I'm a postdoc in the physical sciences, currently writing some job applications.
I have noticed that sometimes, particularly for research oriented faculty positions but also some postdoc positions, you may be asked to list your three best/most significant papers. Presumably the goal here is to enable a meaningful evaluation of your research, without the need to read all of your papers.
My question is: what is the optimal strategy for choosing which papers to put forward as your best three?
Should one simply choose the ones which are in the highest impact journals?
I imagine that long papers might be off-putting to busy people with lots of applications to evaluate, so is it best to only choose short papers? On the other hand maybe throwing one long paper into the mix can, if it's good, demonstrate a capacity for deep scholarly research?
How heavily does first authorship weigh? Does it look bad if one of your best three papers is one where you aren't the first author?
In my specific case, I'm submitting a faculty application where I have to choose three papers and I'm fairly sure of two of my choices. For the third, it comes down to a choice between a well-cited paper in a fairly high impact journal, but where I am second author versus a paper where I am first author, and which I think is a good paper, but it's in a less high profile journal and hasn't attracted very much attention in the field. Are there some general principles for deciding which is a better choice?