I think this conversation is kind of going far afield. Discussions that we can measure the impact of a cable or a switchbox on the signal, and thereby verify its audible transparency, aren't particularly productive in this context. And what context is this? It's the context of an argument between people who believe in that science, and those who do not. So for the purposes of the debate, then, talking about a priori measurements is assuming the premise.
Let's face it, it's the same old discussion! It is simultaneously preaching to the choir of objectivists and tired old nonsense to the subjectivists. Kind of boring for both, actually.
So if we're trying to actually accomplish something new, we objectivists have to find ways to meet subjectivists on their turf. Give them as much reason to be comfortable with the testing conditions as possible. Give them as much rope as possible---to hang themselves with, of course.
This is the part of the approach of the James Randi foundation's $1MM challenge that I liked so much. They worked to hammer out a protocol that was acceptable to both parties, and they bent over backwards to accommodate the challenger's requests, as long as they didn't compromise the validity of the test. Granted, most of the challengers still figured out a way to bail out and accuse JREF of cheating. But the ones that went through with the test, as their video documentation shows, were treated well.
One of the ideas they promote is requiring the challenger to pass a non-blinded version of the test being proposed before they proceed to the full blinded test. That is, every aspect of the test protocol is put in place except for the blinding, and the challenger has to demonstrate they can perform under those conditions. If they can't, the test doesn't move forward.
Relating that to audio... so we need a switchbox? Fine. Put one in place. Sure, it might affect the sound (objectivists roll their eyes). But maybe the subjectivist will still hear differences between cables with it in place anyway. Maybe the differences will be made more subtle (still rolling). So give the subjectivist as much time as he wants to train to these new conditions, to find material that does the best job of exposing the differences that are left.
If he is confident that he is able to resolve differences, great! Put the blinds on and go for it!
If not---well, the tests ends before it starts. Yeah, you may not have been able to prove what you'd hoped to prove, but at least you have demonstrated that a an extremely-electrically-benign switchbox can obliterate all of the perceived improvement of a $30K pair of wires.
It's not the definitive smackdown an objectivist would love to see---but then, it's more than you'll get by boring a subjectivist to sleep with talk about frequency responses and impedance so that they lose interest in doing anything.