In On Chasing Dreams, or Not: A Post Devoid of Coherence, xykademiqz muses about whether to encourage grad students to pursue professorial positions:
Chances of not succeeding at anything competitive are much greater than of succeeding. It’s heartbreaking to fail. But is it always the best idea to discourage the people from even trying? Should we as PIs at good but non-elite schools actively try to stomp out every inclination of all our progeny to even dream of a professorship because the chances are really slim? I try to let my students know as much as I can about what the job is; for most, that’s enough to turn them off. But there are some in whom I see the fire and the combination of skills and determination, and I think they could, with coaching and some luck, make it. Should I tell them to forget it just because, even for them, the odds of not succeeding are high?
In my current field (bioinformatics), there are still reasonable chances of students getting PhD-requiring jobs (in industry or in academia), though only a few of our students will go on to get tenure-track positions in research universities. Based on the alumni information at https://www.soe.ucsc.edu/departments/biomolecular-engineering/programs/msphd-bme-bioinformatics/alumni, about ¼ of our PhD alumni have gone on to become professors (and several are still postdocs, so the number who eventually become professors may be as high as ⅓). Success probabilities of 25% are still quite good (grant funding odds have gotten way lower than that), and many of those who don’t become professors never had any intention of becoming professors, so the odds of success for those actually seeking professorial positions from our department may be as high as 40%.
It is not necessarily the best researchers who became professors, because the attractions of startups, national labs, and industrial labs can be high for those who are solely research-focussed. Instead, it is those who seek both research freedom and teaching opportunities that have been attracted to professorial jobs. In some cases, the jobs are primarily teaching, with only modest research opportunities, while in other cases the jobs are mainly research.
I’m not directly supervising any PhD students now, since I’ve stopped pursuing research grants, but I do still advise students in their first year as grad students, in an informal way. I encourage them to try teaching, and if they like it, to consider getting more practice and more training through programs like the Institute for Scientists and Educators. But I don’t assume that everyone is going to end up teaching nor that it represents the best outcome for most students.
If I were in a field where the opportunities were much more limited (like biomedical research or physics), or if I were at an institution that was not at the top of the field for the subject, I’d have a harder time figuring out how to advise students.