>>This is NOT a sync problem, there are no corrupted trees,<<
I never said it was a sync issue, and your description indicates it could be a corrupted file, it merits consideration.
>>and please assume I know what I'm doing.<
I don't assume, I evaluate the query and the terminology used. Your posts are confusing, it's not clear what you are doing or trying to do.
>>The issue is that the FTM option of exporting a subset of your tree in FTM format results in a file that cannot be successfully imported back into FTM.<<
This doesn't make any sense, "exporting a "subset" ( that's not how it works), "results in a file", (what kind of file?? a gedcom?) , "that cannot be successfully imported back into FTM".
What is the point of creating a file in a program, exporting it to somewhere, and then importing back to the same program?
At the very least, give my method my a try, you have nothing to lose and you might learn something.