Steve Randy Waldman
@interfluidity.com

you don't think there's a lot of overlap, among people interested in playing, between the least competitive of one sex or gender and the most competitive of the other, in most sports? 1/

in reply to this
Steve Randy Waldman
@interfluidity.com

that's all it takes. if the most competitive of one sex or gender would consistently dominate, that's not sufficient to render sex or gender a far from noisy proxy. 2/

in reply to self
Steve Randy Waldman
@interfluidity.com

if there's a middle of the unified distribution that's substantially mixed, then these categories are very noisy proxies for ability. 3/

in reply to self
Steve Randy Waldman
@interfluidity.com

if your only interest is competitiveness to be at the very top of the distribution, then perhaps these categories seem less noisy. (perhaps not, i'm not making the claim, but it might be right!) 4/

in reply to self
Steve Randy Waldman
@interfluidity.com

but most questions surrounding participation in sports are not restricted to this hyperelite level. 5/

in reply to self
Steve Randy Waldman
@interfluidity.com

using less noisy proxies doesn't harm sorting according to living in the very right tail. if the most elite strata by directly relevant proxies turn out to be very disproportionately of a sex, gender, race, whatever, that's fine! 6/

in reply to self
Steve Randy Waldman
@interfluidity.com

segregating by other characteristics or demonstrated capability doesn't preclude those outcomes, while segregating by sex, gender, race guarantees noise beneath the far right tail and uselessly may preclude outliers from unexpected categories who could compete in those tails. 7/

in reply to self
Steve Randy Waldman
@interfluidity.com

for most of sport that's not hyperelite, sex or gender based segregation almost certainly prevents fair competitors from meeting one another. at the hyperelite level, there's no cost to sorting by better proxies for ability whether or not the outcomes prove quite similar. /fin

in reply to self