Allow me to address both of your concerns with some scientific studies.
The idea that a significant portion of trans children are actually "just gay or a tomboy" was pretty much the standard assumption among child psychologists for a very long time, to the point where this was largely taken as axiomatic
without actually studying it. And when it was actually studied, the results genuinely appeared to support that thesis. The most commonly cited study, the one by Steegma's research group, tracked 127 child patients into adolescence, and found that 80 of them had "desisted" by age 15-16.
However, those studies include one
huge method flaw: the study measurement of "desistance" was based on the percentage of children referred to gender clinics by their parents who were still being treated for gender dysphoria at the same clinic on follow-up studies. That sounds straightforward at first, right? But think about how that works demographically:
1. Children who never actually identified as transgender, are being counted as "desistance" cases. With the Steegma study, the initial gender clinic classified 38 of the 127 patient cases as "sub-threshold", i.e. they did not actually meet the diagnostic criteria for gender dysphoria. Steegma nonetheless included these cases as if they had actually been diagnosed as transgender.
2. Children with no follow up are being counted as "desistance" cases. Again, I will take Steegma as an example because hard numbers are available: out of the 80 cases he reported as "desisting", 28 were actually cases where the research team were unable to locate the same patient or the patient's family declined to participate in the follow-up. Instead of marking those as null results, Steegma counted them as "desisting". There were only 52 actual responses of "desistance" and again, that's *without* counting the 38 cases that weren't actually diagnosed transgender at all.
Note that Steegma himself acknowledges these methodology flaws, but now states that the flawed desistance numbers don't really matter because the desistance numbers are
looking at the study backwards in the first place -- he was counting persistence, not desistance.
This was carried forward in Steegma's
2011 follow-on study, where he actually filtered for cases that were
actually diagnosed and then also actually participated in the follow-on interviews, and concluded that there was a critical age range of approximately 10-13, beyond which point all interviewed patients (both persistent and desisting) reported a firmly fixed gender identity. He also found that the
reported motives for gender exploration were distinct between persistent and non-persistent cases.
This led to further study and even more explicit clarification in his
2013 follow-on study, which concluded that "intensity of gender dysphoria" was a strong predictor of persistence; i.e., those cases where children
strongly and definitively stated a specific gender identity, as opposed to those whose parents merely
thought they might be transgender based on "cross gender behavior".
In other words, the evidence suggests that many/most children who exhibit behavior that goes against social gender norms are not necessarily transgender. However, the evidence is highly definitive that when children
strongly and definitively assert a specific gender identity, they are absolutely persistent in that gender identity. There is therefore no need to restrict transitioning in such cases, certainly not beyond the 10-13 age range -- which is actually
less restrictive than current medical practice.