But Mitchell and his cohorts claim that the polling methods favored by mainstream outfits like Monmouth and Quinnipiac are flawed, and purposely so. “All the polls you hear out there are basically just garbage,” he told me when reached by phone Monday.

He cited as proof of this belief that supposedly random samples of Americans always seem to include more self-identified Democrats than Republicans, which he says is on purpose.

But those polls tend to survey an equal number of registered Democrats and Republicans, it’s just that when those people answer the phone, they sometimes informally switch their party affiliation. Pollsters are quick to point out that the phenomenon doesn’t mean the method is unscientific, but that voters are fickle.

In response to criticism from people like Mitchell, Monmouth University’s Patrick Murray recently wrote that his sample was, “evenly divided between Democrats and Republicans” but the registered Republicans were more likely to say they were Democrats. “The question you should be asking yourself, in light of events over the past few weeks,” Murray said, “is why that might be so.”