Has there been a worse home performance in recent memory?

[Fabrice Fernandes no.1 fan]

2 minutes ago, Dman said:

Fuck me. Head against a brick wall. 
 

there are many external factors. Some being; 

- Time and situation of the game - less pressure at 80 minutes when you’re 6-0 up, than it is 0-0 in a relegation battle. 

- pace and accuracy of the pass / cross (specifically for these type of chances). 

- finishing ability of the striker. Regardless of what XG says, there are some strikers who are better than others. 
 

Anyway, agree to disagree, I cba to continue to go over the same stuff. 

This is painful to read, you're arguing different things here to be honest.

XG is a good useful metric. It allows an individual chance to be given a value, by which you can measure if a team has created what should be considered good chances of scoring a goal. Under this measure Chelsea created chances worth 4.4 goals. There might be occasions where we have let teams create more chances expected to lead to a goal.

What XG doesn't tell you is whether a specific striker in a specific instance should have finished their chance. This is because there obviously isn't enough of a dataset to work with. Even someone who has taken as many shots as say Cristiano Ronaldo hasn't shot enough to even have a useful dataset.

If you try and use stats and metrics to explain things they aren't designed to cover the result will always be flawed. 

What the above doesn't include is how easily Chelsea were able to make the chances, and have players in space high up the pitch. The chances they had were good, and so 6 goals is quite a reasonable return for them. 

We were shit.

[Dman]

35 minutes ago, Fabrice Fernandes no.1 fan said:

This is painful to read, you're arguing different things here to be honest.

XG is a good useful metric. It allows an individual chance to be given a value, by which you can measure if a team has created what should be considered good chances of scoring a goal. Under this measure Chelsea created chances worth 4.4 goals. There might be occasions where we have let teams create more chances expected to lead to a goal.

What XG doesn't tell you is whether a specific striker in a specific instance should have finished their chance. This is because there obviously isn't enough of a dataset to work with. Even someone who has taken as many shots as say Cristiano Ronaldo hasn't shot enough to even have a useful dataset.

If you try and use stats and metrics to explain things they aren't designed to cover the result will always be flawed. 

What the above doesn't include is how easily Chelsea were able to make the chances, and have players in space high up the pitch. The chances they had were good, and so 6 goals is quite a reasonable return for them. 

We were shit.

I agree with this. 
 

It was actually Twar who was arguing that we were unlucky to concede 6 given their XG was 4. Anyone reasonable and who actually watched the game can tell you that is nonsense. 

Edited 12 April, 2022 by Dman

[Charlie Wayman]

On 09/04/2022 at 18:28, Suhari said:

Adam fucking Hammill. A fat Scouse in a t-shirt. Remember him coming on once as a sub, and I wondered which kebab House he was signed from. Whatever happened to him?

Riding bikes for Deliveroo

[Whitey Grandad]

1 hour ago, Fabrice Fernandes no.1 fan said:

This is painful to read, you're arguing different things here to be honest.

XG is a good useful metric. It allows an individual chance to be given a value, by which you can measure if a team has created what should be considered good chances of scoring a goal. Under this measure Chelsea created chances worth 4.4 goals. There might be occasions where we have let teams create more chances expected to lead to a goal.

What XG doesn't tell you is whether a specific striker in a specific instance should have finished their chance. This is because there obviously isn't enough of a dataset to work with. Even someone who has taken as many shots as say Cristiano Ronaldo hasn't shot enough to even have a useful dataset.

If you try and use stats and metrics to explain things they aren't designed to cover the result will always be flawed. 

What the above doesn't include is how easily Chelsea were able to make the chances, and have players in space high up the pitch. The chances they had were good, and so 6 goals is quite a reasonable return for them. 

We were shit.

It also doesn't take into account that any individual chance might have been created by that particular striker. Someone 7ft tall is going to get more heading chances than someone who is 5ft 2"

[Dman]

On 12/04/2022 at 14:25, TWar said:

Your external factors are strikers are different levels of clinical. If this was true better forwards would out perform xG but they don't.

😬

 

[smithy]

Stats are stats and have a place for analysis, but they do not take in to account the situation on any given game day, so therefore are a guide and nothing else. Anyone who watched Saturday's game and has played 11 a side football at any level knows that we could and should have been hammered by more than 6. We were 4-0 down after half an hour with multiple woodwork shots, the xG of 4.4 does not reflect that game in any way. I'm a bit shocked that the xG was 4.4 given the chances created by Chelsea. If McCarthy was in goal it's double figures for me as FF made some incredible saves that really did keep the score down. It baffles me that people get blinded by xG and can't or refuse to acknowledge this. 

[TWar]

1 hour ago, Dman said:

😬

 

Not sure what you think you are proving here. Sometimes players will overperform xG, no one is denying that. It just isn't the mark of a good forward. Benzema across the season is performing pretty close to xG.

Just like Salah, Kane, Ronaldo etc.

I don't want to reignite this argument but if you think a good player overperforming xG over 2 games means anything to the overall point I don't think you really understand the article you are replying to.

[Dman]

12 minutes ago, TWar said:

Not sure what you think you are proving here. Sometimes players will overperform xG, no one is denying that. It just isn't the mark of a good forward. Benzema across the season is performing pretty close to xG.

Just like Salah, Kane, Ronaldo etc.

I don't want to reignite this argument but if you think a good player overperforming xG over 2 games means anything to the overall point I don't think you really understand the article you are replying to.

Ahhh see I knew that you’d respond with this, so I’ve done a bit of research. 
 

