Your right, I forgot about waterfall plots...you rarely see enough of those for them to be useful, I wonder why?
Interestingly enough, every speaker I've heard that looks good on either paper, FR or waterfall plot has never overly impressed me. Some mediocre measuring speakers have though. Something's obviously missing in our evaluation methods (or known and just never provided).

Is anyone aware of test where someone measured the properties of a live performance against the recording of that same performance on different speakers, etc??? I don't know what test you'd use for this, but I have a feeling that you need to have a valid point of reference (ie: the real thing) other than +/- 0 dB, 0.00% THD, etc...