A series of new polls released this week shows the Wisconsin recall to be a dead heat between Tom Barrett and Scott Walker: LINK, LINK, LINK. These polls all show the race deadlocked within the margin of error, with Walker failing to crack the critical 50% threshold in the most recent Mellman poll released today. In addition, every single legitimate public poll also continues to show the Walker-Barrett matchup within the margin of error.
Yet despite this substantial evidence the recall sits at the same dead heat amongst a highly-polarized electorate that has been predicted all along, Republicans and some members of the pundit class continue to propagate a pro-Walker narrative based on flawed public polling.
This week’s installment of public polling is especially absurd, with new data being pushed by Walker and friends from an “independent nonpartisan” firm called We Ask America, libertarian Reason Magazine, and St. Norbert’s University. Let’s take these public polls one at a time.
1) We Ask America. Coming out of nowhere two weeks ago and offering results giving Walker his widest leads on record, this firm claims to be independent and nonpartisan. The reality is that We Ask America is a subsidiary of the Illinois Manufacturers Association. They donate almost exclusively to Republicans, and refuse to release any detail regarding the sample or crosstabs of their single-night robo-polling.
2) Reason Magazine. This poll is a week old, and was first reported by Breitbart News as breaking news late Wednesday night. The libertarian Reason poll is particularly amusing, both in the phrasing of its survey questions and its sample. In addition to using a sample in which 20% of respondents self-identify as 50-50 or less to actually vote, the so-called Independents in the survey break 70-30 for Walker. (Yeah, right.) Right-wing spinners have argued a 10 point Obama lead in the presidential horserace shows this sample to be unbiased, but that includes Libertarian candidate Gary Johnson getting 6% of the presidential vote – need anything more be said about this poll?
3) St. Norbert’s. While the result is within the margin of error and shows a 5-point Walker lead, this poll is actually great news for Tom Barrett—it’s the sample that’s the key. Notice the survey breaks out the sample as including 52% Walker 2010 voters and 38% Barrett 2010 voters. Think about that for a moment: the inputs in the survey are +14 for Walker in terms of their 2010 vote, and the result is Walker +5. That means many self-identifying Walker supporters back Barrett, nevermind the fact that a 52-38 Walker-Barrett 2010 split on June 5th isn’t even in the same time zone as the likely electorate.
The bottom line is simple: Not a single legitimate public poll has shown the Walker-Barrett matchup to be outside the margin of error, and yet the effort of pro-Walker forces to field and release polls based on faulty inputs and assumptions about the electorate to drive the “it’s over” meme continues unabated, with no virtually no criticism of these highly-suspect polls from the media regurgitating them in print.
There are 12 days left in a 28-day general election, and this race remains a dead heat. It will remain a dead heat through Election Day. And as much as Scott Walker’s operatives or the DC pundit class want to declare this race over, nothing could be further from the truth. This race was always going to come down to a base-versus-base battle of turnout for which our movement has been building and organizing for 15 months. On Election Day, it’s that very turnout that will determine the outcome, and we remain confident in our trajectory and continue to execute our program as planned.