Advertisement

SKIP ADVERTISEMENT

One Year From Election, Trump Trails Biden but Leads Warren in Battlegrounds

Signs that the president’s advantage in the Electoral College has persisted or even increased since 2016.

Despite low national approval ratings and the specter of impeachment, President Trump remains highly competitive in the battleground states likeliest to decide his re-election, according to a set of new surveys from The New York Times Upshot and Siena College.

How Trump fares among registered voters

Trump vs. Biden Sanders Warren
Mich.Michigan (n=501)
Even
Sanders +2
Trump +6
Penn.Pennsylvania (661)
Biden +3
Sanders +1
Even
Wisc.Wisconsin (651)
Biden +3
Sanders +2
Even
Fla.Florida (650)
Biden +2
Trump +1
Trump +4
Ariz.Arizona (652)
Biden +5
Trump +1
Warren +2
N.C.North Carolina (651)
Trump +2
Trump +3
Trump +3
Based on a New York Times/Siena College poll of 3,766 registered voters from Oct. 13 to Oct. 26.

Across the six closest states that went Republican in 2016, he trails Joe Biden by an average of two points among registered voters but stays within the margin of error.

The Daily Poster

Listen to ‘The Daily’: Who’s Actually Electable in 2020?

A new poll from The New York Times reveals which Democratic candidates have a competitive edge against President Trump in battleground states.
bars
0:00/24:36
-24:36

transcript

Listen to ‘The Daily’: Who’s Actually Electable in 2020?

Hosted by Michael Barbaro, produced by Jonathan Wolfe, Rachel Quester, and Jessica Cheung; with help from Annie Brown and Kelly Prime; edited by M.J. Davis Lin and Lisa Tobin

A new poll from The New York Times reveals which Democratic candidates have a competitive edge against President Trump in battleground states.

taylor buckingham

Hi, may I speak with Jack? Hi, Jack. My name’s Taylor. I’m just calling from the Siena College Research Institute. We’re just speaking briefly today to voters in your area. No one will try to sell you anything. Have I reached you on a landline or a cell phone? Do you have time to take a survey? It just takes about 10 minutes.

O.K., are you sure? It’s very important.

O.K., thank you. You too. Bye.

michael barbaro

From The New York Times, I’m Michael Barbaro. This is “The Daily.” Today —

speaker

Now, if the 2020 U.S. presidential election was held today, who would you vote for if the candidates were Joe Biden or Donald Trump?

michael barbaro

The Times conducts a major new poll that tackles the biggest questions of the 2020 presidential race.

taylor buckingham

And now for the next set of questions, tell me which of the two Democrats you’d be likelier to support for —

michael barbaro

How likely is Donald Trump to be re-elected? And which Democrat is best positioned to defeat him? It’s Tuesday, November 5.

Nate Cohn, this morning The Times released this poll that you’ve been working on that’s, as best I can tell, the biggest poll The Times has done in the 2020 campaign. What is the story of this particular poll?

nate cohn

I think we decided to pursue this project for two main reasons. The first is because of electability. Polls show that Democrats are obsessed with how to beat the president. They say that they would rather vote for a candidate who they thought had the best chance to beat President Trump, even more so than someone who they agree with most on the issues. And we thought that was an issue that we didn’t know very much about, that we didn’t have very much data that could address that question of what kinds of candidates, what kinds of issues, what kinds of messages are likeliest to have an advantage in the general election. The second reason is because in the last election, the polls basically got it wrong. The national polls were O.K. They showed Hillary Clinton ahead, and indeed she won in the national vote. But the election was decided in a series of battleground states, where the polls showed Hillary Clinton with a wide advantage.

archived recording

Now, to decision 2016 coverage tonight in the push for Pennsylvania, which will be a key battleground state —

nate cohn

They showed almost across the board that Hillary Clinton would defeat Donald Trump.

archived recording 1

— Hillary Clinton is holding a 9-point lead over —

archived recording 2

— trailing Hillary Clinton by about 7 points.

archived recording 3

Clinton ahead 46 percent to 36 percent for Trump.

archived recording 4

When you’re losing that bad in all of the swing states after the conventions, not to put too fine a point on it, but if you’re a presidential candidate, you have to change that. That’s an emergency.

nate cohn

And those polls were wrong for a number of reasons. I think the biggest reason is because they underrepresented the number of white voters without a college degree. That’s the president’s base. But the polls that were conducted in these critical states tended to drastically underestimate the number of white, working-class voters who are part of the electorate.

michael barbaro

Hmm. Why would that be?

nate cohn

In general, well-educated voters are likeliest to respond to political surveys. That’s been true for a long time. But many state pollsters, unlike the national polls, I should note, did not adjust for that, which had significant consequences for the overall result. And so in order to accomplish those two goals, we knew that we had to conduct surveys in the six closest states that were carried by the president — Pennsylvania, Wisconsin, Michigan, Florida, Arizona and North Carolina. And we got an Iowa bonus on the side. And we asked voters of all parties whether they would vote for Donald Trump or one of the three leading Democratic candidates in head-to-head matchups — Joe Biden, Bernie Sanders and Elizabeth Warren. And we had to make sure that we did an appropriate job of representing the groups that decided the last election. And we also asked them about their views on a range of issues.

