I am sure that at the top of the performance hierarchy, natural talent matters. Below that, and including at that level, one must be hard-working, practice lots, be reasonably intelligent, disciplined, etc. My feeling is that for most professors in most fields, they are well below the top, and those characteristics of hard work etc matter the most. There are some fields, mathematics, musicianship, ... where you do have to have lots of natural talent to get anyplace at all. And if the university tenures people only after two books or after about ten years (Harvard, MIT, ...) , with stellar references, it's likely that more of its faculty has both high performance and natural talent.
On the other hand, if you have some natural talent, it is easier to get better and that may affect your willingness to work. More generally, matching your self to the potential role/work is important.
What this adds up to is that many more people can be trained/educated for many of these roles dealing with uncertainty. And some people will prove to be naturals, others will not but will be more than good enough and quite committed (the trouble with naturals is whether they are willing to be committed, be coached, realize they are not quite as good as they might believe).