To the latter point, I certainly had a similar experience, and that's in fact why I value track record so highly. It took a very different set of skills when I made my own transition from contracting -> academia -> bigCo, but I'd loop this back to my "thesis", namely that someone observing my weakness after each step would rightfully observe I don't have a track record within that context.

To some extent this seems logical, would we call a home carpenter "great" within the context of industrial skyscraper construction? The moment you step back from a rather tight focus, (and this may be the root question/debate about "what does great mean", and why my approach has generated some highly opposed responses, because I try to dodge that problem by limiting the scope) any commonalities become much harder to define, and limited mostly to "meta-skills."

So basically, you're right, for me it is very context sensitive, and as I said upfront, I have no silver bullet for identifying greatness without digging into that context quite a bit. In prior threads about the trouble with interviewing I've been very eager to both share the techniques I use/find out how others approach the problem, since summarizing after the fact what I found valuable, much much easier than determining it in a loop. I would love techniques that let me determine how well someone's skills in context A transfer to B, but that is a _massively_ dimensioned question that I can't even begin to hand-wave at, beyond that it might go back to my "meta-skills" categorization.