xG is a statistic to be used to analyse players over a season/long time frame in an attempt to measure their finishing ability, or how many goals a team should score based on the quality and quantity of chances they create. xA is useful over the course of a season to attempt to measure how many assists a player should have got (thus alleviating variance due to poor finishing from his teammates).
xG is also somewhat, but not particularly, useful at measuring a team's performance in a match, in an attempt to measure the typical "who deserved to win" debates.
It's not particularly useful at assessing a player's performance in one match. A high xG isn't even a indication of the player's ability - it's the difference between xG and actual goals that matters in terms of finishing ability. You can argue that it is a result of the combination of the player's movement, the chances his teammates create for him, and the quality of the opposition defence, but xG alone is not particularly useful at distinguishing between those factors or accounting for variance.
Even when used correctly, it's a limited and flawed statistic, as there are multitudes of factors that come into play. When used incorrectly, its primary purpose tends to be for some sneering tosser to try and win a debate on the internet.