michael barbaro

Right. Our colleague, “Daily” producer Jessica Cheung, actually went up to Siena to watch polling in action for this actual project.

jessica cheung

Hey, I’m Jess. I’m a producer with “The Daily.”

taylor buckingham

Hi.

jessica cheung

Hi. I was wondering if I could ask you a few questions.

taylor buckingham

Yeah, that’s fine.

nate cohn

Right, so that call center is in the basement of a dormitory.

taylor buckingham

Well, my name’s Taylor Buckingham, and I work for the Siena College Research Institute. And I’m currently a second-semester senior.

connor mccann

I’m Connor McCann, and I’m doing Siena College Research Institute polling.

jim companion

Yeah, my name is Jim Companion. I’m a retired New York State Department of Taxation and Finance worker. And I’m doing this job at Siena as a part-time gig.

nate cohn

The interviewers who are placing these telephone calls are a mix of students and retirees and people who make these calls as their second job.

connor mccann

Oh, O.K. Is this not Jeanette? All right, well, then, I appreciate your time. All right, thank you.

[phone receiver clicks]

[music]

nate cohn

They might spend all day trying to reach only a handful of respondents who actually pick up the phone and complete our survey.

jessica cheung

And what number call was that?

taylor buckingham

That was number 32.

jessica cheung

And how many did you complete?

taylor buckingham

Two so far. I’ve called people, about 200 calls in two hours, and I will get no surveys.

jim companion

23, and I’ve got one.

connor mccann

That was 88. I finally got one.

Well, not bad.

taylor buckingham

I think I had 130 attempts in two hours, and that was my sixth successful call.

jessica cheung

Oh, great.

taylor buckingham

Yes. Yeah.

michael barbaro

Right. Listening to the tape that Jess got from that call center, it is striking just how hard it is to poll people. I mean, logistically, just a huge challenge to get people to answer the phone and then actually answer the questions.

nate cohn

Right. I would guess we made several hundred thousand telephone calls in order to get the 5,000 respondents that we ultimately received.

michael barbaro

Yeah, it feels like just about the most inefficient operation in the universe.

nate cohn

It has to be. And it’s gotten much more difficult over the last few years.

jessica cheung

Do you think tone matters when you pick up the phone and call people?

connor mccann

Yeah, for sure. They tell you that you don’t want to sound like a robot. And sometimes it’s hard, too, because you’re reading off a screen.

nate cohn

A stream of robocallers deters everybody from picking up unknown numbers.

connor mccann

You know, hello, this is Connor. I’m not selling you anything. I’m just doing a poll today. Otherwise, if you deal with a robot voice, I usually just hang up, or you get a pretty vulgar hang-up or something like that.

nehiti tull (ph)

So a bad tone is like, hi, my name is Nehiti Tull (ph). That’s a bad tone. And a good tone is like, oh, hi, my name is Nehiti Tull (ph). How are you? Stuff like that.

taylor buckingham

And if I answer and I’m just like, hi, this is Taylor. I’m calling from the Siena College Research, as opposed to, hi! My name’s Taylor — it’s a different response. Because if people —

nate cohn

Fewer people have landlines than used to be true.

taylor buckingham

People answer, and they just curse me out — just a lot of yelling, screaming.

jim companion

They’ll tell you to you-know-what, F off, stuff like that. Don’t call me again. This is being recorded. You’re in big trouble if you call me again.

nate cohn

And yet, it’s the best option we have for reaching a representative sample of voters, even still.

michael barbaro

And how did those 5,000 or so respondents, out of hundreds of thousands of calls you make — how did that ensure that you were doing as accurate a poll as possible, especially given the history that you just described of problems in 2016?

nate cohn

Right, so there’s really two halves to what we do. One is that we do everything we can to make sure the people who we get are representative. And we go to great lengths to do that. For instance, we make sure that we call the right number of registered Democrats and Republicans in each state, the right number of young voters and older voters, and high-turnout voters and low-turnout voters. And then once we have our sample, we then make statistical adjustments on the back end. Despite the effort to make sure we talk to rural voters, to talk to low-turnout voters, we still have too many people who have a college degree, and we have to give more weight to the people who don’t have one.

michael barbaro

So you encountered the same problems as the state polls in 2016, but it sounds like you think you’re coming up with some sort of a fix, some sort of an adjustment that accounts for that same problem.

nate cohn

Right, so almost all of the top pollsters out there now do this technique called weighting, where they give more weight to respondents from the underrepresented groups. And in our polls, that mainly entails giving more weight to less educated voters of any race.

michael barbaro

And so how long does this whole thing take?

nate cohn

In this particular case, we started polling on October 13 and only wrapped up on October 30, I believe.

michael barbaro

So about three weeks.

nate cohn

Right.

michael barbaro

And then you get the data, which I assume is like Christmas for Nate Cohn.

nate cohn

Very exciting. And I should note, I get an update on it every day. And if I don’t get it in the morning, then I’m waiting for it, and I can’t do anything until I see what it is.

michael barbaro

Panic.

nate cohn

Well, I’m not panicked. I just want to know, right? And so how am I supposed to do anything when I think that I’m about to see something that’s more interesting?

