[Albion] Interesting Expected Points (based on XG) graph

[Machiavelli]

According to the Premier League website Murray conversion rate is around 43% by comparison Sergio Aguero is just 14%, Salah is 12%, Kane 16%, Aubameyang 32% and Hazard 26%.

He's got to be one of, if not the most clinical strikers in the PL based on that stat.

Wow. I'm stunned by that Aguero stat and, to a lesser extent, Kane.

 

[Joey Jo Jo Jr. Shabadoo]

Wow. I'm stunned by that Aguero stat and, to a lesser extent, Kane.

Aguero has taken 51 shots so far this season for his 7 goals, Murray has had just 14 shots for his 6 goals.

 

[SussexSeahawk]

this just further goes to show what a complete load of tosh all these XG and other stats really are. Its a case of emperors new clothes. Its on TV , its on some video game, some pundit mentioned it so it must be important. Its primarily used to give commentators more crap to espouse on TV and betting. companies some reason to give odds. Its actually completely meaningless in the real world

They've been shown to not be complete crap. The whole point of xG is that it is a better indicator of future form than actual points.

The amount of chances you generally concede or create are much more important long term traits than whether you happen to win. You don't have to pay attention to the more refined stats like xG if you like, but it is a genuinely useful tool.

 

[dazzer6666]

According to the Premier League website Murray conversion rate is around 43% by comparison Sergio Aguero is just 14%, Salah is 12%, Kane 16%, Aubameyang 32% and Hazard 26%.

He's got to be one of, if not the most clinical strikers in the PL based on that stat.

Thanks.... That's it.

Sent from my H8314 using Tapatalk

 

[The Antikythera Mechanism]

So I take it from that graph that the Guardian expected Brighton to currently be bottom and Burnley second from bottom, well that doesn't seem right.

Just what’s to be expected from the newspaper with the lowest circulation in the country. #scrapingthebarrel

 

[Ⓩ-Ⓐ-Ⓜ-Ⓞ-Ⓡ-Ⓐ]

Wolves below what they expected . Not having that . . .. unless its what their fans expected , in which case they wouldn't of expected to lose a point.

No! It's bases on statistical analysis. Here is a video explanation of xG which is what was used for the Expected Points calculations that provided the data set for the graph in the OP.

[yt]w7zPZsLGK18[/yt]

 

[Uter]

Aguero has taken 51 shots so far this season for his 7 goals, Murray has had just 14 shots for his 6 goals.

And there's more to his game than that. He's chipping in defensively too.

 

[hans kraay fan club]

Apologies if fixtures, but found this graph plotting Expected Points* against Actual Points.

You can look it it with a glass half empty or glass half full approach I suppose. It does underline the fact that we create very few chances in games which you'd think, in terms of points, would even over itself out over the course of the season - slightly concerning! We're overachieving now - how long can we sustain it?

View attachment 101923
https://www.theguardian.com/footbal...overachieving-the-data-suggests-they-could-be

*Expected points are calculated through expected goals, another advanced metric that’s made its way around the game in recent years, and has been the centre of much contentious debate over its use.Expected goals — or xG — assigns a quantitative value to every shot on goal, based on the quality of the assist, angle of the shot and distance to the goal. Every shot’s xG in a game is totalled and compared against the opponent’s xG, and an “expected winner” is then established and the expected points distributed.

Expected points are calculated through expected goals, another advanced metric that’s made its way around the game in recent years, and has been the centre of much contentious debate over its use.Expected goals — or xG — assigns a quantitative value to every shot on goal, based on the quality of the assist, angle of the shot and distance to the goal. Every shot’s xG in a game is totalled and compared against the opponent’s xG, and an “expected winner” is then established and the expected points distributed.

They've been shown to not be complete crap. The whole point of xG is that it is a better indicator of future form than actual points.

The amount of chances you generally concede or create are much more important long term traits than whether you happen to win. You don't have to pay attention to the more refined stats like xG if you like, but it is a genuinely useful tool.

Sorry, but it is HOPELESSLY flawed, to the edge of pointlessness.

It makes ZERO allowance for the state of the respective games. In that run of three 1-0 wins for example, for a very large period of each game, one side (us as it happens) was ahead in the game, and attempting to see it out -making little effort to create further chances, but instead seeking to bed in and stifle the opposition.

It goes without saying that in those lengthy periods of those matches that opposition are going to rack up some chances, and boost their 'XG'. Effectively you're suggesting that our going ahead in three consecutive matches is an indicator of future poor form.

It's statistical bullshit.

 

[Johnny RoastBeef]

Sorry, but it is HOPELESSLY flawed, to the edge of pointlessness.

It makes ZERO allowance for the state of the respective games. In that run of three 1-0 wins for example, for a very large period of each game, one side (us as it happens) was ahead in the game, and attempting to see it out -making little effort to create further chances, but instead seeking to bed in and stifle the opposition.

It goes without saying that in those lengthy periods of those matches that opposition are going to rack up some chances, and boost their 'XG'. Effectively you're suggesting that our going ahead in three consecutive matches is an indicator of future poor form.

It's statistical bullshit.

xG is a key tool used by bookmakers in their statistical analysis and Tony Bloom at Star Lizard.

