When analysing the performance of teams in one particular match, agreed. When used over a much larger quantity of matches as one of a number of figures you could utilise to assess the quality of strikers, it seems a perfectly reasonable tool.
Yep. On a game by game basis it means very little because it's a probability stat and they're always shit with a small sample size but over longer periods it's as good a stat as you'll get to help focus where to train and identify what is and isn't working.
Even then it's only useful as a pointer for things to consider but it does a great job of highlighting things that can be easily missed otherwise.