michael barbaro

[LAUGHS] So when this data finally arrives, what does it tell you?

nate cohn

We found that the president remains strikingly competitive in these states. I, at least, was surprised by it. He seems to be assembling a coalition that’s not very different from the one that he had in 2016. He still does really well among white voters without a degree.

michael barbaro

And why did that surprise you?

nate cohn

Well, I mean, the national polls are — some of them are pretty cataclysmically bad for him, I would say. There are national polls out there that say that a majority of Americans think he should be impeached and removed from office right now. So for all of that to be true and him to sort of be doggedly competitive in these states, while it’s not stunning, I think, given the differences between these states and the country as a whole, it is still somewhat surprising, I think.

speaker 1

Thinking about the upcoming presidential election in 2020, do you think you’re definitely going to vote for President Trump —

speaker 2

I sure will.

speaker 1

— probably going to vote for Trump, probably going to vote for the Democratic nominee, or definitely going to vote for the Democratic nominee?

speaker 2

No, definitely going to vote for Mr. Trump.

speaker 1

O.K. Do you approve or disapprove of the job that Donald Trump is doing as president?

speaker 2

I definitely approve.

speaker 1

Do you somewhat approve, or do you strongly approve?

speaker 2

I strongly approve.

speaker 1

O.K.

michael barbaro

When you say he’s doing well in these states, can you give us a little specificity?

nate cohn

Yeah, he’s in a close race in every one of these states against all of his potential opponents. Nothing is out of reach for him at this point.

michael barbaro

He could win all six.

nate cohn

He could win all six of them. The reverse is true, too. The Democrats could win all six of them. There’s a year to go. But I think that if you were looking at these national polls, you might have expected him to be at a substantial disadvantage.

michael barbaro

O.K., so that’s how the president is positioned in these six states. What do these polls tell us about how his Democratic competitors stack up against him in those same states?

nate cohn

So we found that Joe Biden would defeat the president if the election were held today in most of these states. He didn’t do exceptionally well, but he ran an average of 4 points better than Hillary Clinton did in 2016. And that would be enough for him to win the election if these results were right. I don’t think that’s great news for Biden.

michael barbaro

Why is that not great news for Biden?

nate cohn

Well, his whole pitch is about electability. And he’s ahead, but it’s not by the sort of decisive margin that would make him seem like a sure bet to win, even if it indicates he has an advantage if the election were held today. Bernie Sanders and the president were locked in a very tight race. But Bernie Sanders did lead among registered voters in the three states that really cost Hillary last time, Michigan, Wisconsin and Pennsylvania. An interesting wrinkle is that a lot of Bernie Sanders’s support comes from people who don’t have a track record of voting or who say they aren’t sure whether they’ll vote. And then the story is a little bit more complicated still for Elizabeth Warren. Her results were worse than I thought they would be. The president led her in five of the six states. He led her in North Carolina and Florida by comfortable margins. He led in Michigan by a comfortable margin, even though Bernie Sanders was ahead there. She only led the president among registered voters in Arizona. And even that dissipated when you looked at the likeliest voters. So overall, she trailed by 2 points across these states among registered voters. That’s the same as Hillary Clinton’s performance. So if the election were held today, and if these results are right, Elizabeth Warren would lose to the president.

michael barbaro

In every —

nate cohn

In all six of these states, I think.

michael barbaro

— one of these states.

connor mccann

Yeah, so I got a hold of a gentleman on his phone. He was a Democrat, but he was kind of on the fence with certain Democratic nominees. He said that he didn’t like Elizabeth Warren, and all the other Democratic nominees he also didn’t agree with, but he doesn’t like Trump. So he’s kind of at a middle ground as to what to do for the 2020 election.

michael barbaro

We’ll be right back.

[music]

michael barbaro

Nate, it’s been my sense that Warren is basically neck and neck, and in some cases, running ahead of Biden in these national Democratic primary polls. So if that’s true, how can it be that she’s lagging so far behind Biden in her ability to beat Trump?

nate cohn

Yeah, I was surprised by how poorly she fared in our polls. But 6 percent of voters told us that they would support Joe Biden against the president but would not support Elizabeth Warren in a head-to-head match-up against Donald Trump. And that 6 percent is going to be hard for her. We asked every one of these voters whether they agreed with the statement that Elizabeth Warren was too far to the left for them to feel comfortable supporting her, and a majority of them said they agreed with that statement. We also asked all of these voters whether they agreed with the statement that most of the women who run for president just aren’t that likable. And 40 percent of them said they agreed with that statement.

michael barbaro

That’s a very specific way of wording that question.

nate cohn

Yeah. We were trying to give people permission to say that, in their mind, there’s just something wrong with the women who have run for office, without making them say they don’t want a woman to be president.

michael barbaro

You’re essentially giving people permission, through indirect language, to be sexist.

nate cohn

Sexist, I think — right.

michael barbaro

So you’re saying that of those who can support Joe Biden over President Trump but cannot support Warren over Trump, their big issues are her policies and her gender. Do you know which of these two are the most important driver?

speaker

For each set of questions, I’m going to describe two hypothetical Democrats. Tell me which of the two Democrats you would be likelier to support for the Iowa —

nate cohn

Well, we asked a number of questions that might help lend some insight on that.

