(I've thought about that, but the consideration that seemed more salient to me at the time was: If you send different signals than expected then those who can notice subtlety will notice a discrepancy given, say, a few hours of interaction. Yes you'll be oft-discounted (and you will have incurred this cost yourself and I don't deny that this is indeed a cost worth considering), but the people who falsely present themselves as more important than they are so vastly outweigh the people who falsely present themselves as less important than they are that causing someone to update their estimate of your importance upwards is more likely to make a (justifiable) positive impression than the alternative case which involves someone having to eventually update their estimate of your importance downwards. It's like the inverse of "don't throw pearls before swine". (I'm drunk, I apologize if I'm stating the obvious.))

Behaving low-status has the advantage of avoiding status fights in your tribe... by giving up. At the proper moment in the ancient environment it could save your life.

That does not necessarily mean the cost-benefit analysis would have the same outcome today.

Right. This is the "evolutionary optimality challenge" of Bostrom and Sandberg, which is "If the proposed intervention would result in an enhancement, why have we not already evolved to be that way?"

Gwern's excellent article on that lists some ways to escape the challenge; I'm not sure which are at play here, but I think dominance is generally a good idea.

Not everyone feels the need to submit, even in the same situations. I'm sceptical of the idea that the genetic component of the idea is the main thing at play here.