Why apply work in an unexpected situation

Stream: new members

[Chengyan Hu]

I believe that if we apply a hypothesis h to our goal, it will only work in case h:P ->Q and goal|-Q, turning our goal into P. However I found out in application that if the goal is P and we apply h:P, it solves the goal. Why is that?
image.png

[Kyle Miller]

In general, if h : P1 -> P2 -> ... -> Pn -> Q and the goal is Q, it creates $n$ goals for P1, P2, ..., Pn.

The tactic still works when $n=0$.