So Blackburn are top of the table yet have lowest XG

[cambridgeshire canary 6,716]

And we are bottom of the table with the highest XG. So uh, does XG even matter?🤔

[TheRock 171]

As shown Hull's first goal on saturday, xG is stupid as a team can be suffocated of the ball most of the match but yet win a cheap turnover and score totally against the run of play.

 

Its a forecast, not a statistic, which in my opinion was simply thought up to encourage more gambling.

[AJ 1,218]

6 minutes ago, cambridgeshire canary said:

And we are bottom of the table with the highest XG. So uh, does XG even matter?🤔

Long term, yes it does as it is a direct result of how many chances you are creating in theory. It means Blackburn are currently over performing, and we are underperforming. End of the season if the XG stays the same we should finish above Blackburn

[Nexus_Canary 1,016]

Xg is for nerds 

All that counts is the amount of times the ball hits the onion bag 

Edited August 16, 2022 by Nexus_Canary

[Jim Smith 2,317]

Of course its points that ultimately matter but what it does do is give a reasonable indication of whether a current situation is sustainable longer term if things continue as they are. Having a low xG for may not matter too much for Blackburn if they have a striker who has the most clinical season of his life or they have a low xG against as well.

But what the current xG data does suggest is that its highly unlikely that we will be bottom in a few weeks time but also that Blackburn also won't sustain their position at the top. Or to put it another way, we are both in slightly false positions based on the quality of the play we are producing. Time will tell.

[Hank shoots Skyler 2,094]

18 minutes ago, cambridgeshire canary said:

And we are bottom of the table with the highest XG. So uh, does XG even matter?🤔

Yes, clearly. 

Sure some teams will outperform and others will underperform even over the course of the season (as this can be a result of goalkeeper brilliance / howlers or the finishing of the attackers) but we are probably talking a 10-20% variance here and all teams will yield points with a clear trend alongside their xG performance. You can pick a joker out of a deck once or maybe even twice but you can't consistently keep on doing it. 

3 games is a tiny sample size so shows a skew, but there is absolutely no chance that come the end of the season the team with the highest xG points is bottom, and lowest is top.

 

Edited August 16, 2022 by Hank shoots Skyler

[Wacky Waving Inflatable Arm Flailing Tube Man 3,820]

Generally speaking, over the course of the season, the table for net xG (xG for minus xG against) will be very similar to the actual table. 

I think last season in the Premier League, the only real outlier was Wolves, who finished several places higher in the real table than their xG suggested they would. 

[Mason 47 1,379]

It's not the statistic that is important but the analysis. 

I'd take it to suggest a couple of things for Blackburn. Either they've been fortunate to score low probability chances or they have players that negate the difficulty of those chances (may well be the case, with Brereton in the side). 

It's all about interpreting the data to make suggestions, rather than black-and-white this-means-that.

 

[Monty13 2,233]

7 minutes ago, Wacky Waving Inflatable Arm Flailing Tube Man said:

Generally speaking, over the course of the season, the table for net xG (xG for minus xG against) will be very similar to the actual table. 

I think last season in the Premier League, the only real outlier was Wolves, who finished several places higher in the real table than their xG suggested they would. 

It’s almost like it’s very hard to read much into 3 games of a 46 game season.

[Don’t be Krul 399]

I find all these XG stats etc. a bit OTT. They measure distance run, go get Mo Farah. They mention sprints, go get Linford Christie. Strength, go get Geoff Capes. All that matters at the end of the day is who scores the most, something XG or no XG aside were pretty cr@p at

[Badger 2,405]

🤦‍♂️

[hogesar 9,667]

59 minutes ago, Jim Smith said:

Of course its points that ultimately matter but what it does do is give a reasonable indication of whether a current situation is sustainable longer term if things continue as they are. Having a low xG for may not matter too much for Blackburn if they have a striker who has the most clinical season of his life or they have a low xG against as well.

But what the current xG data does suggest is that its highly unlikely that we will be bottom in a few weeks time but also that Blackburn also won't sustain their position at the top. Or to put it another way, we are both in slightly false positions based on the quality of the play we are producing. Time will tell.

