Biden’s Massive Lead
President Trump is behind 14 points in the latest NYT poll.
The headline for the NYT‘s report on its latest presidential poll is powerful: “Biden Takes Dominant Lead as Voters Reject Trump on Virus and Race.”
Mr. Biden is currently ahead of Mr. Trump by 14 percentage points, garnering 50 percent of the vote compared with 36 percent for Mr. Trump. That is among the most dismal showings of Mr. Trump’s presidency, and a sign that he is the clear underdog right now in his fight for a second term.
Mr. Trump has been an unpopular president for virtually his entire time in office. He has made few efforts since his election in 2016 to broaden his support beyond the right-wing base that vaulted him into office with only 46 percent of the popular vote and a modest victory in the Electoral College.
But among a striking cross-section of voters, the distaste for Mr. Trump has deepened as his administration failed to stop a deadly disease that crippled the economy and then as he responded to a wave of racial-justice protests with angry bluster and militaristic threats. The dominant picture that emerges from the poll is of a country ready to reject a president whom a strong majority of voters regard as failing the greatest tests confronting his administration.
But Hillary Clinton was ahead in the polls and she lost, right? Not like this.
Mr. Biden leads Mr. Trump by enormous margins with black and Hispanic voters, and women and young people appear on track to choose Mr. Biden by an even wider margin than they favored Hillary Clinton over Mr. Trump in 2016. But the former vice president has also drawn even with Mr. Trump among male voters, whites and people in middle age and older — groups that have typically been the backbones of Republican electoral success, including Mr. Trump’s in 2016.
Biden has a huge lead among women and a within-margin-of-error lead among men. He leads among every age group except 50-64, where Trump leads by 1 point (well within the margin of error; it’s a statistical tie). Trump has only a 1 point (again, statistically insignificant) lead among whites. The only demographic that favors Trump significantly are whites with no college, who favor him by 19 points.
Biden and Trump enjoy identical 85 point leads among those who identify with their political party. That means there’s no enthusiasm gap. But here’s the thing: the poll has 466 Democratic respondents and only 341 Republicans. If that’s a representative national sample, that’s a problem for Trump. As is the fact that Independents (401) outnumber Republicans and support Biden by 21 points.
And, as the headline suggests, the issues may actually matter for a change:
According to the poll, white voters under 45 are overwhelmingly supportive of the Black Lives Matter movement, while older whites are more tepid in their views toward racial justice activism. And nearly 70 percent of whites under 45 said they believed the killing of George Floyd was part of a broader pattern of excessive police violence toward African-Americans rather than an isolated incident.
What’s striking, though, is that even among white seniors, one of Mr. Trump’s strongest constituencies, he has damaged himself with his conduct. About two-fifths of whites over 65 said they disapproved of Mr. Trump’s handling of both the coronavirus and race relations.
[…]
Nearly three-fifths of voters disapprove of Mr. Trump’s handling of the coronavirus pandemic, including majorities of white voters and men. Self-described moderate voters disapproved of Mr. Trump on the coronavirus by a margin of more than two to one.
Most of the country is also rejecting Mr. Trump’s call to reopen the economy as quickly as possible, even at the cost of exposing people to greater health risks. By a 21-point margin, voters said the federal government should prioritize containing the coronavirus, even if it hurts the economy, a view that aligns them with Mr. Biden.
Just a third of voters said the government should focus on restarting the economy even if that entails greater public-health risks.
[…]
The public also does not share Mr. Trump’s resistance to mask wearing. The president has declined to don a mask in nearly all public appearances, even as top health officials in his administration have urged Americans to do so as a precaution against spreading the coronavirus. In the poll, 54 percent of people said they always wear a mask when they expect to be in proximity to other people, while another 22 percent said they usually wear a mask. Just 22 percent said they rarely or never wear a mask.
