Match Thread [FT] Rangers 4 (Lammers 10'; Danilo 78'; Sima 84'; Dowell 90') - 0 Livingston

[Rousseau 10,489]

And God said*, “Let there be xG,” and there was xG. God saw that the xG was good, and he separated the xG from the xGA. God called the xG “Goals For,” and the xGA he called “Goals Against.” And there was attack, and there was defence—the first chance.

 

*Probably

[weebluedevil 1,239]

Okay, so what you’re saying is that, if Haaland had been presented with the same opportunity, he would more likely miss than score? 

[Rousseau 10,489]

5 minutes ago, weebluedevil said:

Okay, so what you’re saying is that, if Haaland had been presented with the same opportunity, he would more likely miss than score? 

Haaland would be more likely to outperform his xG because he is an elite striker. 

 

Sebo would be more likely to underperformed his xG because... Well.

 

The quality of the chance needs to be the same for all players. 

[weebluedevil 1,239]

15 minutes ago, Rousseau said:

Haaland would be more likely to outperform his xG because he is an elite striker. 

 

Sebo would be more likely to underperformed his xG because... Well.

 

The quality of the chance needs to be the same for all players. 

Great, we do agree 😀

[PoohBear 1,147]

On 14/08/2023 at 08:11, Rousseau said:

It's not subjective: they've added up all the times a player actually scored. It's an objective fact. It takes that, to suggest a probability of a player in the same position scoring in the future. 

 

They take hundreds of thousands of past shots, going back decades, from every conceivable position on the pitch. 

 

In practice, that means if a chance has 0.2xG, it should be scored 20% of the time, because out of those hundreds of thousands of past shots they've looked at, the player scored 20% of the time. 

 

They take into account many more variables than just position on the pitch:

• Distance to the goal
• Angle to the goal
• Did the player strike it with his feet or was it a header?
• In what passage of play did it happen? (e.g. open play, direct free-kick, corner kick, counter-attack)
• Has the player just beaten an opponent?

We do it in our heads all the time, using our judgement and past experience of watching games, when we say, 'that's a sitter - he has to score that'.

 

The xG tells you how many times players in the past have actually scored from that position - and taking into account all the other variables. 

 

It's a lot more objective than someone's judgement. Compare how many games an xG model is based on (tens of thousands of matches, hundreds of thousands of shots, going back decades) compared to how many games you or I have watched in our lifetime. It's not even close.

 

I could watch every single Rangers game in the league, so my judgement on whether a player has missed a sitter is based on 38 games a season. An xG model has covered every match of every single club in the league, and every club in the top 15 (and more) European leagues, going back decades. 

 

In that way it's objective. There's no subjectivity in it. 

Sure, but it seems like there are a number of subjective or problematic issues with the system. Has the player just beaten an opponent? That is subjective. Passage of play can be subjective. What about pitch size, number of defenders etc, these all would make a marked difference on Xg, wouldn’t they? I still have a blinding fever, covid is wank.

[Rousseau 10,489]

4 minutes ago, PoohBear said:

Sure, but it seems like there are a number of subjective or problematic issues with the system. Has the player just beaten an opponent? That is subjective. Passage of play can be subjective. What about pitch size, number of defenders etc, these all would make a marked difference on Xg, wouldn’t they? I still have a blinding fever, covid is wank.

They take into account over 35 contextual variables. 

[Tannochsidebear 2,405]

On 15/08/2023 at 19:10, Rousseau said:

I still think you're mistaken. 

 

They're only accurate if you include as many chances as possible. 

 

We can't pick and choose. It renders the metric meaningless. 

You’re starting to come round I see!

 

Seriously though I appreciate you taking the time to explain how the xG system works. I don’t buy into it because of the very way it is calculated as the debate in this thread has shown. To calculate a xG for a certain chance on an average of thousands of similar opportunities across hugely varying standards of football doesn’t tell you accurately if a Rangers player should have scored or not. For that to be accurate it would have to be calculated only against similar standards of player, not all-comers. Even then there are too many variables, with key ingredients missing like the importance of the chance in the match context, the importance of the match. If a player smashes the xG based on the above that would be hugely impressive, but if he only gets above xG when his team are winning 4-0 or in friendlies then it’s less impressive .
I accept for those who can overlook all these important aspects and just buy into it for what it is, it is a great debating point. 

[Rousseau 10,489]

10 hours ago, Tannochsidebear said:

You’re starting to come round I see!

 

Seriously though I appreciate you taking the time to explain how the xG system works. I don’t buy into it because of the very way it is calculated as the debate in this thread has shown. To calculate a xG for a certain chance on an average of thousands of similar opportunities across hugely varying standards of football doesn’t tell you accurately if a Rangers player should have scored or not. For that to be accurate it would have to be calculated only against similar standards of player, not all-comers. Even then there are too many variables, with key ingredients missing like the importance of the chance in the match context, the importance of the match. If a player smashes the xG based on the above that would be hugely impressive, but if he only gets above xG when his team are winning 4-0 or in friendlies then it’s less impressive .
I accept for those who can overlook all these important aspects and just buy into it for what it is, it is a great debating point. 

I mean, xG does describe players, teams and matches quite accurately over the season, so whether you buy into it or not is irrelevant - it works. 

 

xG does take into account game context - off the top of my head, it takes into account 35 contextual variables. 

 

For xG to be meaningful, it has to describe all-comers, to use your term. Then you can compare how players and teams are performing to the statistical average. 

[Graeme Ro55 749]

3 hours ago, Rousseau said:

I mean, xG does describe players, teams and matches quite accurately over the season, so whether you buy into it or not is irrelevant - it works. 

 

xG does take into account game context - off the top of my head, it takes into account 35 contextual variables. 

 

For xG to be meaningful, it has to describe all-comers, to use your term. Then you can compare how players and teams are performing to the statistical average. 

My tupence worth. 
XG is a well designed way to describe what a players current capability is, or what his previous perfomance levels have been concerning taking chances. The problem with this is that "fans" see other aspects of players such as style and character and project that into our expectation or memory. That causes an uncomfortable difference between our opinion of a player and his XG.