Neil Newhouse, Mitt Romney's campaign pollster, suddenly finds himself in an unfamiliar place -- out on a limb.
The survey-taker, who helped build Public Opinion Strategies into the largest Republican polling firm and has a solid reputation among operatives and colleagues, is growing increasingly vocal with reporters and Romney supporters in the campaign's closing days about what he sees as examples of flawed public polling, and his sense of the race -- particularly in Ohio -- as basically even.
Newhouse is not known for seeking attention, but through the course of the presidential campaign, his profile has risen. He has been part of some public state-of-the-race calls, and is a frequent fixture on calls and in conversations with Romney surrogates and donors, telling GOP elites in one discussion last week that a Quinnipiac University survey with the New York Times and CBS was "crap." (Newhouse said he didn't recall using that word.)
Having decamped from Washington to Boston to be part of the campaign last year, Newhouse has gotten comparatively less attention than some of Romney's high command, but he's attained more of a public -- and defensive -- posture in the race's last stages than President Barack Obama's pollster Joel Benenson.
On Wednesday, one of the two men will be called a genius, and the other called a goat, based on assumptions they made about turnout models, especially in the crucial state of Ohio.
This is hardly an academic debate over the nuances of polling science. The basic issue is whether Newhouse's internal forecasts and assumptions about the composition of the 2012 electorate are correct versus the ones made by the Obama campaign, which have tended to look more like public polling. If Newhouse is right, the majority of public pollsters will have egg on their faces. If he's wrong, there will be post-mortems questioning his take.
It is because of Newhouse's strong reputation that many Republicans have believed his assertions that Democrats are misreading the math, and why members of Romney's campaign have stayed positive even when things have seemed publicly to be slipping from reach.
"Neil Newhouse is probably the most respected GOP pollster in the country," said Nick Everhart, president of the Delaware, Ohio-based Strategy Group for Media, who has worked with POS repeatedly. "I think under-appreciated or respected when he pushes back hard on the public polling especially in Ohio, is that I feel like without question there isn't a pollster in this country ever who has conducted or taken as many polls as Neil has in the state of Ohio. He's practically an Ohio-based strategist the amount of work he's done here, so it's VERY hard not to take his comments and critiques of the rest of the public survey data coming out of the state seriously."
Everhart added: "Neil is just not the kind of guy who would be pushing back for the sake of earned media spin, he's pushing back because he's seeing serious flaws in all this public data."
Indeed, amid the partisan fog, what is getting lost is that both sides -- Obama and Romney -- are fiercely convinced that their numbers are the right ones.
"Some people want to believe we're living in a world where the electorate is going to have a partisan composition" like 2008, Newhouse told reporters last week on a conference call, citing the current political environment. "That's a stretch."
"What's fun about Neil is he's kind of the gritty, blue collar, junkyard dog Republican pollster," said Republican pollster Greg Strimple, who believes Newhouse's data is accurate and describes him as generally cautious instead of risky in his assumptions. "He's not a pompous, egotistical type."
At times, the pollster has delved into purer forms of PR during the race, saying at a panel at the Republican National Convention that the campaign would not "let fact-checkers dictate" how they do things.
In a call with reporters last week, Newhouse dismissed a number of the public surveys for an inadequate screen of registered voters versus likely voters, saying they are not properly measuring who will actually turn out. He also pointed out that Romney has done well in with independents in a number of the Ohio polls, and cited the "intensity gap" that a number of Republicans have highlighted.
He has briefed Romney supporters repeatedly, on donor calls and in a series of conversations initiated by Romney backer Ron Kaufman, although in broad terms, according to a source familiar with what's been said.
But typically the pollster has been steadier than that in those settings, pinpointing on one of the calls where the campaign stands with women and independents, broad metrics that he rarely uses to expand into strategic choices. Once, another participant said, Newhouse allowed that 2012 would "not be the same turnout model as 2008," although he has generally declined to say what this year's electorate will look like. Newhouse, not known as a pollster bent on dictating the messaging based on the numbers, tends to give a more circumspect read on the situation, according to those on the calls.
And Newhouse is cleaving to an assumption that Republicans have held for much of the race -- that in a close, roughly two-point campaign nationally, undecideds will ultimately break for the challenger in the closing days. That has not happened yet, with two days to go, but neither have those voters gone for the president.
"Our assumption is certainly not 2008 and it's not 2004, [when] we're looking at what potential turnout is going to be, it's most accurate to say it's a blend of 2004 and 2008," said Vaughn Flasher, an Ohio consultant who has worked repeatedly with Newhouse and vouched for his data.
