Please, give us more pollsters and fewer pundits
By Erik Wemple,
On MSNBC’s “The Daily Rundown” on Wednesday, host Chuck Todd was presenting a variety of electoral college outcomes for the Nov. 6 presidential election. With the aid of a clever interactive monitor, Todd was shifting around various battleground states to explain Mitt Romney’s most plausible route to a White House-clinching 270 electoral votes.
In the middle of it all, his voice exasperated, he issued this caveat: “When I’m doing this, I’m not saying this is where NBC says the race is right now. I’m going through scenarios, so don’t overreact on Twitter.”
Compounded by a treatable case of socialmediaphobia, Todd was displaying a bit of old-fashioned caution, ducking behind the curtain of “I’m just a reporter, folks.” How out of touch. Campaign 2012 has seen news outlets go ever more deeply into making news, not merely reporting on it. They don’t just conduct polls, as they have for years. They have embraced the art of computer modeling, generating a constantly revised picture of the national political scene.
More noise than illumination, you might suppose. Perhaps, but only if you ignore all the noise that the media’s long-standing pundit-centric product has churned out for decades.
Watch for it: Just as fact-checking operations have gone viral across journalism in recent years, modeling and forecasting franchises are poised for multiplication. These days, the New York Times, Real Clear Politics and the Huffington Post run highly trafficked poll-aggregation machines offering a look at toss-up states, who’s got the lead and a lot of other stuff you didn’t realize you were curious about. Other organizations that matter in political coverage — from the major networks, including NBC, to cable outlets to newspapers and universities — sponsor their own polling, offer their own number-crunching services on polls or both.
Bruises attach as easily to the pollsters and forecasters as they do to the fact-checkers and political reporters, all of whom sustain bias allegations and general nastiness in the course of business. The hazard of these occupations is that, at some point, you’ll have to issue information that displeases one side or the other. When you do, there are a bunch of people on Twitter ready to overreact.
Nate Silver, the brain behind the mile-deep polling analysis at the New York Times’ FiveThirtyEight blog, has made news this election cycle as both modeling guru and punching bag. Silver’s blog aggregates and analyzes massive amounts of poll data and organizes them into clean, snapshot impressions of who’s up and who’s down in the presidential race and other contests. The FiveThirtyEight model has consistently favored President Obama over Romney, creating an obvious opening for critics. MSNBC morning host Joe Scarborough, in a remark that reflects many conservatives’ reactions to Silver’s oeuvre, recently suggested that the poll aggregator was an ideologue and a joke.
The swell of Silver-centric commentary in the campaign’s home stretch leaves search engines with a bit of sorting to do. When he’s not getting slammed for putting out results that favor Obama, he’s being applauded for his quantitative rigor or his understanding of his craft. Politico, AtlanticWire, BuzzFeed, The Post and others have all weighed in.
Knocking those who deal in polling data and voter demographics isn’t a towering intellectual challenge. There’s always disparity in survey numbers. In 2008, for example, pollsters whiffed on the New Hampshire Democratic primary, showing then-newcomer Obama with the edge over eventual winner Hillary Rodham Clinton.
Pollsters have never promised precision. That’s why their surveys come equipped with margins of error. Paul J. Lavrakas, president of the American Association for Public Opinion Research, cautions that polling is “not pure science . . . it’s subject to a lot of threats,” including imprecision and “biases.” For example, pollsters have yet to lick the problem of accounting for the growing number of cellphone-only citizens, which increases the inaccuracy of surveys that tend to reach land-line users only.
To its eternal discredit, polling is never omniscient, especially in a tight race such as this one. “The interesting dynamic in this race is the number of toss-up states has expanded in the last three weeks,” says John McIntyre, founder of Real Clear Politics. “It reflects the uncertainty out there. We don’t know what’s going to happen. No one knows what’s going to happen.”
The churn of polling results, too, feeds a fast-twitched Internet not known for sowing a deep understanding of our country. State polls, national polls and surveys of all kinds keep landing, multiple times per day, giving Linkville just what it’s built for: a never-ending string of revisions, corrections, annotations and amplifications, each one worth more page views.