speaker

A Democrat who promises to replace the current health care system with “Medicare for all,” or a Democrat who promises to improve the existing health insurance system?

taylor buckingham

A Democrat who is more moderate than most Democrats, or a Democrat who is more liberal than most Democrats?

speaker

A Democrat who promises to bring politics in Washington back to normal, or a Democrat who promises to fundamentally change America?

nate cohn

We asked people whether they’d like a more moderate Democrat or a more liberal Democrat. And they do say they’d prefer a more moderate one, but not 100 to nothing. It’s like 75 to 25. And then we’d ask these questions about Warren personally.

speaker

For each of the following candidates, tell me whether you’re very confident that they would beat Trump, somewhat confident that they would beat Donald Trump, not very confident, or not at all confident that they would beat Donald Trump. Elizabeth Warren — very confident, somewhat confident, not very confident, not at all confident.

nate cohn

They have an unfavorable view of Warren by about a 2 to 1 margin. That matches up roughly well with the question that we asked them about whether they like female candidates, with about 40 percent of them saying that they agree that they’re unlikable. So when I sort of put it all together, I think that they’re roughly equal in their contribution. One interesting lens for considering this is Bernie Sanders, because he fares better than Warren and —

michael barbaro

Versus Trump.

nate cohn

— versus the president, and worse than Biden does against the president.

michael barbaro

So he’s kind of in the middle of Warren and Biden.

nate cohn

So he’s basically halfway in between the two. And part of the reason why he would underperform Biden is because he would lose some of these voters who are against a more left-leaning, liberal Democrat candidate. But perhaps he could fare better than Warren, despite having a fairly similar set of views on the issues, because there’s just something about Bernie that makes him more attractive to voters than Elizabeth Warren. And I think you could argue that gender is one of the principal factors that we’d have to consider there.

michael barbaro

Nate, who is a voter who would support Joe Biden against Trump but would not support Elizabeth Warren against Trump? Is there a demographic that comes to mind from this poll?

nate cohn

I’ll offer two archetypes. One is an affluent suburban male who’s somewhat conservative. He really doesn’t like President Trump, but he also really doesn’t like Medicare for all or the progressive left wing of the Democratic Party. Gender could be a factor as well. I wouldn’t rule it out, but I think that just on the policy merits, it’s a consistent story that this is someone who doesn’t like Trump but would not accept a particularly left-wing nominee.

michael barbaro

But would support Biden.

nate cohn

But would support Biden. And then I think the second archetype I might offer is a young voter who has some conservative views on cultural issues. Maybe they don’t like P.C. culture, for instance. And for some reason, Elizabeth Warren just doesn’t resonate with them very much. We spoke to one woman from Florida. She said, quote, “There’s just something about her that I just don’t like. I just don’t feel like she’s a genuine candidate. I find her body language to be off-putting. She’s very cold. She’s basically a Hillary Clinton clone.” And when asked about the women running for president more generally, she said, “They’re super unlikable.” So it actually turns out that among persuadable voters, women are a little likelier than men to say they agree that most of the women running for president are unlikable.

michael barbaro

That’s fascinating.

nate cohn

Yeah.

michael barbaro

Nate, this poll would seem to deliver some pretty sobering news to Democrats. The president is performing at a very competitive level in these six crucial swing states. And from everything this poll just told us, the general election voter is nowhere near as far to the left as the Democratic primary voters, who love, for example, Elizabeth Warren.

nate cohn

Yeah, I think it’s a tough poll for Democrats to take right now. They might look at our data and consider other options. They might try and pursue a blue Texas or Georgia that maybe could make up for their weakness in these whiter working-class states. They might consider whether they ought to look at some of these other candidates again, maybe an Amy Klobuchar or a Cory Booker, who knows? Someone who’s closer to a compromise between the two wings of the party. They might conclude that they have to try and change the composition of the electorate, that they may be trailing among registered voters now, but after a year of registering new young and nonwhite voters, perhaps they could take the lead. But with the electorate we have today, and with the question of electability in mind, it seems like voters today are likeliest to supported a moderate Democratic nominee. But there’s a lot more to a candidate than whether you’re a moderate or a liberal. Moderates aren’t assured to win this election. People on the left aren’t assured to lose it, either. Right now, with these particular set of candidates, Elizabeth Warren fares the worst against President Trump. Joe Biden fares the best. And on balance, the voters who seem to be switching from candidate to candidate are concerned that the major Democratic nominees too far to the left. There’s no guarantee that that’s how it will stick, but that’s where things are now.

michael barbaro

Nate, thank you very much.

nate cohn

Thanks for having me.

speaker 1

I think Iowa is much easier to call than Michigan. They’re just much easier to stay on the phone. And really, once you say 10 minutes, they’re willing to do the 10 minutes once it’s about politics. But Michigan was a little bit harder.

speaker 2

Iowa has been a pleasure so far. People are very nice, but I can’t say the same for Michigan. It was very difficult to do Michigan. They were not willing to do the survey, and mostly they hung up on you.

taylor buckingham

When I called Michigan, nobody wanted to answer the phone at all.

speaker 3

I would definitely say Iowa has been the easiest state so far.

jessica cheung

If you were to rank the six states in terms of friendliness, how would you rank them?

speaker 4

Definitely Iowa. Michigan at the bottom. North Carolina —

michael barbaro

We’ll be right back.

