Last week, the New York Times (11/8/19) published yet another article about polling and the Democratic presidential hopefuls, a year before the general election: “Democrats in Battleground States Prefer Moderate Nominee, Poll Shows.” In the second paragraph, writers Jonathan Martin and Katie Glueck declared,
As the Democratic candidates intensify their argument over how best to defeat President Trump, their core voters in Michigan, Wisconsin, Pennsylvania, North Carolina, Arizona and Florida are counseling them to pursue a political middle ground.
The New York Times (11/8/19) says that Democrats in battleground states want a “more moderate” nominee, even as many of them say they plan to vote for progressive candidates.
Curiously, though, the polling graphic the Times featured at the top of the piece showed a different story, in which the top progressive Dems combined (Bernie Sanders and Elizabeth Warren), compared to the major candidates branding themselves as “moderates” (Joe Biden and Pete Buttigieg), receive either greater support in those battleground states (Michigan, Wisconsin, Arizona), or the difference between the camps is within the margin of error (Pennsylvania, North Carolina, Florida). Many voters polled have yet to make up their minds, but those figures certainly ought to give any journalist pause before drawing the conclusion that those voters are “counseling” pursuit of the so-called middle ground.
The Times‘ analysis appears to be based on a different question asked in the poll—“Would you prefer a candidate who would be more moderate than most Democrats or more liberal than most Democrats?”—which saw a 55%–39% split in favor of the former.
It’s a very slippery question to hang an article on. The opposite of “more liberal” would be “more conservative,” not “more moderate.” It’s not clear what “more moderate” would mean if not “more conservative”—but far more Democrats identify as moderates than as conservatives (Gallup, 2/19/19), so offering “more moderate” as the alternative almost certainly increases the popularity of that choice.
The question also asks about what kind of candidate someone would prefer, rather than directly asking who they’d vote for. This could be especially problematic in an election where Democratic voters regularly tell pollsters they’re more concerned with a candidate’s “electability” than their ideology, and therefore might “prefer” a candidate they think others will vote for. Again, this will skew the answers.
But more importantly, who cares? As everyone knows, we pay extraordinary attention to these “battleground” states, not because they reflect the will of the country better than other states do, but because their votes matter more than the votes from other states due to our undemocratic electoral college system. So what battleground voters “counsel” candidates to do matters only in the sense that who they will vote for in the general election—not the primaries—matters.
The article seems to be making the leap of logic that voters in these states will be less likely to vote for the Democrat in the general election if a progressive is on the ticket. But there’s a much simpler way to figure that out: Ask people in those states who they’re most likely to vote for in the general election in head-to-head matchups.
And guess what? The Times poll did ask that question! And they published a lengthy front-page article about it earlier in the week (“One Year From Election, Trump Trails Biden but Leads Warren in Battlegrounds”—11/4/19), in which the Times’ Nate Cohn highlighted comparatively low poll numbers for Warren, raising questions about her “ideology and gender” and largely excluding Sanders—who beats Biden in Michigan and trails him within the margin of error in the other states—from the analysis.
(Note that the Times poll is only one poll; other respected polls show all three leading Dems beating Trump in nearly every battleground state. See RealClearPolitics for poll roundups and averages.)
So the paper recycled its poll to tell essentially the same story about Democrats being too far left for the battleground states in a different and more roundabout way, presenting it as if it were new. (There’s no mention of the previous write-up in this article.) And we’re still a year out from the general election, and polls were notoriously fallible last time around. Welcome to media coverage of Election 2020.
Julie, thank you for this well-considered commentary on the polling. Among other points, you say: “[Voter preference] matters only in the sense that who they will vote for in the general election—not the primaries—matters.”
During the 2016 primary season, the right’s Big Lie was that Clinton was more “electable” than Sanders. Clinton crowed about her primary numbers as if this had anything to do with how people would vote in November.
The right’s fraudulent meme worked all too well. Many well-meaning primary voters told me, “I like Bernie but I’m voting for Hillary because she’s more electable.” Then one day I spotted a “November matchup” poll, pitting Clinton and Sanders not against each other but against the GOP. Sanders outperformed Clinton against leading GOP contenders. But a single result by itself is all but meaningless.
