I was previously a constitutional law and civil rights litigator and am now a journalist. I am the author of three New York Times bestselling books -- "How Would a Patriot Act" (a critique of Bush executive power theories), "Tragic Legacy" (documenting the Bush legacy), and With Liberty and Justice for Some (critiquing America's two-tiered justice system and the collapse of the rule of law for its political and financial elites). My fifth book - No Place to Hide: Edward Snowden, the NSA and the US Surveillance State - will be released on April 29, 2014 by Holt/Metropolitan.

Thursday, November 09, 2006

Hugh Hewitt shows how Bush followers literally deny reality

(updated below)

In the three weeks prior to the election, Hugh Hewitt was repeatedly insisting that Republican candidates were tied or ahead even when the consensus of polls showed those candidates were actually behind, sometimes by substantial margins. He wasn't just predicting that the GOP candidate behind in the polls would win (there's nothing wrong with being hopeful). He was doing something completely different -- he was insisting that the GOP candidates who were behind in the polls were, in fact, ahead in the polls.

In response to e-mails he received, he condescendingly explained his "thinking" behind this outright distortion of reality:

I get a lot of e-mail asking me why I point to polls like the one favoring Steele when I discount some polls favoring some Democrats.

Because this question comes mostly from lefties, I will pause to explain in as uncomplicated a fashion as possible.

Polling methodology and models favors Democrats.

So polls that show Republicans tied or ahead I see as indicating a race in which the Republican is in the lead.

Polls that show a Republican within striking distance I see as a poll indicating a dead heat.

It shouldn't be that hard to grasp, even for a lefty.

So, like everything else in the world -- from war zone reports to intelligence assessments and everything in between -- polling data can be ignored and disregarded at will (when it's unpleasant) because it is oh-so-unfairly biased against the Republicans. Hewitt took the data that he didn't like, literally changed it in his own mind to make it more pleasant, and then embraced the fictitious data as his reality. And he expressly acknowledged doing so by insisting that data is biased.

In fact, polling data for this election was remarkably accurate. It predicted the vast majority of races almost completely accurately. In the instances where there was a discrepancy, the discrepancy was almost in every case favorable to the Republicans -- meaning the poll showed Democrats with a smaller lead than they ended up with (or behind by more than they actually lost by).

Following is a chart comparing the final polls of Rasmussen Reports for the 11 Senate races it (and the rest of the country) identified as the "Nation's Closest Senate Races" (Connecticut is the only race excluded here due to its confused party breakdown). The table shows the final RR poll for each race, the actual election results, the differential between the two, and indicates whether the differential favored the Republican or the Democratic candidate:

STATERR Final PollActual ResultDifferential
MD50-45 Cardin54-44 Cardin5% (against Dem.)
MI56-40 Stabenow57-41 Stabenow0%
MO49-48 McCaskill49-47 McCaskill1% (against Dem.)
MT50-48 Tester49-48 Tester1% (for Dem.)
NJ48-43 Menendez53-45 Menendez3% (against Dem.)
OH54-43 Brown56-44Brown1% (against Dem.)
PA55-42 Casey59-41 Casey5% (against Dem.)
RI52-44 Whitehouse53-47 Whitehouse2% (for Dem.)
TN51-47 Corker51-48 Corker1% (against Dem.)
VA49-49 Tie49-49 Webb0%
WA54-42 Cantwell58-39 Cantwell7% (against Dem.)

In the 11 Senate races identified by Rasmussen as "the closest Senate races" (excluding CT), Rasmussen's polls predicted the exact outcome in 2 of them. For the 9 races where there was a disparity, 7 of the 9 disparities were in favor of the Republicans. Only 2 of the 11 races showed a gap in favor of Democrats, and in those two races (RI and MT), the difference was miniscule -- 1% and 2%.

It wasn't just Rasmussen. Polls in general were either remarkably accurate or, to the extent they were wrong, were largely skewed in favor of Republicans (at least in terms of what they predicted versus the actual result). The Real Clear Politics average final polls (which averages the outcomes of multiple polls from around the country) show that for the same 11 Senate races, the polling disparities favored Republicans in 8 0f the 11 races, often by considerable margins. In the 3 races where the disparities favored Democrats, it was by very small margins of 1-3 points.

The point here is not to mock Hewitt for being wrong in virtually every prediction he made. It's natural for most people to have their partisan desires influence their predictions, and predicting races even within the science of polling, let alone without it, is extremely difficult. I wouldn't hold it against anyone to have predicted wrong.

But mere inaccurate predictions isn't what Hewitt is guilty of. It's the willingness -- so common among Bush followers -- not only to ignore, but also consciously to deny and reject, information that undercuts their desired beliefs, and to insist that even the most objective facts are "biased." As Stephen Colbert (who has the hardest job in the world -- satirizing people who are beyond satire) put it at the White House Correspondents Dinner:

Now, I know there are some polls out there saying this man [the Preident] has a 32% approval rating. But guys like us, we don't pay attention to the polls. We know that polls are just a collection of statistics that reflect what people are thinking in "reality." And reality has a well-known liberal bias.

Isn't that exactly what Hugh Hewitt argued? And Hewitt isn't unique in this regard at all. The President himself said yesterday that he really believed Republicans would win the election despite all of the data to the contrary -- and his rattled, disoriented, almost pity-inducing performance strongly suggested that he was telling the truth about that. He wasn't just being optimistic. He saw the polling data but, like Hewitt, decided that it wasn't real, because it was unpleasant, because it conflicted with his perceptions, and because he did not want it to be real.

Nor is this reality-denial syndrome confined to polling data. It is visible in virtually every realm, most glaringly in Iraq. Dismissing unpleasant polling data and insisting that you're winning is exactly the same dynamic as dismissing reports about a civil war in Iraq and insisting that things are going well, or dismissing reports about the inability of inspectors to find WMDs and insisting that they are really there, hidden somewhere. It's one thing to assess reality and draw the wrong conclusions from it or exercise wrong judgment about it. Everyone does that.

But the President and his followers don't just do that. Like children -- or, more accurately, like those who are driven by an unshakable, faith-driven belief that they are Right -- they just ignore reality when it isn't what they want it to be or think it should be, and they just magically invent a different reality that they like better, and then stay there. Hewitt expressly described how he does that, and he is about as pure an expression of the Bush movement as it gets. The President and his followers simply don't accept or live in reality and literally don't believe in facts.

UPDATE: A closely related reality-denying syndrome is the way in which Bush followers are interpreting the resounding rejection by Americans of their Leader and their political movement as some sort of endorsement of all of the particular views of Bush followers - as in: "hey, Republicans were crushed because they weren't extreme enough and didn't pursue my fanatical views enough on Issue X, Y and Z."

With this twisted reasoning in place, the most virulent supporters of the Iraq War are actually claiming that the election affirms that most Americans agree with them. Just ponder that for a minute. Unsurprisingly, Glenn Reynolds is one of the most vivid illustrations of that reality-detached though process, as Blue Texan -- the world's foremost scientific expert in Instapundit deceitfulness -- amply, and amusingly, documents here.

UPDATE II: This post was actually the one to which I meant to link. Although it debunks the election claims made by Reynolds specifically, it is a fairly thorough compilation of the standard (and false) conventional wisdom assertions that always circulate through right-wing punditry and the national media as to why Democrats are so hated by voters.

My Ecosystem Details