Now granted, I'm not a professional pollster. Hell, I've never even taken a statistics course. But I do have a brain in my head, and I have been known to make tolerably good use of its grey matter from time to time. So you can imagine my utter shock and disbelief when I heard David Paleologos, pollster for Suffolk University, make the following remarks:
"In places like North Carolina, Florida, Virginia - we've already painted those red. We're not polling any of those states again. We're focusing on the remaining states."
Please follow me for a transcript and further discussion below the glowing ember of raging hot Kos.
So this presumptuous little fellow made his grand sweeping pronouncement while interviewing with none other than the grand master bloviator himself, Bill O'Reilly. Here is the transcript in full:
O'Reilly: [Asks a question about other states, such as Pennsylvania, Iowa, Nevada, etc]
Paleologos: ...I think In places like North Carolina, Florida, Virginia - we've already painted those red. We're not polling any of those states again. We're focusing on the remaining states--
O'Reilly: That's interesting. Let me stop you there. So you're convinced - your polling ageny is convinced - that Florida, North Carolina, and Virginia are going to go for Romney?
Paleologos: That's right, and here's why. Before the debate the Suffolk Poll had Obama winning 46-43 in the head to head number - a poor place to be for a couple of reasons. Number 1, his ballot test, his head to head number, was under 47 before the debate. And it's very difficult when you have the known quantity, the incumbent, to claw your way up to 50. So that was a very very poor place for him to be. And so we're looking at this polling data not only in Florida, but in Virginia and in North Carolina, and it's overwhelming.
First, let me put on a tin foil hat for a moment. Bill-O needs to brush up on his choreography a little. Because it was fairly transparent that the only reason he had this guy on his show in the first place was to drop this provocative bomb. He didn't just invite a pollster on his show without knowing in advance what kind of data he had to share. So to begin the interview by talking about "oh.... I don't know. Let's just by chance pick a handful of states other than the ones you're about to talk about, so I can appear shocked - SCHOCKED - by your findings." Yeah, just a tad disingenuous.
OK, now on to the subject matter at hand. So Mr. Paleologos, a very serious pollster who has no interest in making a big splash with controversial news, tells us that without even going back into the field post debate in Virginia, Florida, and North Carolina, he is absolutely convinced that these three states are voting for Romney.
Because it is not possible that whatever movement we're seeing towards Romney right now is an ephemeral bounce? Because yours is the only poll in existence, and no others showed Obama exceeding the 46% mark in any of these states? Because two nights ago PPP didn't show Obama leading with 50% post-debate? Because, oh I don't know, even if you're right and a huge shitbomb just exploded in those three states, other mitigating factors can't impact the race in the next four weeks?
Nope. Those three are done. No more polling.
OK. So even if there were a somewhat justifiable rationale for pulling all polling from these states, is there no hint of journalistic integrity in you at all? Do you have no sense that by making a pronouncement like that, you are actually creating a media narrative that impacts the state of the race? <----- [stupid rhetorical questions]
Is there any reason at all that anyone can think of, any rational and understandable scenario, where a pollster 4 weeks before an election would declare 3 states that have been very much in play for the entire year, and for most of this time trending in the President's favor, would blithely write them off with absolutely no justification other than appealing to numbers from previous polls? Now granted, as I said I have no background in statistics. But is past performance any reliable indicator of future performance? Pretty sure I heard somewhere that past performance has zero bearing on future outcomes.
This fellow needs to be held accountable for this one. No, I'm not talking about pitchforks and torches - either literally or proverbially. But he needs to be held to account. If you're going to make extraordinary claims without extraordinary evidence, then when and if you are proven wrong there needs to be some accountability. In other words, no media outlet (except maybe Fox) should treat any polling analysis from Suffolk University under the direction of David Paleologos with anything but extreme skepticism.