[4reals]

I suspect you are right Zach. The print code variations for the base 91 topps set would confirm your theory. The 5 cards you mentioned with an E*F* code can be found with F* codes but they are significantly more scarce compared to the other sheets that made the same type of sheet code corrections. This would support that the first 5 sheets were corrected much earlier in the run than sheet F and that other subsets like DS were printed in between these corrections.