We’ve certainly seen cases here in the past where an upright vs. horizontal orientation using the 0-degree calibration file showed drooping high-end response with the former – really distressed the fellow when his response looked worse than he felt it sounded. My AudioControl RTA shows a sag in high-end response above 2 kHz when I point the mic off-axis.
I suspect a room’s acoustics can make a difference in these situations, but that’s a hard thing to quantify from one case to the next on a Forum like this. It never hurts to try legitimate measurements with both mic orientations (“legitimate” meaning that the proper calibration for the orientation is in place). However, if the two measurements come out different, I have more confidence in the accuracy of the horizontal orientation w/ a 0-degree calibration file measurement. :T