Factoring the advisor into evaluation of job candidates

When considering applicants for postdoc and junior faculty positions it is natural to focus on the track record of the applicant, particularly journal publications (number, quality, citations, ...). However, I think one really needs to take into account who are the advisor(s) of the candidate.

It is important to remember that you are hiring the student not their advisor.

Consider two extreme candidates and their advisors.

Joan's advisor Susan is at a "highly ranked" university and runs a large well oiled machine that graduates students on time and produces a steady stream of publications, some in luxury journals. Each student project makes extensive use of hardware and software from previous generations of students. Most of the papers have long author lists. Susan tends to work on somewhat interesting but safe problems that are guaranteed to produce papers in a timely manner. Joan met rarely one-to-one with Susan, but was actually mostly supervised by postdocs. Their discussion always focussed on the next step for getting results for a paper. There was little discussion of the big picture, background knowledge, career advice, or other issues. Susan travels extensively, promoting her group's work in highly polished talks, increasing citations. Susan writes beautiful letters of reference that really "sell" her group members.

John's advisor Steve is well known for just a few highly original papers. Steve is not particularly well organised and supervises few students. He assigns them difficult problems that may not bear fruition and few graduate on time, or have many papers at the end of their Ph.D. Each project is self- contained and the students have to write all their own code from scratch. Steve worked closely with John making sure he had a good foundation and learnt all the "tricks of the trade". Steve is adverse to self-promotion and does not like going to conferences, particularly big ones. Steve writes dry letters of reference that are cautious about the strengths of his group members.

Who would be the better hire? Joan or John?
Although his "track record" may not be as impressive, John actually may have the stronger scientific training and have more potential than Joan.

Having made the above point, students and postdocs are heavily influenced by their advisors, for better or worse. There are always exceptions, but most pick up good and bad habits from their advisor. This covers a wide range of issues from attitude to hype, carefully checking results, quality of talks, breadth of interests, focus, .....  True, some students do learn little from their advisor, either because their advisor has no interest in teaching them or the student is too stubborn or proud to learn anything. In the end, hiring someone whose advisor trains their students well and models scientific integrity, is a good move.

What do you think?