Here’s what else you need to know today. On Monday, the Trump administration formally notified the United Nations that it would withdraw from the Paris climate accord, fulfilling a promise made by the president back in 2017. Under the president’s plan, the U.S. will leave the agreement, the most ambitious plan to curb carbon emissions in history, on November 4, 2020, the day after the next presidential election. The U.S. is the first and only country to try to leave the accord amid a growing body of evidence that climate change will lead to food shortages, worsening wildfires and devastating floods. And —

archived recording (jon echols)

These are real lives, real people, with real families, and with real friends. And they get to go home.

michael barbaro

Oklahoma has carried out the largest commutation of prisoners in U.S. history, allowing nearly 500 inmates serving time for low-level nonviolent crimes to walk free and return to their families.

speaker

I’ve been out of her life for three years, but she’s my hope. She’s never given up on me. And all I have to say is thank you.

michael barbaro

The commutation followed a 2016 referendum in which Oklahoma residents voted to shrink the state’s prison population and downgrade many crimes, like drug possession, from felonies to misdemeanors. Many of those released on Monday had been convicted of crimes that would no longer require prison time under the state’s new rules.

speaker 1

Mama!

speaker 2

Which one? I’m sorry.

speaker 1

That one.

speaker 2

This one right here?

speaker 3

Hi, my baby.

michael barbaro

That’s it for “The Daily.” I’m Michael Barbaro. See you tomorrow.

Mr. Trump leads Elizabeth Warren by two points among registered voters, the same margin as his win over Hillary Clinton in these states three years ago.

The poll showed Bernie Sanders deadlocked with the president among registered voters, but trailing among likely voters.

The results suggest that Ms. Warren, who has emerged as a front-runner for the Democratic nomination, might face a number of obstacles in her pursuit of the presidency. The poll supports concerns among some Democrats that her ideology and gender — including the fraught question of “likability” — could hobble her candidacy among a crucial sliver of the electorate. And not only does she underperform her rivals, but the poll also suggests that the race could be close enough for the difference to be decisive.

In national polls, Mr. Trump’s political standing has appeared to be in grave jeopardy. His approval ratings have long been in the low 40s, and he trails Mr. Biden by almost nine points in a national polling average. But as the 2016 race showed, the story in the battleground states can be quite different. Mr. Trump won the election by sweeping Michigan, Pennsylvania, Wisconsin, Florida, Arizona and North Carolina — even while losing the national vote by two points.

Democrats would probably need to win three of the six states to win the White House, assuming other states voted as they did in 2016 — an outcome that is not at all assured.

How Trump fares among likely voters

Trump vs. Biden Sanders Warren
Mich.Michigan (n=501)
Biden +1
Sanders +3
Trump +4
Penn.Pennsylvania (661)
Biden +1
Trump +1
Trump +2
Wisc.Wisconsin (651)
Biden +2
Even
Trump +2
Fla.Florida (650)
Biden +2
Trump +2
Trump +4
Ariz.Arizona (652)
Biden +2
Trump +4
Even
N.C.North Carolina (651)
Trump +2
Trump +4
Trump +4
Based on a New York Times/Siena College poll of 3,766 registered voters from Oct. 13 to Oct. 26.

The Times/Siena results and other data suggest that the president’s advantage in the Electoral College relative to the nation as a whole remains intact or has even grown since 2016, raising the possibility that the Republicans could — for the third time in the past six elections — win the presidency while losing the popular vote.

There is a full year before Election Day, and a lot can change. Ms. Warren is an energetic campaigner. She could moderate her image or motivate young and nonwhite voters, including the millions who might not yet even be included in a poll of today’s registered voters. Mr. Biden could lose the relatively conservative voters who currently back him; the president could be dealt irreparable political damage during the impeachment process.

But on average over the last three cycles, head-to-head polls a year ahead of the election — matching the eventual nominees of each party — have been as close to the final result as those taken the day before. The stability of the president’s approval rating is a reason to think this pattern might hold again for a fourth cycle, at least for the three leading and already well-known Democrats tested in these polls.

While Mr. Biden ranks as the strongest Democrat in the swing states polled, the findings are not necessarily great news for him, either. His appeal to Democrats hinges on the view that he’s a safe bet against the president, yet his lead against Mr. Trump is not nearly so comfortable that he could be considered a sure thing.

The Times/Siena polls depict an exceptionally energized and polarized electorate that remains divided along the lines of the 2016 presidential election. More than 90 percent of registered voters say they’re “almost certain” or “very likely to vote,” exceeding the 87 percent who said the same thing in Times/Siena polls conducted in the final weeks of the 2016 election.

Three years later, more than 90 percent of Mr. Trump’s supporters from 2016 approve of his performance, while more than 90 percent of Mrs. Clinton’s voters disapprove.

The major demographic cleavages of the 2016 election also remain intact. Mr. Trump struggles badly among college-educated white voters and nonwhite voters, though there are signs his standing among the latter group has improved modestly since the last presidential election. He counters with a wide lead among white voters who did not graduate from a four-year college.

