Note that this is a graph metric, not a single attribute. Attractiveness varies WIDELY across person-person edges. It would be interesting to try to measure the variance across incoming attractiveness for specific women. Much more interesting than ignoring that variance and pretending attractiveness is a single value comparable in useful ways.
Also, “egalitarian” is not well defined here. To the extent it’s a relative measure (man X or even median man “would prefer” woman Y over woman Z), it cannot be egalitarian unless there’s an extremely large number of “no preference”. Which I don’t think is your claim. If you collapse it to “would bang, on a desert island”, that may be equal, but that’s because it’s a binary dimension which CANNOT vary by much.
Note that this is a graph metric, not a single attribute. Attractiveness varies WIDELY across person-person edges. It would be interesting to try to measure the variance across incoming attractiveness for specific women. Much more interesting than ignoring that variance and pretending attractiveness is a single value comparable in useful ways.
Also, “egalitarian” is not well defined here. To the extent it’s a relative measure (man X or even median man “would prefer” woman Y over woman Z), it cannot be egalitarian unless there’s an extremely large number of “no preference”. Which I don’t think is your claim. If you collapse it to “would bang, on a desert island”, that may be equal, but that’s because it’s a binary dimension which CANNOT vary by much.