How does Impact Factor affect a professor's ability to run a lab? If someone is getting out one paper per year, would it be better to have a primary (or last) authorship on a low-impact journal, or a middle-authorship on a really prestigious journal?
(For our audience, here's an example. I've brought up the publications for Susan Lindquist, a dynamo researcher. You'll notice that one some articles, her name comes last. This tends to mean that her laboratory is the one that performed the majority of the work and that she was responsible for getting everyone together to do the work. First author tends to be the person who did the majority of the actual bench-work and writing. Middle authors are people who helped.)
http://www.ncbi.nlm.nih.gov/pubmed?term=lindquist s
You'll see that in 2012, she got a 'middle' authorship in the Journal of Neuroscience (a really prestigious journal), but in 2011 she got 'first' authorship in PLoS (the public library of science), which is an awesome journal but not regarded as highly as J. Neurosci.
The only thing that matters in running a lab is getting the $$ to do it. Now of course that requires that you produce results and a have ideas and is highly competitive but there are different strategies. It is best to have both high numbers AND high impact papers. This does not mean all papers need to be high impact but having just a few High impact and nothing else (me pretty much-I just finished a grant renewal and have 3 Science papers in the period but only 7 total) is not great however in the current funding climate you can no longer just churn out reams of stuff no one reads or cares about (at least in my field) it is just too competitive and there will always be people who are highly productive and have some high impact to win the grants over you these days.
As far as authorship order it is most important to be senior (last) author for a PI. I would rather have a senior paper in a lesser journal than a middle in a higher journal. It does depend on how low you go. You do have a reputation and you dont want to be known for publishing crap. Of course crap can be in high impact journals as well.