And, no the NYT poll isn’t an outlier. Here’s the RealClearPolitics average (which hasn’t been updated with this poll):
Literally every poll has a considerable lead for Biden and the more recent ones have very wide margins. Indeed, the Fox News poll has the same margin as the NYT poll.
Interestingly, though, the approve/disapprove numbers aren’t quite as dismal:
To be sure, the numbers here are awful. Trump’s underwater by 12.8 points—which tracks his deficit with Biden. But he’s actually been far, far more unpopular at various points in his presidency, with disapproval numbers as much as 21 points higher than his disapproval.
The obvious caveats are that we still have four and a half months to go to the election and we don’t elect Presidents via a popular vote. But there’s no reason to think Trump is going to get better on the virus and race relations—or anything else, frankly—between now and November. And a 12 point national deficit would almost certainly turn into an Electoral College landslide.
This site is obsessed with polls. Why don’t you post about governors’ or senators’ races that look interesting this year?
Indeed, is likely to get much worse on both of those issues in particular.
@Not the IT Dept.:
A quick skimming of the posts suggests that the last post that discussed polling was about a week ago.
But what else would you suggest we use to discuss the state of the race, our guts?
In regards to governor’s (and even Senate) races it is probably too early to have meaningful polling.
Even with an overwhelming vote for Joe Biden I doubt that Trump’s supporters will accept the reality. Nor will Trump. It will all be “Fake Nooz!” and “Voting Fraud!” and all sorts of excuses to insist that the country’s support is really for Trump.
And I bet the GOP will allow him to get away with it. We don’t have anyone with the guts and the integrity to do the equivalent of what happened with Nixon–going to him and telling him that it was time to leave. The GOP is terrified of Trump’s supporters, and a lot of them are going to be insisting that their Mango-skinned deity get shoved back into office, no matter what the results are.
Adding to Tiny’s misery, is Biden has either small leads, within the margin of error in most of the battle ground states and Biden is behind in a few, normally solid Repug states, within the margin of error. The fact that Texas and Georgia are competitive now is amazing.
It appears that Tiny’s base is cracking and it is reflected in the erosion in the number of previously self identified Repugs who now say they are independents. This bodes ill for the party irrespective of Trump, with drop in frequency of ticket splitting by voters, Biden could prove to have wide and long coattails.
@Steven L. Taylor:
How about brains – brains are good. Especially since the polling is all running in pretty much the same direction.
I didn’t mean the polling. Issues? Personalities? Thoughtful analysis? Even identifying all the candidates in a particular state would be a nice change of topic.
@Not the IT Dept.: ” Why don’t you post about governors’ or senators’ races that look interesting this year?”
There’s an open forum right below this. Why don’t you post about governors’ or senators’ races if they’re interesting to you rather than ask someone else to do it?
@Not the IT Dept.: “I didn’t mean the polling. Issues? Personalities? Thoughtful analysis? Even identifying all the candidates in a particular state would be a nice change of topic.”
And while you’re at it, Dr. Taylor, I’m really interested in the decline of the western as the dominant American film genre after the revelations about My Lai — would you please watch all the westerns from that period and do an analysis on how the genre shifted and what the effect on the box office was? Surely that will be better than writing about polls.
Oh, I’m also interested in Icelandic history, particularly from the 13th to the 19th century. Get on it!!!
Another thing everybody learned in 2016 is that state polls matter more. Remember, Hillary won the popular vote by 3M votes and still lost.
@Not the IT Dept.:
If only I had thought to use my brain on this site. I will give it some thought 😉
@wr: I will get right on that! 🙂
@Scott:
It wasn’t new in 2016 that states mattered, the problem was inadequate attention to some specific ones. Regardless, the current gap, and the fact that Biden is polling consistently at the 50% level, is important to note and is different than 2016.
Like the comment I saw a while back, who could possibly want four more years of this shit?
@Not the IT Dept.:
I follow all those polls and the problem is there aren’t many. Some races are still in the primary stage, so we don’t have head-to-heads. Places like Montana get polled once in a blue moon. Even in a big Senate race, like Susan Collins in Maine, the latest poll I could find is from early March.