In contrast to recent national surveys, the Times/Siena polls find that the president’s lead among white, working-class voters nearly matches his decisive advantage from 2016. This group represents nearly half of registered voters in these states, and a majority in the Northern battlegrounds that decided the last election.

What Different Types of Voters Said

By race

2016 polls Trump v.
Biden
Trump v.
Sanders
Trump v.
Warren
White, college (n=1,321) Clinton +6 Biden +10
Sanders +6
Warren +7
White, no coll. (1,469) Trump +26 Trump +24
Trump +24
Trump +26
Black (328) Clinton +79 Biden +74
Sanders +69
Warren +68
Hispanic (370) Clinton +35 Biden +34
Sanders +31
Warren +24
Other (186) Clinton +11 Biden +10
Sanders +14
Warren +3

The 2016 data is an Upshot compilation of 7,802 battleground state respondents from pre-election polls of registered voters from the New York Times/Siena, CBS/New York Times, Pew Research, ABC/Washington Post and CNN/ORC.

The poll offers little evidence that any Democrat, including Mr. Biden, has made substantial progress toward winning back the white working-class voters who defected to the president in 2016, at least so far. All the leading Democratic candidates trail in the precincts or counties that voted for Barack Obama and then flipped to Mr. Trump.

As a result, Democrats appear to have made little progress in reclaiming their traditional advantage in the Northern battleground states, despite their sweep there in the 2018 midterms. Respondents in these states said they voted for Democratic congressional candidates by an average of six points, all but identical to their actual winning margins.

Nearly two-thirds of the Trump voters who said they voted for Democratic congressional candidates in 2018 say that they’ll back the president against all three named opponents.

Polling the Battlegrounds

We talked to 3,766 voters in 6 of the most competitive states.

Nonetheless, Mr. Biden holds the edge among both registered voters and likely voters, and even among those who cast a ballot in 2016. He has a lead of 55 percent to 22 percent among voters who say they supported minor-party candidates like Gary Johnson and Jill Stein, and among those who say they voted but left the 2016 presidential race blank. It comes on top of a slight shift — just two points in Mr. Biden’s favor — among those who say they voted for either Mrs. Clinton or Mr. Trump.

Ms. Warren and Mr. Sanders, on the other hand, lose a sliver of Mrs. Clinton’s vote and make fewer inroads among Mr. Trump’s supporters.

The wide spread between the three candidates might be a surprise. But even in today’s polarized era, 21 percent of registered voters don’t simply line up for Mr. Trump or his Democratic opponents in the three head-to-head matchups. This includes 6 percent of the electorate that currently supports Mr. Biden against Mr. Trump but not Ms. Warren against Mr. Trump. There is also 3 percent of the electorate currently willing to support Ms. Warren but not Mr. Biden. And then there are the voters who are undecided in either matchup.

How Voters Would Choose in Head-to-Head Matchups

Most of the electorate Is for any Democrat or for Mr. Trump, but about 1 in 6 voters say that they don’t know or that their vote depends on whether Mr. Biden or Ms. Warren is the nominee.

Whom voters would prefer if the election were…
Trump or
Biden
Trump or
Warren
Share of electorate
Biden Warren
40%
Biden Trump
2%
Biden Other
4%
Other Warren
2%
Other Other
6%
Other Trump
<1%
Trump Other
1%
Trump Warren
1%
Trump Trump
43%
“Other” includes voting for a 3rd-party candidate, not voting at all, or instances in which the respondent said he or she didn’t know or refused to answer.

In states likely to be closely fought, even a modest swing among these voters could resolve the election in either side’s favor. In this poll, they swing the election in favor of Mr. Biden but leave Ms. Warren or Mr. Sanders short.

Ms. Warren’s challenge is not just name recognition. She also underperformed her rivals against Mr. Trump in separate head-to-head polling of Iowa, where she and other candidates have been campaigning in earnest ahead of the Democratic caucus.

Ms. Warren trailed Mr. Trump by six points in Iowa, the widest gap among leading Democrats, even though she led the Democratic caucus in our poll. Pete Buttigieg, who is generally not as well known as Ms. Warren, trailed Mr. Trump by four points in Iowa, which was the only state where we included him in head-to-head polling against the president. Mr. Biden trailed by one point and Mr. Sanders by three.

An analysis of the 205 respondents from the six core battleground states who support Mr. Biden but not Ms. Warren suggests that she might struggle to win many of them over.

Over all, 26 percent of these voters say they have a favorable view of Ms. Warren, compared with 47 percent who have an unfavorable view.

They say, by a margin of 74 percent to 24 percent, that they would prefer a more moderate Democrat nominee to a more liberal one. By a nearly identical margin, they would prefer a Democrat who promises to find common ground with Republicans over one who promises to fight for a bold progressive agenda.

Of voters who support Mr. Biden but not Ms. Warren, 52 percent agree with the statement that Ms. Warren is too far to the left for them to feel comfortable supporting her for president, while 26 percent disagree.

A Sample of Voters Who Prefer Biden, but Not Warren

About 6 percent of voters surveyed said they would vote for Joe Biden but not Elizabeth Warren. Below, a sample of these voters, and more details about their views.

