At the time when job openings are declining across the board, it's easy to lie with statistics.

It's like publications saying "research discipline X is gaining traction in recent years" based on search matches in PubMed, but not normalizing for the total number of articles submitted increasing.