Hi Danny,
Tests like these are very complex and rarely ever work as advertised, this one in particular uses a pinna transform. I was actually part of a research project many decades ago that was assigned to the head of the engineering dept at the company I worked for at the time. The way these are typically done is that a mannequin head is used and outer ears (pinna) are attached to the head. A specialized small capsule microphone is placed in each ear and then a highly directional speaker is used to move around the mannequin head - typically 360 degrees around the head in every possible direction - and then again at a different distance. Sine wave seeps or a maximum length sequence signal is generated from the speaker and then the signal received by each microphone is recorded and then transformed into a frequency response.
From this data, one can then make timing and frequency response changes to a left/right stereo signal to simulate what is "heard" which corresponds to a specific position in space based on what I mentioned above.
Directionality is due to having binaural hearing, the ability to hear separate sounds in both ears. It is subtle changes in frequency response and timing from one ear to another that allows our brain to determine a direction. It is something our brain subconsciously learns as soon we start hearing. It is very subtle.
The assumption of, are these speakers capable of reproducing height and depth with this test, is wrong. It isn't whether or not the speakers can, it is whether or not your pinna and the acoustical environment of your room is close enough to what was used when the transform was created.
Looking at your room, it is full of reflections. A reflection is when sound will bounce off a surface. This reflected sound is not necessarily bad, but it will dramatically change the frequency response and timing of the original signal - thus completely throwing off the various spatial cues. In addition, the shape and size of your head, the separation distance between your ears and most importantly, the shape of your pinna - all must be quite similar to the mannequin head or your brain won't translate the intended directionality properly.
Not in the least, in fact - it helps with vertical height as it limits unwanted reflections - which is precisely the reason so many speakers are designed specifically for controlled vertical dispersion.I know the ribbon tweeters have very limited vertical dispersion so not sure if that affects the "height" imaging.
We don't actually hear sounds that are above or below us, or to the left or right - sound waves hit our eardrums all at the same direction. Our brain determines direction by using our very own built in HRTF, otherwise know as a head related transfer function.
In simper terms, the Sierra-2 and Sierra-2EX are extremely accurate speakers and easily able to reproduce the subtle changes in frequency response and timing necessary for proper imaging. In the case of what you are experiencing with that ledr test, the environment and possibly your pinna are simply not a close enough match to the HRTF transform generated by the original equipment.
Back to the project I enjoyed being a part of, which I believe was an early version of QSound. Very few of us heard the intended directionality - in fact, it was interesting because those of us who were younger had better results ( I was young back then ) We had better results when the speakers were placed in a highly treated sound room (very few reflections) and positioned in a perfect equilateral triangle, with a listening distance of 1 meter. Even then, I recall slightly less than 50% of us were able to properly place the sounds - for which we then correlated that those with smaller outer ears had more accurate results than those with larger outer ears. We suspected that the original HRTF was developed using smaller outer ears...
That was a fun week for a few of us and some good memories for me.
I'm rambling - but hopefully I made some sense.