them higher than they were. Canos' header from Rico's cross may have ranked high but he was leaning back to contact it and it would never go in (probably should have gone for the volley). Mbeumo maybe for the one on one, but didn't get enough and was probably too close to the keeper.

We certainly would have won the xG battle as they created very little despite controlling lots of the second half. But we needed to score in that first half you feel .
 

xG

BuzzBeeSi said:
I think we managed to break xG today!

Our top 3 chances:

- 0.67 - Wissa on 28 minutes. I'll have to look at the replays of this but from where I was sitting it didn't "feel" like a 67% chance (I know xG doesn't care about my feelings!).
- 0.56 - Canós on 41 minutes. The header from Mbeumo's cross that was behind him and ended up closer to the corner flag than the goal. 56% chance of going in? Do me a favour!
- 0.40 - Mbeumo on 26 minutes. Canós played a delightful through-ball but Mbeumo was only able to stretch and toke-poke it right at the goalkeeper who had the goal completely covered. I had a perfect view of this and the only chance of Mbeumo scoring was if it managed to squirm through the goalkeeper's legs.

(source: Brentford 0 - 0 Crystal Palace (February 12 2022) | EPL | 2021/2022 | xG | Understat.com)
Click to expand...
Interesting breakdown. I don’t even remember Wissa’s chance but Canos’ header and Bryan’s toe poke both looked like difficult chances to me. Seems like there’s a lot of weight in getting a touch close to goal with very little context factored in
 

xG

Does XG factor in a side's historic inability to hit a cow's arse? Serious question.
 

xG

Alex said:
Depends if they are holding the tail or not.
Click to expand...
:ROFLMAO: Is this a metaphor for how easy the opportunity is. Is this also factored in? Sorry, I know not a lot about this thing.

My original question should maybe have been about individuals. If player x (or Mattias Jensen :ROFLMAO:) has a record of squandering potential tap-ins, or if player y (or Bryan Mbeumo ;)) has a history of striking the woodwork instead of the net, is this factored into the XG calculation?
 

xG

Lionel Bart-At said:
:ROFLMAO: Is this a metaphor for how easy the opportunity is. Is this also factored in? Sorry, I know not a lot about this thing.

My original question should maybe have been about individuals. If player x (or Mattias Jensen :ROFLMAO:) has a record of squandering potential tap-ins, or if player y (or Bryan Mbeumo ;)) has a history of striking the woodwork instead of the net, is this factored into the XG calculation?
Click to expand...
No it isn't, to answer your question! E.g. the xG of a penalty is fixed, regardless of whether it's Ivan Toney taking it, or Pontus Jansson.
 

xG

DDVonHammer said:
View attachment 34871
We are picking up right where we left off last season. Someone make a sacrifice to the xG gods before Saturday, please.
Click to expand...
I thought Tuesday night was another good example of our approach to xG. I think all our shots were from inside the box, whilst almost all of Palace's were outside. In other words, we restrict the opposition to low probability chances whereas in attack we favour higher % chances rather than taking long range shots.

Probably goes part way to explaining why we score very few worldies, whilst at the other end we seem to be a magnet for conceding them!
 

xG

Rob-BEE said:
I thought Tuesday night was another good example of our approach to xG. I think all our shots were from inside the box, whilst almost all of Palace's were outside. In other words, we restrict the opposition to low probability chances whereas in attack we favour higher % chances rather than taking long range shots.

Probably goes part way to explaining why we score very few worldies, whilst at the other end we seem to be a magnet for conceding them!
Click to expand...
It also makes us as fans feel like we’re bad at defending set pieces because pretty much the only close range goals we concede are from corners.
 

xG

Lionel Bart-At said:
:ROFLMAO: Is this a metaphor for how easy the opportunity is. Is this also factored in? Sorry, I know not a lot about this thing.

My original question should maybe have been about individuals. If player x (or Mattias Jensen :ROFLMAO:) has a record of squandering potential tap-ins, or if player y (or Bryan Mbeumo ;)) has a history of striking the woodwork instead of the net, is this factored into the XG calculation?
Click to expand...
Say player y has 5 shots at goal each with an xG of 0.4. Therefore he has an xG of 2.
Suppose he hits the post every time, he has actual goals of 0.

The stats team sees this and thinks y gets into a lot of good positions and didn't get the goals he deserved. Could be undervalued.

They then have to decide whether he'll get the breaks in future and live up to his xG or instead it's a cow's arse/banjo scenario.
 

xG

Rob-BEE said:
No it isn't, to answer your question! E.g. the xG of a penalty is fixed, regardless of whether it's Ivan Toney taking it, or Pontus Jansson.
Click to expand...
OK thanks. So if all our four chances in a given match were open goal opportunities for Pontus, this would rank equally to all those chances falling to Ivan.

If all four were missed, would the xg rating be the same as if they were converted? And will the fact that these chances were missed affect the team's (rather than the individual's) weighting for similar chances in the future?