Also, as I occasionally remind my Twitter followers who whine about my choice of topics: you’re paying what for this?
@Scott:
@Steven L. Taylor:
As Dr T has pointed out the 2016 polls were accurate. Hillary received the percentage of the vote that the final polls anticipated, but the undecided in those last polls, broke for Tiny. The failure in 2016 wasn’t the polls, but the extrapolation as to what the numbers meant. Analyst’s giving Hill a 99% probability of a win committed a miscarriage. Reasonable probability declarations were grabbed on by the statistically challenged as proof of a Hill lock rather than a warning of a possible Tiny win.
A year ago, pundits were warning that Tiny could still be reelected even if the Dem were to have a 5M popular vote margin, so Biden’s lead is closer than it looks. The polls have consistently shown that 12-14% of the voters are undecided, likely 4-5% will vote for a minor party candidate, how the remaining undecided vote will determine who will be president in 2021.
It would be interesting to know the demographic makeup of these undecided voters.
Right now I think the most likely October polling will be a reversion to the mean. Biden will be up by 6-8 points. Normally an incumbent under 50% favorability would be a goner, but I think there is a 4-5 % racist gap (the tendency of racist politicians to do 4-5% better than polling, as people feel more comfortable voting for a racist in a secret ballot than they do telling a stranger over the phone that they support the racist), and Biden probably needs 3-4% of margin to win the EC.
So you put those numbers together and yes, a very, very tight race.
Trump’s floor is high 30’s. His ceiling is mid-40’s. It’s the second number that’s the problem for him. He got 46% of the popular vote. He has never polled above that. He has never polled net positive. For a while people were saying that it would change when instead of a generic Democrat we got down to actual cases. Well, no. The fresh, dynamic and exciting Mr. Biden is destroying him without ever leaving his basement.
A full 50% of registered voters despise him.
Trump is a psychopath with a specific, very limited set of skills. He’s only capable of attack. There is no second stage for him to light up, this rocket only had the one motor.
While this is gratifying, Democrats would be wise to ignore the polls and continue to operate as if we are in a neck-and-neck race.
It is extraordinarily difficult to unseat an incumbent anything – particularly a President. It rarely happens.
Trump will burn the world if he thinks it will help him win, and the presidency gives him the power to do so.
@MarkedMan: You are thinking of, I suspect, the “Bradley Effect”–but that tends to manifest when one of the candidates is black, not with two white candidates.
I would like to think that there are “shy” Trump supporters who are embarrassed to say they support him on a phone poll, but I am not sure that there are really that many who fall into that camp.
@Scott: Those nice folks at fivethirtyeight-dot-com have state polls right on their front page. Of course the first thing you see is their special homogenized number. With the thought that you’d find it useful (if very sparse and early):
MICH Biden + 9.3
PA Biden +5.8
WISC + Biden 7.4
AZ + Biden 4.9
FL + Biden 7.6
NC + Biden 2.0
Still, if you are pointing to the last Presidential election and saying we should worry about the outcome until there is a stake through the heart of this administration (METAPHORICALLY!!), I know millions of Democrats agree with you.
@Steven L. Taylor: I’m somewhat aware of the Bradley effect but I wasn’t citing anything so scientific. I’ve lived in a lot of different states and have always paid attention to politics and it’s my very unscientific gut feeling that when there are two white candidates and one of them is dog whistling so loudly it begins to attract negative comment, the racist outperforms the polls. I should have said “such people outperform polling by 2-3 percent but I think with Trump the dichotomy will be higher, especially as more and more people express contempt for Trump supporters.”
@Not the IT Dept.:
https://www.blogger.com/blogger.g#welcome
Here’s a place where you can start your own blog. You can do thoughtful analysis. You can identify all the candidates in a particular state to your hearts content.
You’re welcome.
