I'd feel better if HSU paid someone else to do the testing, even if it is a "public" venue.
Further, I can't imagine that turning a subwoofer on it's side would impact output 3 to 4 db, regardless of the "patent pending" triangular port design, or circular ports, or square, rectangular or pentagonal (is that a word?).
Unless the sub was a defect, it's hard to believe that 1 sub's measurements were off by that much while the others were within 1 db. Especially if Dr. HSU himself was there for the measurements and nothing odd was noted at the time?
BTW, when HSU first came out with the sub, didn't they post intial CEA-2010 figures, then a few weeks later post slightly higher numbers and mention "the previous were from a prototype model"? Curious if the AH numbers are closer to the previous ones?