The problem is so much quote-unquote "realism", which is more tenuous than you or many others might describe (the idea that the U.S. Women's soccer team would lose to an average high school freshman men's squad is completely laughable).
I'm not making that claim without evidence (although you've slightly altered my claim). You realize that they do play exhibition games against U15 and U17 teams, and that most of the time they do lose? Most of the time these games aren't highly publicized, but I'm sure you'll be able to find the case where they lost to FC Dallas's U15 boys team 5-2, for example.
Are you in fact offended by reality?
Yes, there are currently definitive differences between biologically female and male athletes in a number of ways that are apparent at comparable competitive levels, though we are starting to see the cracks breaking in that "reality" (or at least in the structures that organize around it).
I have no idea what that means.
The problem is reductivism. It's reducing specific genders/races etc. (or even just individual characters) to nothing but a collection of stereotypes. And then acting as if those "truths" are universal. This is what leads to things like female strength penalties (or charisma bonuses; remember that reinforcing positive stereotypes is also problematic.)
No, because I have specifically argued against reductivism, noting for example that there is nothing unrealistic about having a woman much stronger, faster, or in all ways more athletic than myself. Thus, I have fully endorsed the position that only knowing the sex of the person, we can't know which is the more athletic, and I will fully endorse it again. As far as the claim of reductivism goes, I can't help but feel you in particular are now projecting, since I know for a fact that when I've argued that knowing only some quality of a person we can't know some other quality because people's individuality was greater than the collection of groups that they'd been categorized in, you've rejected that when it suited you.
Nor in my outline of anything have I suggested that the outcome of this chargen process must be a stereotype. For example, we could have a chargen process where you rolled a random number and were then assigned a person of the chosen sex from the pool of all the world's females. The character would then be based off a real person, and since real people are almost always more than stereotypes, any claim that the chargen process produced a stereotype would be provably false. You would never know what sort of person you'd end up with, and any attempt to guess based on the 'average person' would probably be a stereotype. I have certainly not argued for or acted like "truths are universal" as you put it, and by making that claim so in opposition to what you are actually responding to, I can only feel we are talking past each other.
But I continue to not understand what you are trying to communicate, nor do I feel you have answered my essential question. In fact, so unrelated is your response to my claims, that it feels deliberately evasive to me.