As a feud, it does not rise to the level of Lyndon Johnson versus Bobby Kennedy or even Jack Benny’s radio war with Fred Allen. But, still, anyone organizing a post-election panel discussion might be wise to put a few chairs between New York Times polling guru Nate Silver and Frank Newport, the editor-in-chief of Gallup.
The triggering events were probably the blunderbuss record of Gallup’s national presidential polling and Silver’s running critique of their surveys. But the most provocative moment came when Newport warned in a post-election commentary that because of the popularity of aggregators like Silver, “We could see significantly fewer polls conducted in the 2016 election—at the state and national level—than in this election.”
Newport’s argument is instantly familiar to anyone who recalls the rise of The Huffington Post: “It’s not easy nor cheap to conduct traditional random sample polls. It’s much easier, cheaper, and mostly less risky to focus on aggregating and analyzing others’ polls.” The issue is not, as some of Newport’s critics have claimed, that polling aggregation has intellectual validity far beyond The Huffington Post recycling news stories. Rather, it is that analysts like Silver require a steady diet of other people’s expensive polls to perform their feats of electoral prestidigitation.
Instead of being frightened by Newport’s nightmare vision of a 2016 presidential campaign with reliable public polling cut back to 1970s levels, it seemed at first blush to be rather beguiling. No longer would it be possible to purport to cover a campaign without straying far from a cable TV green room by merely reciting the polling averages from Real Clear Politics. Suddenly, old-time journalistic techniques like interviewing voters in diners and drinking with county chairmen in swing states would come back in vogue.
Part of the romance of covering politics is getting out of your comfort level and talking with people that you would not normally encounter in civilian life spent among like-minded insiders in Washington or New York. So I will confess a nostalgic bias against the bloodless 21st century traditions of inaccessible candidates, spin-cycle campaign aides and statistical analysis trumping on-the-ground reporting.
Yet the more that I thought about it, the less that I wanted to go back to an era when the release of a Field Poll in California or a Columbus Dispatch Poll in Ohio was the major political event of the week. I did not crave an alternate universe in which Marist College and Quinnipiac University were famous for their football programs. No reporter—whatever his or her Luddite tendencies—wants to opt for a campaign cycle in which there is less reliable information rather than more.
The big point is that polling would not suddenly vanish, but rather that the media would be practicing unilateral disarmament. Campaigns would still predicate every major decision on their own surveys, but the contours of that public-opinion landscape would remain opaque for reporters.
Not long ago, leaked internal campaign polls were a staple of campaign coverage. These polls were usually a form of disinformation (perhaps the only upbeat poll all month), but credulous reporters went with them because they were often the only game in town. That has changed with the growing sophistication of swing-state polling. The 14 Pennsylvania polls published during the final month of the 2012 campaign made it a lot harder for Mitt Romney’s strategists to claim that they had a secret plan to win the state. Take away most of those surveys and we would have seen breathless pre-election headlines like, “Secret Romney Poll Shows Mitt Up by 6 in PA”.
I have been re-reading (as every campaign reporter should) Tim Crouse’s enduring classic, The Boys on the Bus. Even though the 1972 campaign world of David Broder, Johnny Apple and Walter Mears is familiar in many ways, there were comparatively few national or state polls during the Nixon-McGovern race.
As a result, most reporters did not anticipate the extent of the 49-state landslide in which Nixon racked up 61 percent of the vote. Liberal columnist Mary McGrory, who was always an indefatigable reporter, was gulled by a leaked late October McGovern canvass sheet showing the South Dakota senator supposedly fending off Nixon in the blue-collar Detroit suburb of Hamtramck. New York Times reporter Adam Clymer theorized from his own door-knocking in Ohio that voters were embarrassed to admit they were secretly backing McGovern. Smart reporters on the McGovern plane thought that Nixon might win by 5 percentage points, and not, as it turned out, by 23 points.
This is what happens when—devoid of reliable polling—political reporting is governed by instinct, hunch and whispered campaign plane conversations. Maybe there is romance in resisting the gimlet-eyed clarity of regression analyses of the state surveys. But given a choice of a campaign shaped by polls or one in which reporters try to divine the sentiments of the voters through the size and enthusiasm of crowds, please sign me for Politics 2012.