Still, I got curious. I began stalking Real Clear Politics (RCP) for every November-matchup poll that included both Sanders and Clinton, so it would always be apples to apples.
Fast forward. From early March into June, RCP reported 150 such polls. I examined each as it showed up. Remarkably, 135 of 150 times Sanders outperformed Clinton, sometimes by a wide margin. Sanders’ performance against the GOP was so consistently strong, I dubbed him “Iron Man Sanders.”
Meanwhile, looking across dozens of polls, Clinton showed disturbing signs of electoral weakness.
As the weeks rolled by, I reported and interpreted these tallies at a site I set up for the purpose, BernieWorks.com. You can still see results there.
One can argue “woulda, coulda” all day long. But the stunningly consistent pattern of November-matchup polls makes it all but certain that Sanders would have wiped the floor with Trump, and today we would live in a very different world.
Why cry over spilled milk? To learn from this bitter and tragic lesson — something Dems seem hellbent on avoiding.
All this came flashing back when I read your comment about looking at November polls rather than primary polls. In 2020 a Dem candidate will face the GOP, not another Dem. That’s what matters.
Amen, brother Ira.
To your list of points I would add only this one: name recognition holds a lot more influence in primaries than it does after a candidate is selected for the general election, when presumably the DNC MIGHT get off its anti-progressive horse and support someone who stands for real, structural change.
Well, the “Hillary 84 percent likely to win” Nate Cohn did that “analysis” for the NY Times. So what do you expect.
He should have been fired by the NYT on Nov 9th 2019.
But I completely understand the mindset of keeping him on staff.
Thanks for catching this, I couldn’t stomach reading the “analysis” closely.
Should expect this sort of rhetoric to only increase, the Dem leadership and their media cohorts (CNN, MSNBC, WaPo, NYT, etc) are clearly pulling out all the stops to prevent a progressive candidate from winning the Dem nomination. Nevermind the results, the last time we ran a bland mealy-mouthed centrist with decades worth of political baggage and scar-tissue against Donald Trump, way back in 2016- surely next time will be different, because… um… reasons, right?
And more importantly, if a “moderate” (i.e conservative) Dem somehow manages to win the White House, we don’t have to worry about any of that medicare for all, anti-corruption nonsense that threatens the status quo favoring wealthy Dem donors and corporate sponsors. The REAL sign that we’re living in Hell World, is not that Donald Trump is president, or that the GOP has gone full-on white supremacist… its that their counterparts from the left (Biden, Clinton, Pelosi, Shumer, etc.) are only negligibly better.
I do NOT believe those CORPORATE POLLS! Democrats do NOT want to move further to the “REICH”, We did NOT “vote” for Trump in 2016 & I will NOT “choose” Biden in 2020!
Our “elections” are a RIGGED FRAUD we do NOT have a “democracy” this is a WARMONGERING, FACIST OLIGARCHY & I will no longer participate in these FAKE, RIGGED “ELECTIONS”!
BOYCOT THE 2020 FAKE, RIGGED, FRAUDULENT “ELECTIONS”!
better not boycott the election if Bernie somehow manages to win the nomination, he’s basically our only hope for meaningful reform
Don’t boycott the elections-go in and vote third-party. If you boycott, the powers that be take that as you’re concerned
The DNC hierarchy is a bunch of “influencers” who listen to other liberal “influencers.” They interpret polls of the “inflencees” to suit their deep-seated prejudices about who is “electable.” As they have consistently been taken to the woodshed by the GOP in many states over the past decade or two, they naturally assume that to be electable candidates must put forward a more conservative issue. That the GOP has been very adept at tilting elections hasn’t caused them to mirror their tactics (voter ID, gerrymandering, voter suppression, negative ads, and perhapseven rigged voting machines. They are stuck in a mindset that leads them to eschew progressive politics in the vain hope that “moderates” will excite voters, so many of whom have had it with conventional politics. They are totally afraid to breathe fresh air.
How can we be “influencers” to people like that? Any ideas?
The answer is pretty simple: Bernie Sanders needs to win the primary by a landslide. If that happens, then we ARE the influencers.
Remember: us not me…..or us not him.
Sanders isn’t joking about that.