@Steven L. Taylor:
I know of “shy” Trump supporters who won’t admit to voting for him publicly, but I have no way of calculating how many there are. I tend to think there would be more of them in blue states (and in certain professions such as academe, journalism, publishing, and entertainment) rather than red ones.
@Sleeping Dog:
Prominent among those were James Comey who thought he could afford to play politics with the GOP Senate at her expense and the supposedly liberal MSM, led by FTFNYT, who saw no harm in using Trump and HER EMAILS!! for click bait. Would Hillary have lost the famous 70,000 votes if Comey had followed policy and kept his mouth shut and if the MSM hadn’t piled on? At least Comey sort of apologized. I haven’t heard a peep from NYT indicating they understand what happened. They keep bleating about not understanding the electorate in 2016 without acknowledging their own role in shaping it.
@Scott:
I was about to make this same point. Thank you for bringing it up already. I’ve essentially round filed national level polling for anything. Politics is local, and we don’t elect presidents as a nation. I’d prefer that we did, but until we do the only polls that matter are the state level ones (accompanied by analysis of what they portend for the electoral vote distribution).
IMO national polls should be read primarily as indicators of momentum. They’re windsocks.
But let’s get over the semi-mystical dread of a magic Trump victory. He got very lucky in 2016. It was a black swan event, and events are called black swan events because they are extremely rare and tend to be over-interpreted.
A 78 year-old man we’ve all known since forever, operating from his basement, is outpolling the incumbent president by 10 points in Nate Silver’s weighted average, and is well ahead in state polls in FL, MI, WI, PA and AZ. Jut as important, he’s pulling even with, and may be moving ahead of, Trump’s fundraising.
Trump is not a wizard. He’s a psychopath who only knows how to attack. He has zero capacity to unite. He has shown no capacity to even get back to his initial 46%.
To re-use a metaphor, he’s a bad oyster the country is vomiting up.
FWIW, in re: national polling. I would again caution that we should not over-react to 2016 and say that national polls are worthless because they aren’t.
The real lesson of 2016 is that the EC is real and that states can’t be ignored. But, of course, the real lesson is that forecasts have to be prudent and that understanding probabilities and margins is super important.
I would note, again, that if Biden can poll consistently at 50%+, that is hugely important and that large margins in the national numbers can very much be an indication of relative position in the state numbers.
Keep in mind: HRC was consistently in the mid-to-higher 40s and the margin in 2016 was, over time, pretty close.
I am not predicting a Biden win, but the bottom line is that these really are extremely good numbers for him at the moment.
And, of course, as someone noted above, the Dems can’t coast (but I think they know that).
@Michael Reynolds:
Part of me thinks that his ability to stay in the basement, so to speak, helps him remain the “generic Democrat” which, I think, is what he wants to be in this campaign.
@Michael Reynolds:
Weak as he is, his skill with attack is still pertinent which is why I’m taking so little comfort from the polls right now.
When weak and cornered, the feral animal lashes out. Trump is feral. As Tony W notes, Trump will burn the world before he accepts defeat. Just last night in Phoenix, he claimed the coming election would be the “most corrupt” in US history.
Does anyone think Trump will graciously concede? Will he concede at all if the winning margin is small? Would anyone be surprised if he called on his Trumpaloons to protest the outcome, including veiled calls for violence?
Trump needs to be humiliated.
@Steven L. Taylor: Precisely. He’s able to stay the “sane” and moderate (white male) candidate. African-American support got him the nomination, while his ability to not scare anti-Hillaryites who might be wavering on Trump should be a net positive for him.
All he has to do is be a grown up. That’s a low bar, but the GOP side’s bar is subterranean at this point.
OTOH, I worrry that the NYT is pushing the “big lead” narrative in part for ratfkery. I don’t trust them as far as I can spit.
@Steven L. Taylor:
More precisely, Dems have to go for the jugular. A close election will not be enough to bring about the peaceful transition of power that is the tradition for which America is rightly proud.