The Obama team argues that the broader demographic makeup of the 2012 electorate will more closely resemble 2008 -- with high turnout among African-Americans, Latinos and other groups that favor the president. Democrats argue that the percentage of the white vote has declined at a steady, and fairly predictable, level since 1996, reflecting larger demographic shifts in the population. Romney has not made a strong effort to appeal to Hispanic votes nationally, after tacking hard-right during the primaries.
The public polling has been criticized repeatedly by Republicans for partisan lean, and a WSJ/Marist/NBC poll of Ohio likely voters released late Friday night was no exception. Yet the basic assumptions that Benenson and Newhouse make in their surveys, and where the differences lie, fall less on party weighting, a self-identifying process by people surveyed, than the demographic composition they expect of the electorate.
"I've never heard Joel, on the record, saying, 'Gosh, I think the Q-poll is the real Word of God, and I think that's where the race is,' " said Bill McInturff, who helped build POS with Newhouse, and who did the polling for John McCain in 2008. "What I've heard both of them both publicly and in the private political circles [do] is [to] referee and say look there are polls out there that are outliers and that are just very different than our internal stuff."
The difference, McInturff said, is what percentage of the white vote will make up the larger composition of the 2012 electorate, compared to the coalition the Obama team believes it will have. This is especially true in Ohio, where Obama is doing better than he is in some other states with white working-class males, largely because of the auto bailout's popularity.
McInturff added, "We're talking about campaigns that have huge tons of data that are having a margin of error fight...we just happen to be having an election where that margin of error [determines] who becomes president."
Newhouse came up with McInturff and two other POS partners through the Ronald Reagan-tied Wirthlin Group. He has been involved in countless gubernatorial and statehouse races, a number of them in Ohio. He did not do Romney's polling in his 2008 campaign, but frequently works with Romney senior strategist Stuart Stevens, and was not a surprising pick.
"Neil moved up to Boston on day one of the primary process. He was all in from the very beginning," said Romney campaign manager Matt Rhoades. "People on the staff don't view him as 'the pollster,' they view him as a teammate. He quickly became a leader and a staff favorite on the campaign."
Newhouse, a low-key presence at the Romney campaign's Commercial Street headquarters -- and about whom most sentences with operatives begin with the words "Neil is a good guy..." -- has been adamant in internal conversations in the campaign and with supporters that the race remains basically even, despite the bulk of public polling showing a small but steady Obama lead in Ohio, a state critical to both campaigns right now, according to sources familiar with the discussions. His internals had Romney slightly ahead toward the end of last week, multiple sources said.
Ed Goeas, of the Tarrance Group, agreed, saying, "The difference between [firms like his and POS] and the newspaper polls, public polls, is that if we're wrong we lose clients and we don't get any new clients. If the newspaper polls are wrong, they get another news story about the numbers moving. And so we have to be very specific and very right where actually it serves to [public pollsters'] benefit if in fact they're letting the sample float...He's one of the best and certainly has to be seen as the premier pollster on the Republican side this year."
POS has been involved in countless races -- one rival operative described the high-volume firm as "the McDonald's of polling" outfits. The firm's Glen Bolger, involved in the fire-walled, independent expenditure-entity related to POS, put Minnesota in a dead heat at the presidential level in an AFF survey this weekend.
Newhouse gave Scott Brown a large dose of credibility early in his special election race against Martha Coakley for the late Ted Kennedy's Senate seat. And his colleagues were was part of the Gov. Bob McDonnell team in 2009, giving him a window into some of the GOP wave energy of that cycle.
Both presidential campaigns rely on in-state data help -- the Democrats have three streams of data, and the Republicans have a few pollsters aiding them, among them American Viewpoint, according to sources familiar with the numbers.
In terms of the 2012 math and whether Obama can achieve levels of turnout among Hispanics, unmarried women and black voters that he needs, it's a guess on either side, according to Republican pollster Whit Ayres.
"Nobody really has any tool that will give you a precise measure," said Ayres, who polls for the group Resurgent Republic, and said that Romney must rely on enthusiasm if the racial proportions more closely resemble 2008, which has been a Newhouse argument. "And that's the fundamental difference among the pollsters right now."
"Neil Newhouse is a great guy. We share a passion for sports and politics and I'm sure when this is all over we're going to sit down over a few beers and kick around what we were both saying and what we thought it meant," said Benenson.
Correction: An earlier version of this piece listed McDonnell's election as taking place in 2010.
The Gazette now offers Facebook Comments on its stories. You must be logged into your Facebook account to add comments. If you do not want your comment to post to your personal page, uncheck the box below the comment. Comments deemed offensive by the moderators will be removed, and commenters who persist may be banned from commenting on the site.