This speech was posted to MetaFilter a few years ago and generated a much deeper discussoin than it seems to here. http://www.metafilter.com/64034/The-Waw-effect Definitely worth reading some less glowing discussion about the talk.
I do agree though that it starts out good, and becomes less and less substantiated, but on the whole it is interesting.
One argument was that the adventurous sailor would be subsumed by the cobbler or tavernkeeper in the mating game. True, but he wouldn't be subsumed by the beggar, now would he?
Another argument is to find a counterexample. "There were female scientists. Good female scientists. Your claim is now disproven!". Except for the fact that the claim is statistical and you just picked an outlier. There are a couple of main claims, which are quite objective:
Are males more extremely distributed than females? Is the minimum wage pertubing the mean wage (yes, it is - but significantly?) Is the absolute grade scale pertubing the results (yes, but by how much?) If all three are confirmed, you could say the societys obsession with the mean is what causes all the trouble.
I think your point about less and less substantiality is warranted. It is the first couple of paragraphs which has the most meat and is the most objective. It would be bad to dismiss them, if the latter parts are weaker. Yet, the weaker parts are the ones people will attack in an argument - they are the low-hanging-fruit. Especially if they have an agenda to be fulfilled in the gender wars. Though the more objective parts are the parts you should seek to refute.
I think the future will be even more interesting than now. In the most developed parts of the world, women surpasses men in higher education. Note that this fits the presumption: males have a more extreme ability distribution. This will have a great impact. There will be less well-educated men for the well-educated women. And this is a first in the history (of Europe). It will be interesting to see how that one plays out.
Well, if the beggar is Diogenes...
The main counter argument is that most world conquerors are not especially fertile; Napoleon had maybe six kids (max.); Alexander had two and was maybe gay; Hitler had zero. How would the "conqueror" trait have so much influence on humanity if conquerors did not father many children (and there were not enough successful conquerors to start with).
I try to develop a more substantiated analysis of this article here: http://blog.medusis.eu/of-parrots-and-men
The argument is Evolutionary Psychology, which is a rather soft science. So is economics, but that doesn't mean it isn't useful.
However, in order to make it useful you have to think statistically about the behavior of humans in LONG LONG period of pre-history.
Leaving aside the fact that nature does not design anything, if evolution is to select a trait that is "mostly unsuccessful", then this trait is to be incredibly successful when it actually succeeds.
And in fact, Genghis Khan is the exception; other well-known world conquerors are not exceptionally prolific, casting doubt on the possibility that "unknown world conquerors" (if such people ever existed?) would have been prolific enough and numerous enough to have an impact on the human species.
I'm well aware of Evolutionary Psychology, its premises and its shortcomings; it isn't "soft science" but rather "non-science" since it can't be refuted. Most of its claims may be right, or wrong, but that doesn't matter because we have no way to tell.