And what are the different weightings for different types of chances? Presumably this depends on distance from goal, angle, number of opposition players in proximity, dead ball or not etc.?

Again, apologies for being a dinosaur and/or for not reading through the entire thread to find these answers.
 

xG

Lionel Bart-At said:
OK thanks. So if all our four chances in a given match were open goal opportunities for Pontus, this would rank equally to all those chances falling to Ivan.

If all four were missed, would the xg rating be the same as if they were converted? And will the fact that these chances were missed affect the team's (rather than the individual's) weighting for similar chances in the future?

And what are the different weightings for different types of chances? Presumably this depends on distance from goal, angle, number of opposition players in proximity, dead ball or not etc.?

Again, apologies for being a dinosaur and/or for not reading through the entire thread to find these answers.
Click to expand...
I'm pretty sure xG isn't specific to either a team or a player, so none of those factors would influence future xG weightings.

As to the different weightings, I think it's pretty complicated and different sources use different factors, which is why you sometimes see conflicting xG stats. But as far as I know, all those things you listed would be taken into account.

One big fallacy though (I think!) is that a player has to take a shot for it to register as an xG. So, let's say if someone is through on goal 1v1 but decides to pass to a teammate and misses the pass it would be 0 xG. But if they'd taken a shot it might have been, say, 0.5 xG. Even though essentially the chance is the same in both cases. Not that anyone is pretending xG is a perfect tool...
 

xG

Lionel Bart-At said:
OK thanks. So if all our four chances in a given match were open goal opportunities for Pontus, this would rank equally to all those chances falling to Ivan.

If all four were missed, would the xg rating be the same as if they were converted? And will the fact that these chances were missed affect the team's (rather than the individual's) weighting for similar chances in the future?

And what are the different weightings for different types of chances? Presumably this depends on distance from goal, angle, number of opposition players in proximity, dead ball or not etc.?

Again, apologies for being a dinosaur and/or for not reading through the entire thread to find these answers.
Click to expand...
The Pontus v Ivan thing is how we can see those who under- and over-perform on the basis of xG. A CB will typically underperform, a striker overperform. Across the whole population of players and chances over thousands of games you get a par score which is the xG.

It’s why teams like us set up to do stuff like not shoot from distance except when we have a JDS or Said who can regularly outperform from low xG situations (and why Jensen almost always passes when in a position a Tielemans might shoot, whereas he’d have had a go 3 years ago). It also explains why in extremis it isn’t bonkers to shove Goode up top because if you’re aiming balls right into the 6 yard box they’re so high xG that even CBs have a good chance of converting and need to be defended against, distracting from defending against the other more skilled attackers. Mee’s effort v Palace and his goal v Man U would have only been negligibly better executed by a specialist striker.
 

xG

AB said:
Mee’s effort v Palace and his goal v Man U would have only been negligibly better executed by a specialist striker.
Click to expand...
I would take Mee over Mbuemo for those sorts of chances!
 

xG

Rob-BEE said:
I'm pretty sure xG isn't specific to either a team or a player, so none of those factors would influence future xG weightings.

As to the different weightings, I think it's pretty complicated and different sources use different factors, which is why you sometimes see conflicting xG stats. But as far as I know, all those things you listed would be taken into account.

One big fallacy though (I think!) is that a player has to take a shot for it to register as an xG. So, let's say if someone is through on goal 1v1 but decides to pass to a teammate and misses the pass it would be 0 xG. But if they'd taken a shot it might have been, say, 0.5 xG. Even though essentially the chance is the same in both cases. Not that anyone is pretending xG is a perfect tool...
Click to expand...
By fallacy I think you mean failing?

So what would the xG of Ivan mis kick from the free kick 2nd half register on xG, even though he (barely) made contact and arguably was a good chance?
 

xG

GuyBee said:
By fallacy I think you mean failing?

So what would the xG of Ivan mis kick from the free kick 2nd half register on xG, even though he (barely) made contact and arguably was a good chance?
Click to expand...
One of the basics of statistics is that aggregating a large volume of numbers means that one outlier event barely registers. If it was going to be sophisticated, you'd remove the outliers, for example using a standard deviation, before your calculation.
 

xG

GuyBee said:
By fallacy I think you mean failing?

So what would the xG of Ivan mis kick from the free kick 2nd half register on xG, even though he (barely) made contact and arguably was a good chance?
Click to expand...
Probably not the official stats but registered as xG of 0.17 on the Fotmob app
Header from a corner in the 78th minute registered as 0.04
 

xG

I’d like to see a breakdown of our xG differential in open play this season compared to last season. It *feels* like we are doing better but I’d like to see it confirmed. Before the injury bug hit last season, it felt like we were defending well and then relying on set piece opportunities to create scoring chances. Then the injuries piled up and we were just poor. Then we signed Eriksen and were finally healthy and looking somewhat dangerous in open play. It feels like we’ve improved this year, albeit in a small sample size
 
41 More Posts...
You must log in or register to reply here.

Share this page

Back
Top Bottom