Areas where voters who support Biden, but not Warren, disagree with Warren
In Warren v. Trump,
would vote for…
Unfav. view of Warren Women cands. not ‘likeable’ Warren too far left Against single-payer health Prefers moderate Dem. Supports Rep. tax law
Alex
Mich.
Trump
Alex
Wis.
Trump
Alexa
N.C.
Trump
Alexandra
Fla.
Trump
Alexis
Ariz.
Another cand.
Amber
Mich.
Doesn’t know
Angela
N.C.
Doesn’t know
Ann
Mich.
Doesn’t know
Ann
N.C.
Trump
Anne
Fla.
Doesn’t know
Anthony
Mich.
Doesn’t know
Anthony
Ariz.
Another cand.
Araceli
Ariz.
Trump
Arlene
Mich.
Doesn’t know
Austin
N.C.
Trump
Barbara
Mich.
Doesn’t know
Barbara
Ariz.
Wouldn’t vote
Bessie
Mich.
Doesn’t know
Bethany
Pa.
Doesn’t know
Betsy
Pa.
Doesn’t know
Betty
Wis.
Wouldn’t vote
Billy
Mich.
Trump
Bradley
Wis.
Trump
Brendan
Pa.
Trump
Carl
Fla.
Trump
Carmen
Fla.
Doesn’t know
Cathy
Ariz.
Trump
Charles
Fla.
Trump
Charles
N.C.
Another cand.
Christine
N.C.
Doesn’t know
Christopher
Fla.
Another cand.
Christopher
N.C.
Trump
Colin
Mich.
Doesn’t know
Curtis
Fla.
Trump
Cynthia
Ariz.
Another cand.
Damian
Pa.
Trump
Daniel
Mich.
Wouldn’t vote
Daniel
Pa.
Trump
Daniel
Pa.
Wouldn’t vote
Daniel
Ariz.
Another cand.
Danielle
Pa.
Another cand.
David
Wis.
Trump
David
Wis.
Trump
David
Wis.
Trump
Delbert
Fla.
Trump
Dennis
Mich.
Trump
Don
N.C.
Another cand.
Donald
Pa.
Another cand.
Doris
Fla.
Another cand.
Dorothy
Wis.
Trump
Douglas
Ariz.
Trump
Edward
Wis.
Trump
Elizabeth
Ariz.
Doesn’t know
Elmer
Ariz.
Trump
Eric
Mich.
Trump
Evan
Pa.
Trump
Faith
Fla.
Trump
Francisco
Ariz.
Wouldn’t vote
Frank
Pa.
Doesn’t know
Frankie
N.C.
Another cand.
Frederick
Wis.
Doesn’t know
Gabriel
Wis.
Trump
Garrett
Fla.
Trump
Gayle
Fla.
Doesn’t know
Geneva
N.C.
Another cand.
Glen
Mich.
Trump
Glenda
Pa.
Another cand.
Heather
Wis.
Trump
Hector
Ariz.
Doesn’t know
Jack
Mich.
Wouldn’t vote
Jacob
N.C.
Trump
Jakob
Wis.
Trump
James
Wis.
Trump
James
Ariz.
Trump
James
Ariz.
Another cand.
James
Fla.
Doesn’t know
Jason
Ariz.
Trump
Javier
Ariz.
Wouldn’t vote
Jennifer
Mich.
Wouldn’t vote
Jennifer
Wis.
Doesn’t know
Jennifer
Fla.
Doesn’t know
Jerry
Fla.
Trump
Jessica
Mich.
Doesn’t know
Jessica
Wis.
Trump
John
Mich.
Another cand.
John
Fla.
Trump
John
N.C.
Another cand.
Joseph
Wis.
Another cand.
Joseph
Wis.
Wouldn’t vote
Joseph
Pa.
Doesn’t know
Joshua
Wis.
Doesn’t know
Joshua
Wis.
Trump
Joshua
Pa.
Trump
Juan
Wis.
Trump
Juan
Fla.
Doesn’t know
Judith
N.C.
Trump
Justin
Pa.
Trump
Justin
Ariz.
Trump
Kathleen
Fla.
Wouldn’t vote
Keith
N.C.
Another cand.
Kenneth
Pa.
Wouldn’t vote
Krista
Wis.
Trump
Lillie
Fla.
Wouldn’t vote
Linda
N.C.
Wouldn’t vote
Loretta
Fla.
Trump
Lorraine
Mich.
Doesn’t know
Lucia
Fla.
Wouldn’t vote
Lucy
N.C.
Doesn’t know
Luke
Wis.
Doesn’t know
Marcus
Mich.
Trump
Margaret
Pa.
Doesn’t know
Margie
Ariz.
Another cand.
Mark
Wis.
Doesn’t know
Mark
Fla.
Trump
Martha
Fla.
Wouldn’t vote
Martina
Fla.
Doesn’t know
Mary
Ariz.
Wouldn’t vote
Maryjo
Pa.
Trump
Matthew
Pa.
Doesn’t know
Michelle
Wis.
Another cand.
Miriam
Fla.
Doesn’t know
Nancy
Mich.
Trump
Nancy
Wis.
Another cand.
Patricia
Fla.
Wouldn’t vote
Patrick
N.C.
Trump
Paula
Ariz.
Wouldn’t vote
Pedro
Ariz.
Doesn’t know
Peter
Fla.
Doesn’t know
Rachel
Pa.
Doesn’t know
Raymond
N.C.
Doesn’t know
Robert
Mich.
Trump
Robert
Pa.
Another cand.
Roberto
N.C.
Trump
Roger
N.C.
Trump
Ronald
Wis.
Another cand.
Roxann
Pa.
Trump
Ruben
Ariz.
Trump
Sabrina
Fla.
Wouldn’t vote
Sandra
N.C.
Wouldn’t vote
Scott
Mich.
Wouldn’t vote
Sean
Pa.
Trump
Sharron
Ariz.
Doesn’t know
Sheila
Wis.
Doesn’t know
Sheree
Fla.
Another cand.
Sherman
Fla.
Doesn’t know
Shirley
Mich.
Trump
Stephen
Wis.
Trump
Steven
Fla.
Trump
Stuart
Wis.
Doesn’t know
Susan
Ariz.
Trump
Susan
Fla.
Trump
Susan
N.C.
Wouldn’t vote
Tara
N.C.
Doesn’t know
Teresa
Fla.
Doesn’t know
Theodore
Pa.
Another cand.
Toni
Mich.
Trump
Troy
Ariz.
Trump
Victor
Ariz.
Doesn’t know
Virginia
Fla.
Trump
Vivian
Pa.
Wouldn’t vote
Walter
N.C.
Doesn’t know
Walter
N.C.
Trump
Zachary
Mich.
Wouldn’t vote