I will defer to the last sane man to hold the office;
Well said, Mr. Obama.
@Michael Reynolds:
Despite the fact that a Senator from Montana counts as much towards who is majority leader as any other Senator. In the last Montana poll I recall, Bullock led the incumbent Republican Daines by +7. Real Clear Politics’ poll tracking average has Kelly up +10 in Arizona over incumbent Republican McSally. In Colorado, also seldom polled, Hickenlooper was most recently — like six weeks ago — up +18 against incumbent Republican Gardner (even though Hick hasn’t actually won his primary yet).
OTOH, speaking as a Mountain West Democrat, it may be just as well that the national party and media aren’t trying to throw their weight and opinions around in these races.
@Scott F.:
I have long argued, going back to when Doug was still around and with former regular Andy about Bernie Sanders, that every vote, even in a deep Red state against Trump matters as we, as a nation, need him to be as thoroughly repudiated as possible.
@Michael Cain: Polling is expenisve–that’s the bottom line. And quality polling is especially expensive.
@Scott: @HarvardLaw92:
It doesn’t help that people don’t understand margin of error in polling.
MOE applies to both individual numbers you see, not the difference between them.
Ex:
Republican 45%
Dem 47%
MOE +/- 3%
What that means:
Republican support is 42-48%
Dem support is 44-50%
The possible differential is much larger than 3%.
If you look at the last few state polls in PA,WI, and MI from 2016, they were not that far off the final result if MOE is applied correctly.
But there’s no reason to think Trump is going to get better on the virus and race relations—or anything else, frankly—between now and November.
And he’s now added DOJ corruption and foreign relations treason into the mix. The NYSD firing is going to backfire. Bolton’s book is revealing even more corruption. And Mary Trump’s book is going to expose his venality even more.
@Pylon:
Not to mention that the EU is likely not to allow US citizens free access to Europe due to virus situation here. Think about how well that is going to go over with the business community.
While the polls are interesting, we have the makings of an electoral disaster this fall. States are cutting way back on polling places and the Republicans are fighting mail-in votes. If COVID comes back for a second wave, we could be looking at a complete meltdown and an election either side can claim is illegitimate.
@Hal_10000: This a very real concern.
I will say that the wider the margins, the easier it will be the ameliorate some of that very real chaos potential.
The “Whatabout 2016?” hedges are hovering over a lot of people’s predictions; people are afraid of being wrong like they were in the last cycle, and this fear is making people a lot more cautious than they otherwise would be. The betting markets currently have Biden’s chances at about 55%–a favorite, and certainly higher than earlier this year when they projected Trump as the favorite. But it’s still much lower than the 75+% they were giving Hillary four years ago. They were burnt by what happened, and now they’re overcompensating. I really don’t think it’s any more complex than that.
@Kylopod: 100%
@Kurtz:
I think that applies to PA and MI. WI is a bit more complicated. RCP’s final average was 7 points off from the final results. However, the average included only two polls from the month of November, one that went no later than Nov. 1 and the other no later than Nov. 2. 538 had a bit more data, and two polls from WI in the final week of the campaign had Hillary up by only +3 or +1, which is indeed within the MOE.
People overlook the fact that Trump’s numbers surged in the final week of the campaign, and it happened so quickly that many pollsters failed to pick it up. So part of the apparent inaccuracy was due to the numbers simply being outdated. One of the lessons is that pollsters should have been paying more attention to those states; another is that a candidate’s level of support isn’t static and can change rapidly. But there’s very little supporting the conclusion that the polling was massively inaccurate per se.
@gVOR08: the worst NYT article was the one about Huma Abedin and Anthony Weiner’s divorce. The Times talked about how it created questions about Clinton’s judgment (how, exactly?), and would remind voters of Clinton’s own marital troubles. And of all things, they asked thrice-married Trump to opine on how bad this looked for Clinton (!).