Yet whatever its drawbacks, a media world populated by more and more Nate Silvers and their reams of data promises a brighter future than the one to which Americans had resigned themselves: the world of the pundit. As Silver himself says, “I think we represent a counterweight to a lot of the BS, frankly, that you hear in the mainstream media.”
One example here: “momentum,” a poisonous word for Silver but one frequently used by pundits to describe a campaign’s ups and downs. “When the term ‘momentum’ is used, I think that’s a red flag that the coverage you’re reading is suffering from bias,” Silver says, noting that the bias could take any form, from a pundit’s political leanings to simply a desire for a close contest that’s more fun to cover.
Everyone agrees that Romney secured “momentum” around the time of his shellacking of the president in the first debate on Oct. 3. But how long did it last? In an Oct. 25 post, Silver argued that if public opinion mattered in computing momentum, Romney’s version would have petered out a week or two after the debate. However, he said, pundits kept referring to it as an active dynamic in the race for weeks.
On the “Today” show on Oct. 27, Scarborough minted this take on the election: “So for Romney, I think he’s going to be hoping more for this momentum that is sweeping from the first debate to continue forward and carry him over the finish line. And it’s just momentum versus the Obama ground game.”
There’s some perfect punditry for you. In 40 words, Scarborough managed to weave a few strands of conventional wisdom into a compelling-sounding, even-handed wrap-up of where things stand.
Silver himself sometimes envies the wishy-washiness afforded to pundits. “I’d be more comfortable if Obama were a slight favorite [in his modeling]. . . . But that’s not where the math — it’s not where I think it takes you,” he says.
On Thursday, Silver’s model was giving Obama a 79 percent chance of winning the election. The luxury of his contraption is its built-in defense against being wrong. He’s not calculating the size of an Obama victory, only the probability that the president will end up on top. If Obama loses, Silver can always say, “Well, that’s okay, my model gave Romney a 21 percent chance of pulling it out.”
When Silver says that 80 to 90 percent of punditry contributes nothing of journalistic value, his modeling is being charitable by about 10 percentage points. On television, the goal is to panel up and talk, with the hope of stirring a disagreement that gets shared like crazy on Facebook. On the Web, the goal is volume, and nothing yields volume quite like political analysis. Politico, the king of volume and analysis, printed this paragraph before the debate between Vice President Biden and Paul Ryan:
“The best case scenario for each? A clear win. The worst case scenario for each? A clear loss. The murkiest scenario for each? A debate that gets fought to a draw, which the press will interpret in different ways but which also likely won’t stop the GOP ticket’s momentum.”
That’s important context. Because no matter your view of polling, no matter how biased you deem the people behind the questions and the modeling, think of the choice: impression-based horse-race coverage by pundits vs. data-based horse-race coverage by statisticians.
Given Silver’s new fame, there’s an 84 percent chance that media outlets will christen at least 10 new poll-modeling gizmos in time for election 2016. “I think I’m taking advantage, in part, of an underserved market,” Silver says. “The fact that I get kind of undeserved attention, both negative and mostly positive, would suggest there’s a market deficiency — that you have a gang of maybe 500 pundits.”
And not enough modelers.
Much has been made of the possibility that pollsters and even modelers will be proved wrong come Nov. 7. Lavrakas says that outcome would be a “major setback” with “dire consequences.”
Silver has a milder take. “I know as a matter of practice that I’m going to have more opportunities if my prediction looks good and fewer if it doesn’t,” he says.
When’s the last time a cable pundit had to face such consequences?
I’m more forgiving on the forecasting models. If the model doesn’t work, tweak it. If that one doesn’t work, tweak it again. But keep the reality-based analysis coming. I’ll take a deeply explained essay on the eccentricities of New Hampshire political forecasting or a disquisition on the house biases of polling organizations over conventional punditry every time.
Erik Wemple reports and writes about the media in his Washington Post opinion blog.