Wasn't it Blackburn who started another season, one of our championship title wins, with a ridiculously good start but their xG was incredibly low? I remember the highlights show making out like they were title contenders but they ended up finishing midtable as their xG suggested they would...

[Nuff Said 5,137]

I think it was our first promotion season under Farke that Reading were for a month or two the runaway surprise at, or near, the top of the table. They were far out-performing their xG and we were looking good stats-wise but not getting results. We all know how that ended.

[TIL 1010 4,744]

The thread title says all you need to know.

[Jim Smith 2,317]

22 minutes ago, Nuff Said said:

I think it was our first promotion season under Farke that Reading were for a month or two the runaway surprise at, or near, the top of the table. They were far out-performing their xG and we were looking good stats-wise but not getting results. We all know how that ended.

Yes there was a season where Reading were basically scoring from almost every shot they took for the opening couple of months.

[Ken Hairy 3,772]

Just shows exactly what a **** stat this xg **** is. The only people who use it are people who try to make the game something it's not. 

[hogesar 9,667]

38 minutes ago, Ken Hairy said:

Just shows exactly what a **** stat this xg **** is. The only people who use it are people who try to make the game something it's not. 

In which case, you lump money on Blackburn being promoted, I'll lump money on them finishing midtable. 

[Nexus_Canary 1,016]

39 minutes ago, Ken Hairy said:

Just shows exactly what a **** stat this xg **** is. The only people who use it are people who try to make the game something it's not. 

Its fully integrated on the FM series now and when it first dropped I was excited to see how my modern day train spotting could be improved.
Very quickly realised its a load of **** and just went back to playing the game.

Edited August 16, 2022 by Nexus_Canary

[hogesar 9,667]

1 hour ago, Nuff Said said:

I think it was our first promotion season under Farke that Reading were for a month or two the runaway surprise at, or near, the top of the table. They were far out-performing their xG and we were looking good stats-wise but not getting results. We all know how that ended.

Ah, Reading was it? That sounds familiar now.

Again, some posters on here have completely missed the point. There is no exact science but xG would suggest that Blackburn won't maintain their conversion rate and their results will drop off. It could be they just have insanely clinical players but even that's unlikely, with Messi being the only player in world football who's regularly exceeded his xG.

[norfolkngood 1,087]

22 minutes ago, hogesar said:

In which case, you lump money on Blackburn being promoted, I'll lump money on them finishing midtable. 

Blackburns XG could change or they could be under pressure for 40 games and win 1 nill with one shot on goal 

unlikely but XG will not win you the league on its own 

our XG is high but we do not score goals and we have not for a while now untill that changes we will not win football games 

[The Bunny 309]

Tiny sample sizes result in weird results. News at 11.

[king canary 7,605]

21 minutes ago, hogesar said:

Ah, Reading was it? That sounds familiar now.

Again, some posters on here have completely missed the point. There is no exact science but xG would suggest that Blackburn won't maintain their conversion rate and their results will drop off. It could be they just have insanely clinical players but even that's unlikely, with Messi being the only player in world football who's regularly exceeded his xG.

I remember we were the outliers when we started hot under Alex Neil too. Top after 12 games and I think it was Ted Knutson who mentioned we were far outperforming our xG and were likely to come back to the mean which we did. 

[Nuff Said 5,137]

44 minutes ago, hogesar said:

Ah, Reading was it? That sounds familiar now.

Again, some posters on here have completely missed the point. There is no exact science but xG would suggest that Blackburn won't maintain their conversion rate and their results will drop off. It could be they just have insanely clinical players but even that's unlikely, with Messi being the only player in world football who's regularly exceeded his xG.

New stat needed then - MxG. Messi expected Goals.
 

Perhaps also SxG - Sargent expected Goals, which is always zero?

[sonyc 5,517]

10 minutes ago, Nuff Said said:

New stat needed then - MxG. Messi expected Goals.
 

