Saddleworth2
Well-Known Member
- Joined
- 27 Jan 2014
- Messages
- 20,466
This from the Daily Fail, reporting on the introduction of an expected goal metric to Motd this season
So, what is an expected goal?
An expected goal is a way of working out the percentage likelihood of a player scoring from a position on the pitch.
It is a new analytical tool provided by Opta which shows how many goals a player or a team should have scored based on the chances they had in a game or over the course of a season.
Each chance is ascribed a 'quality value' (xG) based on thousands of chances analysed by Opta over the history of the Premier League.
The higher that figure - and one is the maximum value - the more likely a chance will be converted.
What can they tell us about players?
It is a very good way of establishing whether a player is scoring more or less goals than his chances should dictate.
So, for example, a striker could finish as a club's top scorer, but he might actually be one of the weaker finishers in the side.
It also lets a fan judge at a glance whether a striker is an elite level player or whether he is simply being bolstered by the creative talents around him in the team.
Each of last season's top eight goalscorers, barring Sergio Aguero, scored more goals than their expected goals would suggest likely. This means that each of them are expert finishers.
And for 29-goal Harry Kane, who scored over 10 goals more than expected, it flags him up as a world-class striker.
The same system applies to players who are performing worse than expected. It allows you to flag up whether a player is being wasteful in front of net, beyond someone in the pub saying, 'He had to score that!'.
ACTUAL GOALS V EXPECTED GOALS
Player Actual goals /Expected goals (xG)
Harry Kane 29/ 18.6
Romelu Lukaku 25/ 15.3
Alexis Sanchez 24/ 17.6
Sergio Aguero 20/ 20.5
Diego Costa 20/ 14.2
Dele Alli 18/ 13.3
Zlatan Ibrahimovic 17/ 14.4
What decides the likelihood of a goal?
Opta have analysed over 300,000 shots to calculate the likelihood of an attempt being scored from that specific position on the football pitch.
That has also included what phase of play the chance comes along in.
These are the key factors:
It lets an analyst judge a team's future performances if their current results are not matching up to displays.
Say there is a team that is dominating matches and making a ton of chances but losing or drawing more frequently than their play would suggest.
Sometimes that would mean a manager would be up for the chop. Expected goals lets that fear disappear to an extent and proves that a turning point is likely to just be around the corner.
Juventus in 2015-16 are an example cited by the BBC. They won only three of their first 10 games but analysis shows they were vastly outperforming that level of results.
And after a turning point came, they eventually cantered to the title. Expected goals can be a valuable tool for fans, journalists and pundits alike.
Basically this stat is saying the Kun is scoring what you would expect him to score based on the quality of chances he is getting, but of the top 7 goalscorer, he is the only one that hasn't scored more than their 'expected goal' score.
I guess my reaction to that is its probably validating what we saw with our own eyes, he is a great striker but he misses chances and doesn't really score the worldies he maybe once did because he has dropped a little in pace. If you look at what it says about Kane, again it smacks of the truth. He does score a lot of goals that you would rate as half chances. Will be interesting to see how this stat develops as it can also be applied to whole teams as it was to Juventus in the example given. I bet we didn't beat our expected goal score last season!
So, what is an expected goal?
An expected goal is a way of working out the percentage likelihood of a player scoring from a position on the pitch.
It is a new analytical tool provided by Opta which shows how many goals a player or a team should have scored based on the chances they had in a game or over the course of a season.
Each chance is ascribed a 'quality value' (xG) based on thousands of chances analysed by Opta over the history of the Premier League.
The higher that figure - and one is the maximum value - the more likely a chance will be converted.
What can they tell us about players?
It is a very good way of establishing whether a player is scoring more or less goals than his chances should dictate.
So, for example, a striker could finish as a club's top scorer, but he might actually be one of the weaker finishers in the side.
It also lets a fan judge at a glance whether a striker is an elite level player or whether he is simply being bolstered by the creative talents around him in the team.
Each of last season's top eight goalscorers, barring Sergio Aguero, scored more goals than their expected goals would suggest likely. This means that each of them are expert finishers.
And for 29-goal Harry Kane, who scored over 10 goals more than expected, it flags him up as a world-class striker.
The same system applies to players who are performing worse than expected. It allows you to flag up whether a player is being wasteful in front of net, beyond someone in the pub saying, 'He had to score that!'.
ACTUAL GOALS V EXPECTED GOALS
Player Actual goals /Expected goals (xG)
Harry Kane 29/ 18.6
Romelu Lukaku 25/ 15.3
Alexis Sanchez 24/ 17.6
Sergio Aguero 20/ 20.5
Diego Costa 20/ 14.2
Dele Alli 18/ 13.3
Zlatan Ibrahimovic 17/ 14.4
What decides the likelihood of a goal?
Opta have analysed over 300,000 shots to calculate the likelihood of an attempt being scored from that specific position on the football pitch.
That has also included what phase of play the chance comes along in.
These are the key factors:
- Distance from goal
- Angle of the shot
- Was it a shot or a header?
- Has the player just gone around an opponent?
- Was it a one on one or were defenders involved?
- What was the assist like? (e.g. long ball, pull-back, cross, through ball)
- In what part of the game did the chance occur? (e.g. open play, direct free-kick, corner kick)
- Is it a rebound?
It lets an analyst judge a team's future performances if their current results are not matching up to displays.
Say there is a team that is dominating matches and making a ton of chances but losing or drawing more frequently than their play would suggest.
Sometimes that would mean a manager would be up for the chop. Expected goals lets that fear disappear to an extent and proves that a turning point is likely to just be around the corner.
Juventus in 2015-16 are an example cited by the BBC. They won only three of their first 10 games but analysis shows they were vastly outperforming that level of results.
And after a turning point came, they eventually cantered to the title. Expected goals can be a valuable tool for fans, journalists and pundits alike.
Basically this stat is saying the Kun is scoring what you would expect him to score based on the quality of chances he is getting, but of the top 7 goalscorer, he is the only one that hasn't scored more than their 'expected goal' score.
I guess my reaction to that is its probably validating what we saw with our own eyes, he is a great striker but he misses chances and doesn't really score the worldies he maybe once did because he has dropped a little in pace. If you look at what it says about Kane, again it smacks of the truth. He does score a lot of goals that you would rate as half chances. Will be interesting to see how this stat develops as it can also be applied to whole teams as it was to Juventus in the example given. I bet we didn't beat our expected goal score last season!