Brown Nose
Well-Known Member
I mentioned this metric in the post match thread for the Liverpool game.
Lots of stats minded people seem to believe that it is a more reflective way of measuring performances than actual results. There are many, many articles going on about how Team A has an xG of this and Team B has an xG of that as a way of explaining why individual results may not be the best way of assessing how a team is actually doing. There is certainly some merit in saying that results are often not reflective of performances.
xG is designed to measure the quality of chances created and conceded which, over a sustained period, should tell you more about how your team is performing than their results. Match of the Day have been using it for a couple of years now in their post match analysis.
I was puzzled by the fact that our xG against Liverpool was deemed to be only 0.10. In contrast, Liverpool had an xG of 3.75. This suggested that we were very lucky to score and by rights should have lost the game 4-0. So I decided to do some reading up on the topic.
For information, our 0.10 that day consisted of the Maddison goal (rated as 0.07) and the shot from Praet (0.03). Liverpool's 3.75 consisted of 0.76 for the penalty, Mane's goal was given a 0.32 xG and then there were a further 15 attempts that 'scored' xG for Liverpool during the match.
If you accept that the metric is of interest, and this is very much something that is growing in usage and popularity, how are Leicester really doing this season?
If results were determined by xG rather than goals, we would have 11 points rather than 14 and we would actually be 11th in the xG table. Man City would be above Liverpool at the top of the league and Watford would be 8th.
Leicester have very low xG for and against this season. Only Newcastle have fewer expected goals than we do. On the other hand, only Liverpool and Man Utd have better defensive records. This strongly suggests that our positive start is mostly down to us being strong defensively and we're actually pretty poor going forward.
In terms of individual players, Vardy has 5 goals this season but his xG would only give him 2. This clearly reflects well on Vardy in that he is making more of his opportunities than you'd expect. Perez has been our most wasteful attacking player. Most xG assists (passes that lead to a shot) have come from Albrighton (0.46 per appearance) and Gray (0.43 per appearance). Maddison by comparison only scores 0.15 per game.
Does any of this convince me of the merits of xG? Not really. However, I am going to continue to monitor it as the season progresses to see if xG and actual results/performances converge more. I'll use this thread to pick up on notable things xG related as the season progresses.
Lots of stats minded people seem to believe that it is a more reflective way of measuring performances than actual results. There are many, many articles going on about how Team A has an xG of this and Team B has an xG of that as a way of explaining why individual results may not be the best way of assessing how a team is actually doing. There is certainly some merit in saying that results are often not reflective of performances.
xG is designed to measure the quality of chances created and conceded which, over a sustained period, should tell you more about how your team is performing than their results. Match of the Day have been using it for a couple of years now in their post match analysis.
I was puzzled by the fact that our xG against Liverpool was deemed to be only 0.10. In contrast, Liverpool had an xG of 3.75. This suggested that we were very lucky to score and by rights should have lost the game 4-0. So I decided to do some reading up on the topic.
For information, our 0.10 that day consisted of the Maddison goal (rated as 0.07) and the shot from Praet (0.03). Liverpool's 3.75 consisted of 0.76 for the penalty, Mane's goal was given a 0.32 xG and then there were a further 15 attempts that 'scored' xG for Liverpool during the match.
If you accept that the metric is of interest, and this is very much something that is growing in usage and popularity, how are Leicester really doing this season?
If results were determined by xG rather than goals, we would have 11 points rather than 14 and we would actually be 11th in the xG table. Man City would be above Liverpool at the top of the league and Watford would be 8th.
Leicester have very low xG for and against this season. Only Newcastle have fewer expected goals than we do. On the other hand, only Liverpool and Man Utd have better defensive records. This strongly suggests that our positive start is mostly down to us being strong defensively and we're actually pretty poor going forward.
In terms of individual players, Vardy has 5 goals this season but his xG would only give him 2. This clearly reflects well on Vardy in that he is making more of his opportunities than you'd expect. Perez has been our most wasteful attacking player. Most xG assists (passes that lead to a shot) have come from Albrighton (0.46 per appearance) and Gray (0.43 per appearance). Maddison by comparison only scores 0.15 per game.
Does any of this convince me of the merits of xG? Not really. However, I am going to continue to monitor it as the season progresses to see if xG and actual results/performances converge more. I'll use this thread to pick up on notable things xG related as the season progresses.
