Attacking and defensive performance from 2017 tp 2020 (xG based) : Gunners
Small point: teams with more defensive set ups tend to (but not always) concede a higher xG against. For reference: Sheffield United conceded 13 goals less than expected this season, Burnley in 2017/18 (the year they got into Europe) conceded 13 less goals than expected, and Mourinho’s United conceded 15 goals less than expected in the year they finished second. If you also look at Atlético Madrid’s xG against, they always succeed in conceding less than expected (14 less in 2017 as an example).
There is a case to be made that the quality of the keeper is the reason, but it’s “a” reason rather than “the” reason. Tito football covered Burnley’s tactics in 2017, and explained why they invited certain shots to be taken at their goal: https://youtu.be/NKHwUVYH4hU
That spike in xG against at the end of the graph is right around when we went to a back 5, and right around the time we looked at our most defensively secure. It was also around the time we played City twice, Liverpool twice, and chelsea. It also includes that horrid Watford game that we did not give a fuck about.
The xG being low isn’t a surprise, and we know the reason. Just thought I’d add some context to the xG against stat
