“Bad explanations are easy to vary while good explanations are hard to vary.” – David Deutsch
The key issue with an theory whether it’s a myth or a scientific theory is that not only must it have testable predictions but more importantly it’s “explanation” of objective reality must be hard to vary; in other words, there can’t be many or any variations of the explanation. The narrower the explanation that has success making predictions the closer that hypothesis (theory) is to objective reality. The more variants to the hypothesis (theory) that have as accurate (or inaccurate as the case may be) predictive power the less likely those hypotheses (or theories) have to do with objective reality.
For example, Newton’s Gravitation Theory is very accurate and has tremendous success with predictions however it was long known that it failed in some cases. Einstein came along and filled in that gap with a more accurate theory, General and Special Relativity.
Slice and dice, Occam’s Razor now has a corollary in “bad explanations are easy to vary while good explanations are hard to vary”.
One of the four strands of reality in The Fabric of Reality is the strand of “Karl Popper’s epistemology, especially its anti-inductivism and its requiring a realist (non-instrumental) interpretation of scientific theories, and its emphasis on taking seriously those bold conjectures that resist falsification.“
Paths To Knowledge (dot Science) 