There's clearly a group of folks who aren't particularly concerned about MPG and how it's calculated and I respect that. But the thread title is "MPG for those interested" and a number of other people who posted earlier clearly are interested, so perhaps I could respectfully ask those who aren't interested to allow that discussion to continue please ?
For me, I'm just interested in whether anyone can put up any plausible theories as to why two methods of measuring the same thing produce significantly different results. Doesn't mean I'm anal about the last fraction of an MPG, it's just an objective academic interest (call it an enquiring mind). I don't buy the "That's just the way it is" theory, there's got to be some underlying reasons. And some of the reasons suggested so far may also affect the supposedly accurate manual fill method ... which my sad mind finds quite interesting.