The peaks do not designate "cis", you can be cis and fall anywhere on the chart - being cis is about the sex you were arbitrarily assigned at birth (and whether that assignment aligns or conflicts with your actual gender identity).

And when doctors change assignments, it's really unclear whether you're cis or not if you transition - e.g. a baby assigned female at birth who is then weeks later assigned male at birth later transitions to be a girl, she was originally assigned female at birth - is she trans or cis?

Instead the peaks represent the most common combination of male and female sex traits in humans, with the slopes representing less common combinations of traits, e.g. to the left of the male peak might be men who experience excessive androgenization like lots of body hair, maybe precocious puberty, early balding, and so on (more male traits than average).

This chart as a model of sex actually doesn't make much sense, since sex has been redefined in light of how complex sex is and the differences in sexual development that occur.

Where on the chart would we put someone with complete androgen insensitivity syndrome (CAIS)? With CAIS a person is born with XY chromosomes and thus has a typical male karyotype, but their androgen receptors do not respond to androgens, so none of the masculinization is able to occur - leading the person to look, develop, and usually live as a woman.

The chart implies a spectrum, when the reality of biological sex is much more complex than a simple spectrum would allow - more like a constellation. Each sex differentiated trait is an axis / spectrum of its own, and there are thousands of ways differentiation can happen.

EDIT: oh, and to answer your question, it sounds like your question is really whether on a bimodal distribution if the peaks represent a smaller number than the tails in aggregate, and the answer is that it depends on how you select your aggregates and how much of the peak you lump together. I think the entire point of the bimodal distribution, though, is to show that the majority fall on the peaks while the tails represent a minority.

That said, a MRI study found that when examining brain sex, >90% of people (mostly cis) were not able to be classed as having fully male or female brains, so realistically I think it's fair to say most people are sexually divergent in some way.

Yes, hyperreal genders do exist, but are not stable outside lab conditions.

Well, clearly. If you define a male characteristic as something that's more common in men than in women and vice-versa, then e.g. being tall would be a "male characteristic".

Height isn't a binary thing with men being exactly Xcm tall and women exactly Ycm, so there's people who have more of said male characteristic and people who have less. And you also have women who have more of this characteristic and some men (e.g. there are some women that are taller than some men).

The same can be done for every characteristic that's associated with a gender. Genitals are on a spectrum (large clitoris vs micropenis), fat distribution is on a spectrum (e.g. there are men with breasts and women without), body hair is on a spectrum, hormone distribution is on a spectrum and so on and so on.

If you take a lot of characteristics at once it becomes clear in most cases whether the person you are dealing with is a man or a woman (though there are some where that's more difficult or impossible), but if you take just a single characteristic (e.g. height) it's impossible to say whether the person you are dealing with is definitively a man or a woman.

It means that traditionally understood cis male can still have some female characteristics (no facial hair, higher pitched voice, bad at driving) but some males will have none.

I don't think it's an accepted term anymore, but you reminded me that they used to call the triple X chromosome syndrome by the term Super-Female-Syndrome.

Probably not what the author intended though.

yes.