Perhaps also SxG - Sargent expected Goals, which is always zero?

Or ...

Hx(-h/c=c)G

Hanley expected goals - of headers from corners into the crowd.

Note: cue other equations relating to our players🙂

Edited August 16, 2022 by sonyc

[Mr Angry 1,555]

56 minutes ago, hogesar said:

Ah, Reading was it? That sounds familiar now.

Again, some posters on here have completely missed the point. There is no exact science but xG would suggest that Blackburn won't maintain their conversion rate and their results will drop off. It could be they just have insanely clinical players but even that's unlikely, with Messi being the only player in world football who's regularly exceeded his xG.

Not missing the point, more likely ignoring the point completely so they can carry on raging.

[Hank shoots Skyler 2,094]

1 hour ago, Ken Hairy said:

Just shows exactly what a **** stat this xg **** is. The only people who use it are people who try to make the game something it's not. 

Nah, I think the main thing xG shows is who did and didn't like maths at school... I'm going to go ahead and guess that you did not...? 

At the end of last season in the prem, the average 'expected' points total was accurate on average within 15% of teams' actual points totals. Shown below. Looks rather similar although there are some swings around midtable where points are much more congested.

So it isn't a ****** **** stat, there is a correlation with this stat and actual points, pretty accurately but of course teams over and underperform as well. This is the kind of element that makes football the best sport in the world - the aspect of luck, that you don't need to always be the better team to come away with 3 points, the fact that a single outstanding striker or goalkeeper could tip the balance in your favour etc.

You could also use stats like shots, shots on target, touches in opposition box, and even possession to a lesser extent to draw similar parallels. 

I agree that it shouldn't be used as a bible, but I can feel pretty confident in saying that we won't be bottom of the table much longer, and I'd expect us to be up to the play off spots in the coming weeks if we can keep dominating games like the last two league matches (and not shoot ourselves in the foot!). 

 

 

[hogesar 9,667]

Just now, Hank shoots Skyler said:

Nah, I think the main thing xG shows is who did and didn't like maths at school... I'm going to go ahead and guess that you did not...? 

At the end of last season in the prem, the average 'expected' points total was accurate on average within 15% of teams' actual points totals. Shown below. Looks rather similar although there are some swings around midtable where points are much more congested.

So it isn't a ****** **** stat, there is a correlation with this stat and actual points, pretty accurately but of course teams over and underperform as well. This is the kind of element that makes football the best sport in the world - the aspect of luck, that you don't need to always be the better team to come away with 3 points, the fact that a single outstanding striker or goalkeeper could tip the balance in your favour etc.

You could also use stats like shots, shots on target, touches in opposition box, and even possession to a lesser extent to draw similar parallels. 

I agree that it shouldn't be used as a bible, but I can feel pretty confident in saying that we won't be bottom of the table much longer, and I'd expect us to be up to the play off spots in the coming weeks if we can keep dominating games like the last two league matches (and not shoot ourselves in the foot!). 

 

 

Ran out of reactions, but well put, much more eloquently than I could be bothered 😄 

[Mr Angry 1,555]

1 hour ago, Nexus_Canary said:

Its fully integrated on the FM series now and when it first dropped I was excited to see how my modern day train spotting could be improved.
Very quickly realised its a load of **** and just went back to playing the game.

I’m still playing FM20, XG must have appeared after that?

[Hank shoots Skyler 2,094]

33 minutes ago, king canary said:

I remember we were the outliers when we started hot under Alex Neil too. Top after 12 games and I think it was Ted Knutson who mentioned we were far outperforming our xG and were likely to come back to the mean which we did. 

Yep, I remember doing a thread at that time saying 'worrying performances' and getting absolutely slated because we were top of the league... Feels like the reverse is true at the moment. Although the elements around individual mistakes and failure to convert chances are pretty significant concerns right now, I do feel like they will iron out, just needs an early goal tonight and the monkey is off our back!

[Clint 221]

It’s completely irrelevant. It shows we’re not taking our chances and who can say that is categorically going to change?