Joel Spolky's article 'Hitting the High Notes' asks whether it makes sense to talk about having the 'best programmers?'.
He provides some interesting data showing how some programmers can crank the code out 70% to 250% faster than others for the same 10 assignments at a programming course at Yale university.
"There's just nothing to see here, and that's the point. The quality of the work and the amount of time spent are simply uncorrelated."
...If the only difference between programmers were productivity, you might think that you could substitute five mediocre programmers for one really good programmer. That obviously doesn't work. Brooks' Law, "adding manpower to a late software project makes it later," is why. A single good programmer working on a single task has no coordination or communication overhead. Five programmers working on the same task must coordinate and communicate. That takes a lot of time. There are added benefits to using the smallest team possible; the man-month really is mythical."