I interpreted it as a potential explainer for why this sort of result is unlikely to spread broadly.

Do you have knowledge of, eg of New Hampshire (which is mentioned as a counter example in the article?)