Well, I'm not sure if you're saying you disagree with the hypothesis or with that being what manufacturers point at.
The general jist of things is explained by this graphic from Conti:http://velonews.competitor.com/files/20 ... 46x421.jpg
Hed has claimed this as a benefit from about 2009 but as I said, the only data that I've seen is very neutral. Al Morrison showed that the same tire on a traditional 20mm and a 23mm rim, back to back on rollers, had the same resistance within margin of error.
And from an article by Greg Kopecky over at ST:
"Does that automatically mean that the Crr is lower? Some manufacturers suggest that this is the case, but nobody is producing hard numbers to prove it – or at least showing them to any of us. Zipp Technical Director, Josh Poertner, challenges the idea. He suggests that there may be a very, very small improvement – but neither their testing, nor any independent testing they’ve seen can prove it outside of the margin of error of the test (on the order of .0000125 Crr improvement). My suggestion is that saying that the wide rim itself improves Crr, is like claiming that the chicken came first, when it’s really the egg. Yes, a wide rim allows you to run a wider tire and lower pressure – which lead to lower Crr. But it isn’t the rim itself that lends to this improvement; or at least not enough that we can actually measure it. I’ll go ahead and insert the voice of reason for you: Go worry about something else! Sleep more! Eat less Twinkies!"