I already know the difference between los and nlos. Go geeky all you like. The fact that people in the REAL world, are receiving these REAL WORLD results, means something, more so than any "lab tests". We all know the huge differences between lab controlled environments and real world scenarios...
And the fact that even Clear@home modems, usb sticks, and the EVO all suffer similar (but ranging) huge amounts of degradation, there's simply variable unaccounted for, be it the transmitters Sprint's using, the receiver chips, the antennae, or the spectrum being used. More than likely, it is a combination of all of the above. Until they start sticking high gain antennae in the receiving units, all these problems will exist, which otherwise would not and do not exist with lower spectrum, using the same or similar hardware.