More on Skewed Polls

Hard to dislike. But can he take a pass?

Okay, folks can’t stop talking about the polls being wrong. The big problem is that they show a partisan breakdown that suggests a good deal more Democrats will go to the polls this time than in 2004 or even 2008, a banner year for Democrat turnout. This seems intrinsically unlikely.

In fact, there is a second apparent anomaly. The polls seem to show a higher percentage of Democrats voting for Obama, and a lower percentage of Republicans voting for Romney, than is the historic norm for either party. There are always some party members who break ranks, but this is usually much less common among Republicans. This year, the polls suggest a big crossover vote for Obama.

Let’s dismiss the possibility of a vast left-wing conspiracy. Yet there is definitely an anomaly here. Rasmussen has been keeping tabs on party affiliation, and he has seen no sudden move in the electorate towards the Democrats. In fact, the Republicans’ support is above that of the Democratic Party—except in these polls, when asked about the Presidential election. What else might explain this?

Here’s a thought: might these anomalies be signs that we are about to see a sizable “Shy Tory” effect on voting day? That is, people are telling pollsters they are Democrats in this poll, when they are really Republicans or Independents, and we can accordingly assume that at least the same proportions, probably more, are saying they will vote for Obama when they really intend to vote for Romney. And they are doing this because it is more socially acceptable, and they are concerned with what the pollster might think of them.

The original Shy Tories Coming Out Party.

It just feels right to me. Granted that we did not get a significant “Bradley” or “Shy Tory” effect in 2008—the voting fit the polls fairly well. But this time may be different. There was good reason then for people to really want change; the fact that it was more socially acceptable to support Obama might not have mattered much, because folks wanted to support him anyway. This time is the time that dynamic might well have changed. More people are less happy with him. We know that. But because he is black, voting against him now, especially among those who voted for him last time, might look to them like racism. And he seems to be a nice guy; I can see a lot of people deciding he cannot handle the job, but because he is such a nice guy, they do not want to say it outright.

How many points does this give Romney that the polls do not show? Any figure is a guess, but my guess is about five. A tied race.