I simply must disagree with your basic presumption.
As you say, any cable (or anything else) will make some measurable change - which may or may not be audible.
In the case of an interconnect cable, the most likely change is that it will slightly alter the frequency response (probably by rolling off the high end a tiny bit).
Less likely, it may introduce some type of distortion, either by it's own failure to do its job, or because it influences some other component.
The amount in which a cable affects the signal is inversely proportional to the quality of the cable itself, and the other components involved.
(Audio components should not be sensitive to slight increases in capacitance or resistance of the interconnects,
and good interconnects shouldn't change those parameters enough to influence a good component to an audible degree.
NO cable is going to magically REMOVE distortion that's already there.)
As for whether this is a positive change.... let's take a visual analogy.
Let's pick a brand of glass to put in front of our favorite paintings....
Personally, I prefer the clearest glass possible, so I can see the paintings as they are.
Choosing an interconnect that makes some change is like choosing blue-tinted glass....
even though it may make a certain painting look better (to you), I don't think it's the job of the glass to modify the painting.
If you like blue paintings, then BUY blue paintings!
Don't buy pink paintings and "balance them out" with blue glass.
For that matter, don't paint your walls red, then put blue glass over the paintings to try and cancel out the reflections from the walls.
Gray walls, clear glass, and paintings you actually LIKE is your best combination.
Likewise, buy components you like rather than relying on a cable to (hopefully) fix what's wrong with them.
I agree with your statement "all cables must be considered as good or bad processors" however,
as far as I'm concerned, in this case, the best one is the one that you can hear making the least audible difference....
and, with even decent equipment, I have no problem finding $50 cables that make no difference <I> can hear.
[Of course, I am also considering the possibility that, when an expensive cable sounds different, it has actually
been designed to deliberately sound different so it has "a product differentiator" - in which case it may well simply
be coloring the sound - and so is a *worse* quality interconnect than the cheap one.]
Considering how good human hearing is (OK), and the fact that audio frequencies are actually a very simple signal to get
from point A to point B without alteration (as long as your equipment and cables are at least somewhat well designed),
it really shouldn't be that big a deal. Many (most) cable vendors are selling "technology" that, if it's legitimate at all,
only matters at microwave frequencies or under other ridiculous conditions.
As if that isn't crazy enough, the studio where they recorded the stuff you're listening to is almost certainly
using $0.50 a foot Canare Starquad for THEIR interconnects... they are quite pleased with it and your audio probably survived
hundreds of feet of it before getting to the record or CD you're listening to..
As far as I'm concerned.... if you can hear the difference between two cables, then one or both of them MUST be wrong.
If they sound the same, at least it's possible that they're both arbitrarily good enough to be right.
Balancing components by matching their flaws to try and have them cancel each other out will lead to
spending lots of money, probably won't work out very well anyway, and means that every time you
change any bit of the puzzle you have to start over again.
Much better (and easier, and, in the end, cheaper) to carefully select each component to be right to begin with.
I can not agree with your view. Any cable will change the signal. IMHO, some of them will improve the sound quality of a system, which is by definition good.
If some one could create a top high-end system that could not be improved by the choice of cables your position could become much stronger. But until some one creates such a system all cables must be considered as good or bad processors. If the cables can improve our systems why should we deny it just in the name of the measured electrical signal integrity?
Sound reproduction is an imperfect process. Any thing that can really improve it should be considered as good. Surely we can debate if the improvement was real.