"The problem with this though is it conveniently ignores some parts of the data that just don't fit. Such as the voice radio transmission sector change acknowledgement AFTER the transponder was shutdown, etc."
Some of the problems we are having, is that we have to few facts, and to much fantasy and theories.
How do we know the transponder was shut down before last voice transmission?
Depending on the failure they faced, we do not know what systems that would have failed.