Results v Performance. City v Liverpool

[Cobwebcat]

I put this post in a different thread but I thought it might be of interest in its own right.

These days it’s possible to have a look at a few things objectively.

https://understat.com/league/EPL

If you click on the end column you will see what the table “should” look like as a result of a team’s performance. Of course these are never going to be the same as the real table because there is a lot of luck in footy as a result of it being a low scoring sport. However, at this stage of the season, I have never seen a team obtaining TEN more points than their performances deserve. That looks like there is more going on that JUST luck.

Talking of results v performance you can also see how good teams rank on these two criteria.

This first ranking is purely results based so if you dominate but lose 1-0 to a dodgy decision it’s tough.

http://clubelo.com/

This next ranking is performance based and strips out most of the luck in the game. It’s a better predictor of how good a team is now and, on average, how it will perform going forward.

https://projects.fivethirtyeight.com/global-club-soccer-rankings/

In both methods City and Liverpool are the top two teams; but reversed. There’s no doubt that they are the two best teams in the World right now.

It does seem however that Liverpool are getting help to add to the fact that they are already very good.

 

[Gorton_Tubster]

They're getting help with luck mainly. Just one major injury and they wobble. And they are supremely confident too because of the media positivity.

Bias is there, but I don't think the PL /PGMOL are actively organised cheating.

Interesting stats though. Cheers

 

[born2bblu]

They're getting help with luck mainly. Just one major injury and they wobble. And they are supremely confident too because of the media positivity.

Bias is there, but I don't think the PL /PGMOL are actively organised cheating.

Interesting stats though. Cheers

What about VAR? It IS actively organized cheating.

 

[Marvin]

I put this post in a different thread but I thought it might be of interest in its own right.

These days it’s possible to have a look at a few things objectively.

https://understat.com/league/EPL

If you click on the end column you will see what the table “should” look like as a result of a team’s performance. Of course these are never going to be the same as the real table because there is a lot of luck in footy as a result of it being a low scoring sport. However, at this stage of the season, I have never seen a team obtaining TEN more points than their performances deserve. That looks like there is more going on that JUST luck.

Talking of results v performance you can also see how good teams rank on these two criteria.

This first ranking is purely results based so if you dominate but lose 1-0 to a dodgy decision it’s tough.

http://clubelo.com/

This next ranking is performance based and strips out most of the luck in the game. It’s a better predictor of how good a team is now and, on average, how it will perform going forward.

https://projects.fivethirtyeight.com/global-club-soccer-rankings/

In both methods City and Liverpool are the top two teams; but reversed. There’s no doubt that they are the two best teams in the World right now.

It does seem however that Liverpool are getting help to add to the fact that they are already very good.

How are the XGoals, XGA and Xpoints compiled?

How do you explain the difference between City's points and expected points. Is it luck? If it's poor finishing, well then that's deserved. Perhaps you can say that if a team has scored more goals than would be expected from the XG stat then that is because they are very clincial and therefore they deserve to be there.

I am sorry I don't know much about this data.

I am not sure that it makes sense to conclude that City and Liverpool are the best in the world based on this type of data because it is relative to the competition?

 

[Cobwebcat]

How are the XGoals, XGA and Xpoints compiled?

How do you explain the difference between City's points and expected points. Is it luck? If it's poor finishing, well then that's deserved. Perhaps you can say that if a team has scored more goals than would be expected from the XG stat then that is because they are very clincial and therefore they deserve to be there.

I am sorry I don't know much about this data.

I am not sure that it makes sense to conclude that City and Liverpool are the best in the world based on this type of data because it is relative to the competition?

They’re good questions.

Answers:

The difference is mostly luck. City create a lot of chances and proportionately a lot of misses. When it comes to being clinical then it’s quite surprising that virtually all strikers, given a big enough data set, are as clinical as each other. What makes someone like Messi special is the positions he takes up when he shoots in that he gets into positions where he is expected to score more.

Expected figures are compiled slightly differently by different organisations but Understat are considered one of the best. How they are calculated differ therefore but the general principles are the same and any YouTube clip on them will explain better than I could.

The two sets of rankings both take into consideration the level of competition and much more besides. Virtually all the European and World rankings including those based on bookmakers odds have us 1 and 2. The two I’ve listed tend to be the most accurate though.

 

[BosnianBlue]

Sorry mate, but things about Messi getting into positions where he is expected to score more is bulls. He is far less in those positions than any regular striker. He is not one messing (see what I did) around 5 yards as regular nines. His stats are high because he is just miles ahead in finishing than anyone else. if we had anyone close to Messi in finishing, we would win everything in 2nd gear.

 

[Cobwebcat]

Sorry mate, but things about Messi getting into positions where he is expected to score more is bulls. He is far less in those positions than any regular striker. He is not one messing (see what I did) around 5 yards as regular nines. His stats are high because he is just miles ahead in finishing than anyone else. if we had anyone close to Messi in finishing, we would win everything in 2nd gear.

Nope that’s not correct. Statsbomb did a podcast about it specifically about a year ago. I was surprised too but there you go.

 

[Father Paul]

Sorry mate, but things about Messi getting into positions where he is expected to score more is bulls. He is far less in those positions than any regular striker. He is not one messing (see what I did) around 5 yards as regular nines. His stats are high because he is just miles ahead in finishing than anyone else. if we had anyone close to Messi in finishing, we would win everything in 2nd gear.

Not only finishing its what he sets up and manufactures out of nothing. I'm not a big fan of expected figures but I suppose they are an interesting side show.