Nothing I said in that statement you excerpted denied the requirement of replicability (if that's not a word, then it should be!). In fact, I have stated the ability to replicate results using the same methodology as important in many of my other posts. What you said here was true, but it was more of a supplement to what I was saying, not a contradiction or contrast to it.