The True Crucible: Whether It's Reproducible

You hear a lot about peer-review, as if it were some kind of academic gold standard. Well, how good is it?

The unspoken rule is that at least 50% of the studies published even in top tier academic journals – Science, Nature, Cell, PNAS, etc… – can’t be repeated with the same conclusions by an industrial lab. In particular, key animal models often don’t reproduce.

There's a science-humor magazine: The Journal of Irreproducible Results. It sounds like they should expand, and those other journals should shrink!

If it fails when you redo,
despite past peer review,
it's probably untrue.