Hmm, that seems correct then. This is interesting, would you consider extending the explanation and posting it as an answer to this question? https://quantumcomputing.stackexchange.com/q/5459/4153