Gene expression is a good point. It was in the back of my mind (and I glossed over it completely), but you've succinctly explained why it's actually an important issue.
Do you have a notion as to the relative time for gene promoters to hook to a gene with sex dependent fitness? I imagine that the mathematics of evolution would depend to a large degree on gene <-> selector or gene <-> meme coupling.
While we're at it, similar mechanisms account for the difference between a neuron and a muscle fiber, which share identical genomes.
b) A very successful man can end up with hundreds of concubines and children. A significant percentage of Asians are direct descendents of Genghis Khan, and there was a King Niall in Ulster who is said to be an ancestor of 25% of the men in Northwest Ireland. Monogamy somewhat limits this in our culture, but most people for most of the time humans have existed have been polygynous. For a woman, this is impossible - she can have a dozen or so babies at best.