In my opinion this thread has become nothing more than an intellectual pissing contest. The bottom line is that whatever measurements you take from input to output does not mean squat because you will only ever HEAR it if it is in a SYSTEM. Perfect null test? Whoopie do. Give me a colored hodge podge that in the end whose measurements hews closely to the original waveform in actual use to one that all measures supposedly flat individually but falls on it's face when measured from the listening position. In practical terms it's nothing but a security blanket or a crutch. They will only ever be guides for use and/or guides for quality control, etc. They will never be predictive because WE ordinary folks will never be able to look at those measurements and imagine what a piece of gear will sound like given every piece of music because we don't even know a single darned performance by heart. If one is self delusional enough to believe he knows at least one performance down to the tiniest level of audibility, whatever piece that might be, the sound of that will always be grounded on the impressions we had of that piece and NOT the actual sound. It's ludicrous.
+1
And as mentioned earlier, do you know anyone who has bought components and assembled a system based soley on the equipment's measurements? I for one don't.