I like this author, but here is the rub. The entire article is predicated on a "koin" flipping program where one can manipulate the rules that govern probabilit

I like this author, but here is the rub. The entire article is predicated on a "koin" flipping program where one can manipulate the rules that govern probability. This clearly does not make flip-(n) independent from flip-(n-1) and since the flips are no longer independent, one might reasonably expect that the traditional gambler's fallacy in not a fallacy.