Don't you think the burden of proof is the other way around? We see selection effects all the time. Shouldn't you expect super programmers to get more done because they're simply better programmers?
It's a strawman argument to say that, "well, Lisp is more powerful than assembly." Of course. But is it more powerful than C#? Maybe, but the Lisp advocate should be the one asked for proof.