Can team velocity be measured?

Can team velocity / work estimates really be measured accurately? If yes, to what degree?

I read an article recently (can't find it at the moment) around how story points can be improperly used as a measure of time which turns into a way of non-technical leadership imposing unrealistic timelines upon engineers.

It's a story as old as the industry itself: non-technical management not fully understanding the nature of the work at hand, which leads to distrust, overly-optimistic estimates, and burn out.

How should story points and velocity be used? How do you deal with this aspect of agile development with your leadership? Book recommendations?