Selecting power cables requires no more knowledge than wire gauge vs. length, and maximum amperage draw/voltage requirement of the equipment. I (and others) say that what happens on the primary side of the transformer is of relatively little consequence provided standard, ordinary AC engineering is followed. (i.e., not high-end doofus tweako engineering.)
Speaker cables and interconnects are ruled by the Big Four: L, C, R, + noise.
The original poster would be singing a different tune about speaker cables if he'd blown-up a perfectly good amplifier using Polk Cobra Cables. Some amps don't play well with speaker cables having lots of capacitance. Resistance is futile, but capacitance and inductance are also important.
LCR is relatively easy--a scientist or committed hobbyist can model the frequency-response and amplitude change for a signal delivered to the speaker if the amp output impedance, the cable L, C, and R, and speaker input impedance are known. The difficult one is the "...plus noise" part of the spec. Not only do you have to measure the amount of noise, you have to describe the distribution of noise. I think that's where the Golden Ears go all soft in the head and long in the tongue. I suspect that most all of us have bigger problems than the spectral distribution of noise inherent in our interconnects and speaker cables. My room needs more acoustic treatment--about a thousand times more critical than my speaker cables.
