Well, I was intentionally trying to describe the approach, not the details. Note the "if true" part in my hypothetical reply, for instance.

Part of establishing what is true could be exactly what you said: citing the guidelines for the Olympics. Someone else might charge that the Olympics is driven by politics more than by science, and that the optics were better to defer to a pro-trans view even at the expense of their athletes (in support of the "politics beats safety and reality" charge, they might cite age limits on gymnasts and figure skaters, ostensibly to "protect children" but in practice young athletes simply have to train just as hard for even longer until they meet the age limits, doing more damage to their bodies not less). So then you'd come back with the evidence that whichever Olympic committee used to justify its decision(s) regarding trans athletes to show that it was evidence-based not another case of picking whatever seemed best for business, and so on.

The point would be that concerns are taken seriously and addressed, not dismissed; and a realist approach would probably include potentially transiently accepting differential treatment even in the case where there wasn't evidence of a problem but widespread concern and a lack of evidence that there's not a problem (assuming that there is some way to gather evidence so it would be a temporary situation--everlasting differential treatment based on an untested and untestable supposition is pretty much the definition of discrimination).