Good finishing or good long distance shooter ?
A thread
A thread
The “G-xG“ metric is used a lot on social media to judge the obscure quality that is “finishing“
Before starting the thread, we need to remember what finishing is…
Before starting the thread, we need to remember what finishing is…
Finishing means to complete the attack by scoring (converting a goal scoring opportunity into a goal)
If we explain in term of statistics, it means scoring high quality chances (> 0.25 xG)
If we explain in term of statistics, it means scoring high quality chances (> 0.25 xG)
“shooting“ is different, generally it’s lower chances (< 0.10 xG)
There is not real line between those two but you can see the difference, in finishing people expect you to score but when you take a long shot it’s good if you score but people don’t expect it to go in
There is not real line between those two but you can see the difference, in finishing people expect you to score but when you take a long shot it’s good if you score but people don’t expect it to go in
So it leads to a problem… what if when using the “G-xG“ or “xG performance %“ metrics, the player in question scored more than expected just because he scored 10 0.03 xG chances in the season. He wouldn’t be a good finisher, more like a good long shot taker
Let’s use Coutinho as an example. Since 14/15, he scored 56 league goals from 48.4 xG, which makes a 16% overperformance
Based on that, we would say that he was a constant good finisher, but if we look into detail…
Based on that, we would say that he was a constant good finisher, but if we look into detail…
22 of those goals were from out of box, from an xG of 10.75 which makes a 105% overperformance ! But his xG per shot was only 0.03
However, when you look from inside of the box, he scored 34 goals from 37.7 xG which makes a -9.8% xG performance, his xG per shot was 0.16
However, when you look from inside of the box, he scored 34 goals from 37.7 xG which makes a -9.8% xG performance, his xG per shot was 0.16
On this graph we can see that De Bruyne’s xG performance suddenly dropped last season, why ? He was asked to play further up hence to shoot in closer situations, his xG per shot was at 0.12 compared to his usual 0.08 average in his career
Coutinho and De Bruyne are amazing at long shots, but it doesn’t make them good finishers.
Messi too ? Many people are saying that Messi only gets such a high xG performance because of long shots, let’s verify. Since 2014 he scored 53 long shots from 33.9 xG, amazing…
Messi too ? Many people are saying that Messi only gets such a high xG performance because of long shots, let’s verify. Since 2014 he scored 53 long shots from 33.9 xG, amazing…
… but he also scored 178 goals from 158.4 xG inside the box, no player had a higher +/- difference than him in that area. He’s just an excellent finisher and long shooter.
At the end you just need to give context, don’t use G-xG alone to judge finishing
At the end you just need to give context, don’t use G-xG alone to judge finishing
Data from understat.com
جاري تحميل الاقتراحات...