What can one say? If one strips away the table-pounding about sound science, Freeman basically said something trivial: the exit poll results differed from the official returns, beyond random sampling error. He probably didn't intend it as a straw man assertion. But various gee-whiz statistics along these lines are routinely cited as evidence that the election was stolen -- and they aren't. So, maybe the statistic is irrelevant and/or a straw man assertion, not necessarily both. But if it is only irrelevant, then why are we arguing about it? Hmm. Does anyone think that RFK cited this figure to set up a teaching moment about sources of survey error? I don't. On the contrary, he seems to do everything he can to convey the impression that non-random survey error is unthinkable.

Two paragraphs earlier, RFK asserts, "Against these [exit poll] numbers, the statistical likelihood of Bush winning was less than one in 450,000." The "Tale of the Exit Polls" chart says, "Since the poll results were beyond the margin of error, Bush's odds of victory were less than one in 450,000. But when the ballots were tallied, the four states 'flipped' to Bush...." Irrelevant? a straw man assertion? outright gibberish? You decide. (These "odds," of course, assume that the polls are guaranteed bias-free.)

Let's try an historical analogy. The 1936 Literary Digest poll, with over two million respondents, gave Alf Landon an estimated 57% of the popular vote. Since the poll results were far beyond the margin of error, FDR's odds of beating Alf Landon were longer than my calculator can figure. But when the ballots were tallied, state after state "flipped" to Roosevelt. As much as we can say in sound science that something is impossible, it is impossible that the 1936 discrepancies could have been due to chance or random error. So what?

Ah, but isn't exit polling an "exact science" with "exquisitely accurate" results? Well, no, as Manjoo explains. As far as I can tell, RFK mainly uses his statistical factoids to convey an aura of scientific certainty where it isn't warranted. False certainty should not be part of the progressive agenda.