According to the latest WSJ/NBC poll, the health care bill is “particularly unpopular in the districts that matter most in the Republicans’ effort to retake the House”:

According to the latest Wall Street Journal/NBC News poll, 52% of respondents said they would be more likely to vote for a candidate who wanted to give the health law a chance to work and then make changes to it as needed, while 45% said they would be more likely to vote for a candidate who wanted to repeal the law entirely and start over.

But in the 92 House districts considered most competitive, that support flipped, with 42% voting for the candidate who wanted to keep the law and 55% voting for the candidate who wanted it repealed.

Josh Kraushaar argues that Democrats who opposed the bill are doing better in conservative districts than those who supported it:

[P]olling recently commissioned by the National Republican Congressional Committee in 65 of the most competitive congressional seats held by Democrats provides more. When participants were asked an open-ended question about what gives them the biggest pause about voting for their sitting member of Congress, a solid plurality said it was health care – ahead of the economy and jobs. In a follow-up focus group in Erie, Pa., with some of the poll’s participants, one of the organizers said it was striking to see how many women’s votes were driven by health care. Some came to the focus group reciting chapter and verse the provisions of the law they didn’t like. Many said they were Obama supporters in 2008, but the more they heard about the health care bill, the more frustrated they became.

The surprisingly strong opposition from women may explain why the latest NYT/CBS poll finds that the  GOP has “wiped out the advantage held by Democrats in recent election cycles among women”:

If women choose Republicans over Democrats in House races on Tuesday, it will be the first time they have done so since exit polls began tracking the breakdown in 1982.

For those who weren’t following the contretemps at the time, FDL did polling in January of this year about the opposition in swing districts to the health care bill.  We used SurveyUSA, a polling firm with one of the most accurate track records in recent years, and they wrote the questions and determined the methodology (as they always d0, contractually):

The results were indisputable: if these members of Congress were forced to vote for the health care bill as-is, their election prospects were in serious jeopardy.

The DCCC, whose internal polling no doubt showed the same thing, went on the offensive to try and discredit the poll.  There were any number of people anxious to help them.

Nate Silver:

Is this supposed to be an impressive finding? If anything, it suggests that the downside to Republican argumentation about the mandate is pretty limited — even in a world where the other side doesn’t get the favor of arguing back.

James L, Swing State Project:

I have no interest in discussing the healthcare portions of this poll (and neither should you).

Chris Cillizza:

National Democratic strategists dismiss the surveys as rank fear-mongering in an attempt to pull the party to the ideological left and note that in several of the polls some of the demographic data — most notably the percentage of voters age 18-34 — was far lower than it is in the voters files for each state.

Alan Abramowitz, Emory University:

In the past few weeks an ostensibly left-wing blog, Firedoglake, has sponsored a series of polls in House districts currently held by moderate Democrats including Indiana 9, New York 1, and Ohio 1. The polls, each of which was conducted by well known automated pollster Survey USA, showed the Democratic incumbents trailing potential Republican challengers by margins ranging from a few points to double digits….[T]here appear to be serious flaws with both the questions and the samples used in the Firedoglake surveys that should give pause to anyone about drawing any firm conclusions about the opinions of voters in these three House districts.

SurveyUSA’s Jay Leve responded to the criticism of  the polls:

The highway to high office is littered with the road kill of political operatives who find it easier to campaign against a pollster than an opponent. If the DCCC has polls showing congressional Democrats doing better, release them.

Here’s what Chris Van Hollen told Alex Isenstadt of Politico:

“Let me just say that with respect to those polls, they are totally off,” Rep. Chris Van Hollen of Maryland, chairman of the Democratic Congressional Campaign Committee, told POLITICO. “We had the opportunity to compare those polls with polls that have been done in a professional manner, [and] the others are very different. So they’re not credible.”

And then there’s this gem:

One House Democratic aide said the party decided to engage in a concerted pushback on the polls because, “in some cases, people had the perception the sky was falling,” and the questions about the surveys “needed to be told.”

We agreed that the the truth “needed to be told,” so we challenged the DCCC to release their own polls, and even offered to poll the districts together.  They refused.  Markos Moulitsas said that the DCCC should release their own polls if they continued to criticize ours, but none of the so-called “experts” who were so very concerned about the integrity of SurveyUSA’s methodology apparently had that level of intellectual curiosity.  None of them expressed any desire to see them.

Then again,  Markos has always exhibited an impressive commitment to honesty and integrity about polling that you always want to see in someone who analyzes polling data.

Instead, the DCCC lied to their members about what a vote on the health care bill would mean, when they clearly knew the truth from their own internal polls.

I’m really looking forward to looking at how well the polls predicted the impact of the health care bill in these districts after the election, and compare SurveyUSA’s methodology with the outcome in each.  I hope those who willingly did the hatchet work of the DCCC will do likewise.