[Brutha]

The only reason to cite Alexa ranking is to bamboozle people; after all, it's inaccurate, as you readily agree. So what's the point behind citing inaccuracies as "proof"?

Steve doesn't even belief in statistical proof. If you are familiar with Steves writes (especially the Blackbelt article) you should know that there is no possible statistical data that would be a prove for Steve.
An Alexa rating is better than nothing, but it proves nothing.
Steve doesn't suggest that statistical measure can prove success. It's natural that the numbers he cities don't prove success if he doesn't intend it.

My post was not about anything other than the misleading practice of citing Alexa ranking in sales copy.

It's misleading because Alexa is not accurate. For Steve to cite it as evidence that SBI! works when he knows full well that Alexa is worth very little as a measure of traffic is not helpful for someone who claims to want to help others realize their potential.

Steves doesn't suggest that something else besides your own commitment is helpful for realizing someones potential (and you can read that belief in his copy and the discussion afterwards in the forum).

Steve pays no attention to Alexa not because he has his server logs, but because he knows Alexa is inaccurate compared to those server logs

Alexa is supposed to solve a different problem than self evaluation.
Whether it solves that problem doesn't has something to do with whether it is good for self evaluation.

I don't think that Alexas inaccuracy matter much if you average out over houndreds of sites.