Statistical models can be built using xG that can be weighted to allow for any possible in game variable, including current score or even which player the chance fell to and that players individual xG.

The graph seems to be using xG in its most basic form, but you cant deny that we don't create many chances, regardless of if we are winning or not.

 

[hans kraay fan club]

xG is a key tool used by bookmakers in their statistical analysis and Tony Bloom at Star Lizard.

Statistical models can be built using xG that can be weighted to allow for any possible in game variable, including current score or even which player the chance fell to and that players individual xG.

The graph seems to be using xG in its most basic form, but you cant deny that we don't create many chances, regardless of if we are winning or not.

You miss the point entirely, I think.

xG alone is vaguely interesting - it tells you not just how many chances were created, but factors in the quality of those chances. Agree with that much.

What this graph misses is the state of games. Its plainly obvious that a (low ranked) side holding a lead is going to be pushing to create further chances, less than a side trailing by a goal. They've done so by design - they've chosen to bed in and stay compact. It really doesn't tell you much about future form at all.

 

[CaptainDaveUK]

Nope, this graph is not based on opinion or fixtures but on statistical analysis based on xG. This was posted above, but incase you missed the explanation:

Expected points are calculated through expected goals, another advanced metric that’s made its way around the game in recent years, and has been the centre of much contentious debate over its use.Expected goals — or xG — assigns a quantitative value to every shot on goal, based on the quality of the assist, angle of the shot and distance to the goal. Every shot’s xG in a game is totalled and compared against the opponent’s xG, and an “expected winner” is then established and the expected points distributed.

I realised that and is why I stated that it didn't seem right. xG currently needs to be taken with a huge pinch of salt as predicting that Brighton and Burnley would currently be bottom of the league clearly shows that it is still very much a work in progress.

 

[Timbo]

We have 14 points from 11 games.

 

[Jim in the West]

Surely you have to factor in the quality of the striker, somehow. As has been pointed out, Murray has a fantastic conversion rate. Hence, whilst we may have a low xG score, if those chances happen to fall at the feet of someone with a very high conversion rate, then we will get more goals (and points) than one might "expect". Surely that is the lesson from this season so far? And logically, there is no reason why it shouldn't continue to be the case.

Apologies if I've misunderstood - or simply repeated what others have effectively been saying.

 

[Timbo]

Apologies if fixtures, but found this graph plotting Expected Points* against Actual Points.

You can look it it with a glass half empty or glass half full approach I suppose. It does underline the fact that we create very few chances in games which you'd think, in terms of points, would even over itself out over the course of the season - slightly concerning! We're overachieving now - how long can we sustain it?

View attachment 101923
https://www.theguardian.com/footbal...overachieving-the-data-suggests-they-could-be

*Expected points are calculated through expected goals, another advanced metric that’s made its way around the game in recent years, and has been the centre of much contentious debate over its use.Expected goals — or xG — assigns a quantitative value to every shot on goal, based on the quality of the assist, angle of the shot and distance to the goal. Every shot’s xG in a game is totalled and compared against the opponent’s xG, and an “expected winner” is then established and the expected points distributed.

What do the winners get at the end of the season?

 

[Finchley Seagull]

Apologies if fixtures, but found this graph plotting Expected Points* against Actual Points.

You can look it it with a glass half empty or glass half full approach I suppose. It does underline the fact that we create very few chances in games which you'd think, in terms of points, would even over itself out over the course of the season - slightly concerning! We're overachieving now - how long can we sustain it?

View attachment 101923
https://www.theguardian.com/footbal...overachieving-the-data-suggests-they-could-be

*Expected points are calculated through expected goals, another advanced metric that’s made its way around the game in recent years, and has been the centre of much contentious debate over its use.Expected goals — or xG — assigns a quantitative value to every shot on goal, based on the quality of the assist, angle of the shot and distance to the goal. Every shot’s xG in a game is totalled and compared against the opponent’s xG, and an “expected winner” is then established and the expected points distributed.

I get why this can be interesting but surely for this to even itself out over the course of the season, as you suggest, every team would have to have the same quality of goalkeeper and a striker with the same conversion rate. As others have said, we have a high quality keeper and striker so it doesn't say anything except we don't create a lot of chances, which we all knew already

 

[DavidinSouthampton]

Yeah. Lefty fake news peddling Guardian. They ought to be ashamed of themselves.

……... Which is why there was a poll published the other day which showed that the Guardian was by far the most trusted newspaper in the country across all age groups by a long way.

 

[dazzer6666]

……... Which is why there was a poll published the other day which showed that the Guardian was by far the most trusted newspaper in the country across all age groups by a long way.

Poll of Guardian readers, commissioned by the Guardian and published in the Guardian [emoji106]

 

[DavidinSouthampton]

Poll of Guardian readers, commissioned by the Guardian and published in the Guardian [emoji106]

That would be the only way to get a sensible answer.....

 

[dazzer6666]

That would be the only way to get a sensible answer.....

[emoji23][emoji23][emoji23][emoji23][emoji122][emoji122][emoji122][emoji122][emoji122]

 

[The Antikythera Mechanism]

https://www.statista.com/statistics/529060/uk-newspaper-market-by-circulation/