I don't know if I totally agree with this. Wouldn't the tires be carrying their weight in roughly the same distance from the axel since the goal is to not change overall diameter? The 15 tires tread is just as far from the center as the 19" tires if the overall diameter remains the same, and it should.
The treads would be the same distance from the axle but the beads would be further from the axle. The overall effect, assuming the tires had about the same weight, is the larger one carries it's weight further from the axle. However, this difference is minimal compared to the larger rims carrying a very significant portion of their weight further from the axle.
I'd like to see the results with that test throwing in a very lightweight wheel in the mix with a larger size like the 18". It would answer how much of a difference wheel weights matter. I'd think a 17lb 18" wheel would yield much different results then the heavier wheels they used
Yes, going with the lightest 18" wheel and the heaviest 17" wheel, the lighter 18" wheel may very well have less rotational inertia (aka 'moment of inertia'). However, applying that same lightweight wheel technology to a 17" rim would provide even greater benefit (and at a considerably lower cost).