You certainly may be correct.
Question: If this is indeed happening, would you see it being reported in the log? (Turning off verbose mode so that you can see the error.) I know I have caught endstop errors in the log for the Z-axis when playing around with the bed and nozzle overrides.
I only know enough about this stuff to be dangerous
I don't know how STL files are parsed. I'm going to "assume" that when an error is found, the slicer has to decide how to deal with it. Either ignore it or try to fix it. All I do know is that certain files seem to work with some slicers and not others. As I said earlier, if there is a problem found I would prefer that the slicer throw up a warning. I have read that a new STL format is being developed that allows "curved" triangles. This would likely eliminate a lot of these types of problems.
My main interest is simply knowing the true root cause of the problem.