scientific double blinded cable test

Just to mention, the word repeatable can be misleading. The reason being if you mean repeatable by the same person we must insist on independent tests. Even then the word repeatable is a little tricky since the very nature of the initial test could have been flawed. So repeating a flawed test wouldn’t have any value, would it? The tests should be independent in every sense of the word. Different system, different test conductor, different test subject(s). The results, now those should be repeatable.

It’s unfortunate that this thread - which is about the science of double blind audio testing - has become one of the ugliest ever on Audiogon. It’s proof how some want to politicize this issue - and I say that even as I argue that while such testing has its place, it’s of limited value to most audiophiles.
geoffkait

Just to mention, the word repeatable can be misleading

I think "repeatable" in this context means that the test can be replicated by others. That requires that the test protocol be explained to at least allow the possibility of independent verification even if - as sometimes happens in science - another conducts the ostensibly same test, but arrives at a different result.

If the testing protocol is inherently unscientific (as proposed by gdhal here), it saves other experimenters the trouble of trying to replicate the test. If you’re trying to apply science, it’s futile to pursue an unscientific protocol.

That’s how science works.

No, it means the results must be repeatable. Remember Cold Fusion. They had rigged the rest or whatever. You certainly don’t want to replicate phony test procedures. That’s why you want completely independent tests. As I said before one test all by itself has no meaning if the results are negative. Ideally you want a lot if independent tests that come up with the same results. If one is negative you throw that one out. Problem solved.