There are topics that we have discussed ad nauseum in this thread, and which many of us get tired of having to reiterate. But I'll give it another go.
Females that identify as transmen and take hormones pass significantly better as men than the other way around. Testosterone is a one-way street. It causes physiological changes that cannot be undone. Muscle mass is the very least of it, and the only thing that is somewhat "reversible" if a female stops taking testosterone. Facial hair, on the other hand, does not go away. Once the biological switches have been tripped, even if a female stops taking testosterone, they will not stop producing facial hair. Similarly, once their voice drops - which is a result of testosterone - it will never rise again. These changes are permanent in females.
Males that identify as transwomen and take hormones almost never pass even a little bit unless they have considerable cosmetic surgery. Estrogen alone will result in a repositioning of fat on the body, leading to marginally softer hips and face, as well as deposits of fatty tissue in the breast. But estrogen will not stop a male from growing facial hair, nor will it raise their voice, nor will it alter their skeletal structure. If a male stops taking estrogen, they will revert back to a male-typical muscle and fat distribution in relatively short order. The visual changes caused to a male by taking exogenous estrogen are not permanent.
In humans, females are better at sexing other humans than males are, although both are very good at it. Shown faces alone, in monochrome, without hair or make-up or clothing indicators, females correctly sex an unaltered human about 98% of the time, and males do so about 90% of the time. I have my speculations about why that is, but it's a bit off topic.
Facial hair is a strong indicator of sex in humans, as it's an element that is dimorphic in all but extremely rare cases of female hirsuitism. If a person has a beard, we will - all of us - assume they are male and rarely look closer.
The net result of all of these factors is that females who identify as transmen are less likely to be accurately sexed by cursory glance than are males who identify as transwoman.
+++++++++++++++++++++++++++++++++++++++++++++++
On the topic of females who identify as transmen being required to use the single sex spaces of their biology... well, to be frank, most of us don't care which spaces they use. Some males would undoubtedly be put out by the presence of a female body being naked in their presence, and I can definitely empathize with that.
But the situations are not parallels.
I know some people don't want to admit it, but statistically males are materially more likely to be aggressive and violent, and to be sexual predators than females are. So there's an element of agency and risk involved in which spaces a transgender person uses, dependent on their sex.
I'll attempt a poor analogy here. Let's make the illustrative assumption that males are foxes and females are hens. If a hen decides that they are a "fox in spirit" and decides to go visit the fox's den... well, that's on them. They are the one who is accepting the known risk to themselves. The foxes may not like the hen in their den... and the foxes are completely capable of forcing that hen to either leave or face the easily foreseeable consequences of their presence.
On the other hand... if a fox decides they are a "hen in spirt" and decides to wander on into the hen coop... well that's something altogether different, isn't it? The hens are certainly not a threat to the fox, even if they greatly outnumber the fox. A single fox in a henhouse could cause immense harm if they so desired. And despite there being many hens to only one fox, it is the fox that represents the threat.
To break this analogy down to its salient point: When a female that identifies as a transman enters a male-only space, they are willingly accepting a risk to themselves. When a male that identifies as a transwoman enters a female-only space, they are imposing a risk on everyone except themselves.