It would be interesting to see if any great researcher could tie back their success to a teacher. My suspicion (based on nothing more than gut feeling) is that while a good/great teacher has some impact, that beyond a certain level it's 99.9% the individual and their environment/peers/parents/contacts, rather than any particular teacher.

Another way of saying this from a mercenary monetary point of view is that I don't think investing more than $250K/year on a K-12 teacher makes sense, though you might be able to make an argument that people who teach teachers might have greater leverage.