GOP strategist: Carolyn Goodman is the favorite but she’s vulnerable as “quasi-incumbent,” under 50 percent in poll

Saturday, Feb. 5, 2011 | 1:46 p.m.

That's what John Yob, a top adviser to John McCain's 2008 White House campaign and a cog in the Sharron Angle team, told me today about that mayoral poll he oversaw. He had some interesting thoughts when I asked him about the race and on the accuracy of robopolling (Interactive Voice Response, or IVR). Here's what he said:

"I would argue that IVR is more accurate than live polling for small universe elections if the sample size is higher. If you are polling an election with 70,000 votes, and you get a sample size of 900 people who traditionally vote in these elections, you have a good sized sample of the electorate.

"In general, I think we have moved past the point of questioning IVR for tracking surveys. Rasmussen, SurveyUSA, PPP, etc have proven to be some of the more accurate over the last few cycles.

"That said, it is much more difficult when you try to do 20-30 question benchmark surveys because of the drop off.

"I use IVR for all my tracking work in campaigns across the country, including McCain in 2008 Primaries.

"BUT....I think you are getting at something relative to turnout. ANY polling is only as accurate as the model it uses for turnout. Common sense dictates that people who have voted in municipal elections in the past are likely to vote in municipal elections in the future, particularly if they indicate that they are likely to vote.

"The biggest question, that I haven't seen anyone report on, is whether or not the tea party energy of 2010 that dominated Primary ballot boxes in small-universe states across the country (AK, DE, NV, moreso than CA, TX) will show up in municipal races.

"Clearly low turnout, and small state, elections are more susceptible to a tea party (or any non-traditional turnout) surge.

"Can you imagine if a candidate for Mayor in a contest that can be won with 30,000 votes had a tea-party fueled surge of support like Angle, O'Donnell, or Miller? More importantly, such a surge would only show up in polling data if the turnout model accounted for the possibility.

"My company handled GOTV in the primaries for both Angle, O'Donnell, and Snyder. It would be interesting to test the same methodologies in a low-turnout municipal race with a candidate who could inspire non-traditional municipal voters.

"The main thing that stuck out to me is that there is room for a conservative, or GOP-leaning, candidate to emerge in the primary and become the alternative to Goodman. The Dem leaning candidates are breaking up their vote and it probably hurts all of them. Goodman is the favorite but she is vulnerable because she is a quasi-incumbent, well below 50% and there is potential for non-traditional municipal-voter surge."