Note: To preserve privacy, voters with less common names are not shown.

The Biden voters who say Ms. Warren is too far to the left are relatively well educated and disproportionately reside in precincts that flipped from Mitt Romney in 2012 to Mrs. Clinton four years later. They oppose single-payer health care or free college, and they support the Republicans’ 2017 tax law. They are not natural Democratic voters: 41 percent consider themselves conservative; 20 percent say they’re Republican; 33 percent supported Mr. Trump or Mr. Johnson in 2016.

Dawn Marshall, an independent from Tampa, Fla., said that with the exception of Mr. Biden, the Democrats running for president are too left-leaning for her.

“They want to be socialists, and this is not a socialistic country,” she said. “This is a working country where people go out, do the best that they can do, find jobs. I am so sick and tired of having to support other folks. We can’t be equal.”

Ms. Marshall, a telecom engineer who is black and Native American, would not seem to be representative of her demographic group. Yet nonwhite Biden supporters are likelier than white Biden voters to say they would choose Mr. Trump over Ms. Warren.

At the same time, 41 percent of the voters who support Mr. Biden but not Ms. Warren say they agree with the statement that most of the women who run for president “just aren’t that likable,” likely bolstering concerns among some Democrats that sexism could be a burden on her candidacy.

These Biden supporters are disproportionately male and working class. This group holds a variety of conservative views on cultural issues: 55 percent agree that discrimination against whites has become as big a problem as discrimination against minorities; 79 percent agree that political correctness has gone too far; 54 percent would reduce legal immigration.

Some women also fall into this group. Elysha Savarese, 26, works in victims advocacy in Florida. She voted for Mr. Trump and said she would not do so again.

But she wouldn’t vote for Ms. Warren, either.

“There’s just something about her that I just don’t like,” she said. “I just don’t feel like she’s a genuine candidate. I find her body language to be very off-putting. She’s very cold. She’s basically a Hillary Clinton clone.”

As for female presidential candidates in general, she said, “They’re super unlikable.”

The poll does show a natural area of potential growth for Ms. Warren: the 32 percent of Biden-but-not-Warren voters who do not agree that most female presidential candidates are unlikable or that Ms. Warren is too far to the left.

These voters like Mr. Biden — he has a 92 percent favorable rating among them — but 52 percent say they don’t know enough about Ms. Warren to have an opinion. Fifty-nine percent are nonwhite. Mr. Sanders has a wide lead against the president among these voters.

If these respondents had backed Ms. Warren in the Times/Siena survey, the poll results would have shown her and Mr. Trump deadlocked in an election highly reminiscent of 2016.


Claire Cain Miller contributed reporting.

The Times/Siena poll of 3,766 registered voters was conducted from Oct. 13 to Oct. 26. The margin of sampling error for an individual state poll is plus or minus 4.4, except for Michigan at plus or minus 5.1 points. Together, the battleground sample has a margin of error of plus or minus 1.8 percentage points.

The Times/Siena poll of 1,435 registered voters in Iowa was conducted from Oct. 25 to Oct. 30 and has a margin of error of plus or minus 3.1 percentage points.

Full cross-tabs and methodology are available here.

In 2018, the Times/Siena polls had an average error of 2.5 points in 10 polls in these states over the final three weeks of the campaign. If they had been joined together as one large poll, as is the case here, the final result would have been within 1 point.

Nate Cohn is a domestic correspondent for The Upshot. He covers elections, polling and demographics. Before joining The Times in 2013, he worked as a staff writer for The New Republic. More about Nate Cohn

A version of this article appears in print on  , Section A, Page 1 of the New York edition with the headline: Trump Is Remaining Competitive if Not Popular. Order Reprints | Today’s Paper | Subscribe

Advertisement

SKIP ADVERTISEMENT