The majority of big players out performed their XG over the period of 8 seasons, a few stand outs though are: 

 

Kane - https://understat.com/player/647 +21

Messi - https://understat.com/player/2097 + 34

Halaand (over 3 seasons) - https://understat.com/player/8260 + 11

Then you have the likes of Vardy, Argueo and Benzema, who over perform pretty much every season as well. And that was just from a quick off the top of my head glance at what I’d consider the best finishers in football. 
 

So better forwards do in fact out perform their XG, don’t they….

[TWar]

24 minutes ago, Dman said:

Ahhh see I knew that you’d respond with this, so I’ve done a bit of research. 
 

The majority of big players out performed their XG over the period of 8 seasons, a few stand outs though are: 

 

Kane - https://understat.com/player/647 +21

Messi - https://understat.com/player/2097 + 34

Halaand (over 3 seasons) - https://understat.com/player/8260 + 11

Then you have the likes of Vardy, Argueo and Benzema, who over perform pretty much every season as well. And that was just from a quick off the top of my head glance at what I’d consider the best finishers in football. 
 

So better forwards do in fact out perform their XG, don’t they….

Vardy doesn't overperform every season, he has underperformed 2 out of every 4 seasons, last season with a big one. Benzema has underperformed 3 of his last 6. Aguero is generally pretty good but he underperformed his last season. 

Of your three Kane and Messi both are underperforming this season. 

All you've done here is shown that over a big enough dataset with the correctly selected players you can find overperformers. But each of them frequently have seasons they underperform.

I know you are new to statistics but this is a textbook example selecting a data set arbitrarily to prove your point rather than for any actual reason.

[Dman]

12 minutes ago, TWar said:

Vardy doesn't overperform every season, he has underperformed 2 out of every 4 seasons, last season with a big one. Benzema has underperformed 3 of his last 6. Aguero is generally pretty good but he underperformed his last season. 

Of your three Kane and Messi both are underperforming this season. 

All you've done here is shown that over a big enough dataset with the correctly selected players you can find overperformers. But each of them frequently have seasons they underperform.

I know you are new to statistics but this is a textbook example selecting a data set arbitrarily to prove your point rather than for any actual reason.

The figures don’t lie. You said the best strikers don’t out perform their XG, they do. I’ve proven they do.

Naturally, there are variants season on season and players will have good and bad seasons (for a number of reasons), but over a significant data set (generally in statistics the larger the data set the more accurate the numbers), they over perform. Fact. 

It’s been an extremely long winded debate, but in short, your argument was that it wouldn’t have been more with another striker in place of Werner. Statistically,  had Lukaku played (+6), there is a greater chance that he’d have scored the chances Werner (-6) missed. 
 

Keep digging though, the more you post, the more you show how much you haven’t a clue. 

[Singapore Saint]

On 12/04/2022 at 22:07, Fabrice Fernandes no.1 fan said:

This is painful to read, you're arguing different things here to be honest.

XG is a good useful metric. It allows an individual chance to be given a value, by which you can measure if a team has created what should be considered good chances of scoring a goal. Under this measure Chelsea created chances worth 4.4 goals. There might be occasions where we have let teams create more chances expected to lead to a goal.

What XG doesn't tell you is whether a specific striker in a specific instance should have finished their chance. This is because there obviously isn't enough of a dataset to work with. Even someone who has taken as many shots as say Cristiano Ronaldo hasn't shot enough to even have a useful dataset.

If you try and use stats and metrics to explain things they aren't designed to cover the result will always be flawed. 

What the above doesn't include is how easily Chelsea were able to make the chances, and have players in space high up the pitch. The chances they had were good, and so 6 goals is quite a reasonable return for them. 

We were shit.

Nice analysis and interesting nick! I can't imagine there are too many Fabrice Fernandes fans here. When we went down in 2005, he refused to play in the Championship so we had to sell him. Great at keeping the ball but "no end product" (per MLT). When a Saints fan said, "You played well today," he replied, "Only today?"

 

Edited 15 April, 2022 by Singapore Saint

[TWar]

8 hours ago, Dman said:

The figures don’t lie. You said the best strikers don’t out perform their XG, they do. I’ve proven they do.

Naturally, there are variants season on season and players will have good and bad seasons (for a number of reasons), but over a significant data set (generally in statistics the larger the data set the more accurate the numbers), they over perform. Fact. 

It’s been an extremely long winded debate, but in short, your argument was that it wouldn’t have been more with another striker in place of Werner. Statistically,  had Lukaku played (+6), there is a greater chance that he’d have scored the chances Werner (-6) missed. 
 

Keep digging though, the more you post, the more you show how much you haven’t a clue. 

The figures do lie if you specifically pick 3 players who make your point and lie saying 3 more players always overperform when they objectively don't. What you need is a meta analysis across all top strikers, like in the source I posted. Thinking cherry picking 3 strikers out of thousands is statistically conclusive is the sort of data analysis you'd expect from someone who fundamentally doesn't trust stats. Maybe leave it to those who know what they are on about. 

[Fabrice Fernandes no.1 fan]

On 15/04/2022 at 04:35, Singapore Saint said:

Nice analysis and interesting nick! I can't imagine there are too many Fabrice Fernandes fans here. When we went down in 2005, he refused to play in the Championship so we had to sell him. Great at keeping the ball but "no end product" (per MLT). When a Saints fan said, "You played well today," he replied, "Only today?"

 

It is quite tongue in cheek I have to say. I was convinced he was a wizard of a footballer at the time but I think I was lured in by a few wonder goals!