This article does not make its case. He shows a graph of "random" data, and then just kind of keeps going. But that random data is the meat of the whole thing.

Cut out 60% of the useless text, and focus on explaining why random data should look like that.