Every shot a team takes is assigned a score from 0 to 1 based on how likely it is that someone would score from where/how that shot was taken. For example, a penalty kick counts as 0.79 xG because PKs score roughly 79% of the time. Shots from within the 18 yard box are generally going to have a higher xG score than those from outside the box. Shots from in the 6 are going to be even higher, etc.

So the full-time xG scores presented here are the total of all the xG scores of all the shots that both teams took. We didn’t have very high quality shots throughout the game. City had slightly better, but still not great.

Going purely off the history/stats of similar shots to what both teams took, it was fairly unexpected for either team to have scored a goal in the match, which I think was kinda borne out by the eye test, and the fact that our one goal only came from fluking in off of Ake’s face.

Further reading: https://theanalyst.com/na/2023/08/what-is-expected-goals-xg

I’m sure most of man city’s xg came from the opening few minutes. After that we were the better team, but never produced any chances as good as Ake’s early miss.