That's a good presentation of what I think are valid data. If you search through some of the discussions here on the GT forum (my note-sharing with PonyOne on pot values comes to mind) you'll find lots of anecdotal support for what he's saying.

One reason I avoid single-coil setups is I don't like being held hostage by whatever I plug into. My humbucker guitars aren't quite as subject to drastic changes imposed by cable length, although I do use 1 Megohm pots. I generally use a 20-foot cable so that I can wander around a bit, and if I'm playing my VT-22, it's definitely unwise to get too close! Someone recently was asking about linking multiple amps like Tony Iommi's rig with six 100-Watt Laney heads. I pointed out that he used a treble booster to compensate for the treble loss through the long cables.

Note his finding that George L cables have lower capacitance than most brands. Word-of-mouth reports of improved clarity (less treble loss) with these cables tend to tally with his data. Bear in mind that his model only deals with a guitar plugged straight into an amp input. Adding pedals changes the situation dramatically. The output of most modern pedals is a low-impedance op-amp, which will not be nearly as sensitive to cable capacitance. Something vintage, like an Arbiter Fuzz-Face is something else again.

In summary: no earth-shaking revelations here - just solid backing for what critical observers have long remarked on. Thanks for the link. It's nice to have some hard evidence to point to. :)