Any actual basis to the notion that >70% of people are overly cautious for their own good, considering the risk of status loss (real or perceived), and the fact that they are also over-confident about the success rate?

I'd say that for well over 70% of people worldwide, what you said about environment of evolutionary adaptedness, still applies.

Speaking of which. Almost everyone gone through 'environment of social adaptation' in the learning sense, i.e. daycare then school, and the late stages of it are similar enough to 'environment of evolutionary adaptedness' - with the striving for status, small number of apparent mating opportunities, potentials for scary punishments - the products of conditioning by which can all provide ample food for evolutionary psychologists, and ample set of strange biases. The people who didn't go through school, are small and biased sample; the cultures where there is no school are naturally living in something even more similar to supposed 'environment of evolutionary adaptedness'.

It does seem that the sexual attraction, as a novel feeling, confuses the hell out of brain when it first appears, especially in puritan cultures. Kids even do certain thing, that involves looking at pictures of women (or imagining) , and being afraid that someone would walk in and catch them. That, plus Pavlov's conditioning, already provides ample explanation in terms of well studied, uncontroversial phenomena. It may not be as intellectually engaging as imagining what cavemen were like, but that's uncontroversial theory which is well tested, and predicts social anxiety of men around women in many cultures; EP can only predict additional anxiety on top of this. I've seen enough of different people of different nations, with different approach to talking to women, to know that significant anxiety is not universal. Keep in mind that i am from former soviet union, in which many people of different nations with diverse cultures and religions were moved around.