Did the Clinton campaign really stop polling at the end?
By STEVEN SHEPARD
09/29/2017 01:38 PM EDT
An explosive allegation from prominent Democratic pollster Stan Greenberg last week is sparking a new wave of criticism and recriminations about Hillary Clinton’s campaign.
The charge — that the campaign went the final three weeks of last year’s presidential election without polling the battleground states — amounts to an accusation of campaign negligence that resulted in Donald Trump’s election.
It’s laid bare the emotions still gripping many Democrats nearly 11 months after Clinton’s defeat. And it has exposed a simmering — and often generational — debate over the extent to which Democratic campaigns should rely on advanced, Obama-style analytics at the expense of more traditional methods. The Clinton campaign itself divided along these lines.
“Astonishingly, the 2016 Clinton campaign conducted no state polls in the final three weeks of the general election and relied primarily on data analytics to project turnout and the state vote,” Greenberg wrote in The American Prospect, published online late last week. “They paid little attention to qualitative focus groups or feedback from the field, and their brief daily poll didn’t measure which candidate was defining the election or getting people engaged.”
That divide between pollsters and those advocating for more campaign analytics has bubbled beneath the surface in Democratic politics for a decade. One of the fault lines is generational — often pitting experienced campaign pollsters against younger data scientists.
Barack Obama’s campaign was credited with successfully marrying those two approaches, especially in 2012.
Democratic pollster Mark Mellman — who polled for then-Sen. John Kerry’s 2004 presidential campaign — said that while Greenberg’s article focused on Clinton’s 2016 campaign, some of the themes transcend last year’s presidential race.
“It is the typical pointing of fingers on one hand, but there is an underlying issue that lies behind it,” said Mellman. “There has been an ongoing debate. … And that debate has been occasioned by some analytics firms that have advertised them as better than traditional pollsters without empirical evidence.”
While pollsters largely fault Clinton’s campaign for prioritizing analytics over polling for much of the race, most acutely at the end, many campaign staffers largely view the battle between pollsters and analytics as a settled issue.
“This is a generational battle that I thought was over,” said a Clinton campaign official who requested anonymity to discuss the campaign’s data operation. “[Obama’s campaigns] made spending decisions based on analytics surveys, too.”
In the public statements campaign staffers have made — and in Clinton's own book — most of the blame is placed on exogenous factors, rather than the data or its polling decisions.
Forty-eight hours after the election, Nayak sent a memo to senior campaign staff, arguing that James Comey, the then-FBI director, threw the election to Trump by re-opening — and then closing — the investigation into Clinton’s use of a private email server.
In an interview this week, Nayak reiterated that the late-breaking events in the race meant that any polling — whether message or analytics surveys — wouldn’t have predicted Clinton’s defeat.
“There’s empirical evidence that if the election were held on Oct. 27 — even if you have the same amount of error in everyone’s polling — that she would have won,” Nayak said. “Our sense was that despite that late movement, we were still positioned to win.”
Greenberg stands by his basic assertion: that suspending even some of its state polling in the closing weeks of the campaign was foolhardy, and that more diagnostic polling might have seen the trouble brewing.
“I don’t understand,” said Greenberg. “Real elections — in which you win or lose — you get every state that you can win, and you look as closely as you can. And I think it’s malpractice not to have real state polls right until the end.”
Did the Clinton campaign really stop polling at the end?
By STEVEN SHEPARD
09/29/2017 01:38 PM EDT
An explosive allegation from prominent Democratic pollster Stan Greenberg last week is sparking a new wave of criticism and recriminations about Hillary Clinton’s campaign.
The charge — that the campaign went the final three weeks of last year’s presidential election without polling the battleground states — amounts to an accusation of campaign negligence that resulted in Donald Trump’s election.
It’s laid bare the emotions still gripping many Democrats nearly 11 months after Clinton’s defeat. And it has exposed a simmering — and often generational — debate over the extent to which Democratic campaigns should rely on advanced, Obama-style analytics at the expense of more traditional methods. The Clinton campaign itself divided along these lines.
“Astonishingly, the 2016 Clinton campaign conducted no state polls in the final three weeks of the general election and relied primarily on data analytics to project turnout and the state vote,” Greenberg wrote in The American Prospect, published online late last week. “They paid little attention to qualitative focus groups or feedback from the field, and their brief daily poll didn’t measure which candidate was defining the election or getting people engaged.”
That divide between pollsters and those advocating for more campaign analytics has bubbled beneath the surface in Democratic politics for a decade. One of the fault lines is generational — often pitting experienced campaign pollsters against younger data scientists.
Barack Obama’s campaign was credited with successfully marrying those two approaches, especially in 2012.
Democratic pollster Mark Mellman — who polled for then-Sen. John Kerry’s 2004 presidential campaign — said that while Greenberg’s article focused on Clinton’s 2016 campaign, some of the themes transcend last year’s presidential race.
“It is the typical pointing of fingers on one hand, but there is an underlying issue that lies behind it,” said Mellman. “There has been an ongoing debate. … And that debate has been occasioned by some analytics firms that have advertised them as better than traditional pollsters without empirical evidence.”
While pollsters largely fault Clinton’s campaign for prioritizing analytics over polling for much of the race, most acutely at the end, many campaign staffers largely view the battle between pollsters and analytics as a settled issue.
“This is a generational battle that I thought was over,” said a Clinton campaign official who requested anonymity to discuss the campaign’s data operation. “[Obama’s campaigns] made spending decisions based on analytics surveys, too.”
In the public statements campaign staffers have made — and in Clinton's own book — most of the blame is placed on exogenous factors, rather than the data or its polling decisions.
Forty-eight hours after the election, Nayak sent a memo to senior campaign staff, arguing that James Comey, the then-FBI director, threw the election to Trump by re-opening — and then closing — the investigation into Clinton’s use of a private email server.
In an interview this week, Nayak reiterated that the late-breaking events in the race meant that any polling — whether message or analytics surveys — wouldn’t have predicted Clinton’s defeat.
“There’s empirical evidence that if the election were held on Oct. 27 — even if you have the same amount of error in everyone’s polling — that she would have won,” Nayak said. “Our sense was that despite that late movement, we were still positioned to win.”
Greenberg stands by his basic assertion: that suspending even some of its state polling in the closing weeks of the campaign was foolhardy, and that more diagnostic polling might have seen the trouble brewing.
“I don’t understand,” said Greenberg. “Real elections — in which you win or lose — you get every state that you can win, and you look as closely as you can. And I think it’s malpractice not to have real state polls right until the end.”
Did the Clinton campaign really stop polling at the end?