@Blue Galangal:
They are literally reporting the results of their own poll, which is nearly identical to all the other national polls and the four-year rolling average of Trump approval/disapproval numbers. That’s quite the long con they’re playing.
@Kurtz:
That’s true of any individual poll. Anyone hyping up a single survey as especially predictive is either an ignoramus or a huckster. That’s why just about everyone points to indexes of polls and trends.
That Biden is up 12 points in this poll doesn’t tell us much of anything.That he’s up roughly that much in an average of multiple reputable polls does. And the fact that Trump has been underwater in every average of polls since a week into his presidency tells us a whole lot.
@James Joyner:
Absolutely, but I was referencing state polls because the typical dumb conversation elsewhere looks like this:
“The polls were accurate in 2016. Hillary won by ~3%.”
“National polls don’t matter. State polls do. We have this thing called the electoral college. Stop whining, you lost. #lockherup”
At this point the conversation either stops or devolves into more ignorant BS. State polls are typically noisier and fewer in number, which tends to reduce the efficacy of aggregation models.
As far as this year goes, the race will likely tighten. But I think the absolute maximum popular vote deficit that Trump could overcome in the EC is 5%. But the closer to 5 it is, the more unlikely it becomes. Trump likely needs to bring the spread down to around the same it was in 2016, and hope for another fluke.
@Kylopod:
I took a quick look, and the poll summary from WI I saw looked dire for Trump at the top line. But the MOE included the possibility of Trump +1ish. It ended up +0.7.
The latest one I looked at was from Clarity. I have to look elsewhere for the Marquette and Remington polls, because the links where broken. IIRC, they were all on the edges of MOE if interpreted correctly.
@Sleeping Dog: “Think about how well that is going to go over with the business community.”
I’m part of that business community — I do a great deal of work with a Dutch company. And I feel like I’m living on Molokai in the 19th century.
We’re going to get Trump’s wall all right — but it’s going to run all the way around this leper colony of a country, and it won’t be to keep people out…
@Michael Cain:
Indeed.
@Kurtz: Here are all the WI polls that 538 collected that included the month of November:
CCES (10/4-11/6) +5
Ipsos (10/24-11/6) +6
PPP (10/31-11/1) +7
Loras (10/31-11/1) +6
Remington (11/1-2) +8
Clarity (11/1-2) +4
SurveyMonkey (10/30-11/6) +1
Gravis (11/3-6) +3
SurveyMonkey (11/1-7) +2
Google (11/1-7) +12
Notice that the further the poll goes into November, the tighter it gets, and there are several that are well within the MOE of the final results. The only outlier is that Google poll that showed Hillary ahead by 12.
Of course there simply isn’t sufficient data here to draw any conclusions one way or the other. As James pointed out, any individual poll has barely any informative value, but the averages matter, and there just weren’t enough WI polls taken in the final week of the campaign to construct a reliable average. Yet it does seem that most of the pollsters that did conduct polls in that final week did indeed detect the Trump surge that was happening, and while none of them actually showed Trump in the lead, they were roughly accurate as polls go.
@Kylopod:
I’m not sure why you’re giving that information to me. Really, I’m not even sure exactly what this exchange is about. I argued that the people who say the polling was wrong typically don’t understand polling.
Specifically, my OP was just about the media and general population not understanding how MOE applies to polling. They look at the topline spread without adjusting both numbers.
Anyway, the numbers you posted validate my grouping WI with MI and PA.
@Kurtz:
I wasn’t disagreeing with anything you said, just adding to the discussion about putting the 2016 polls in context and realizing that they weren’t massively inaccurate as they’ve been commonly remembered. I see three basic points here. First, there’s the issue of MOE, as you outlined. There’s the fact that poll numbers can change throughout an election cycle, even though, generally speaking, candidates who have held a wide and consistent lead in national polls by this point have usually gone on to win. (Besides 2016, another exception was 1988, when Dukakis held a strong lead throughout the summer, though unlike in 2016, the polls shifted markedly toward Bush by September, so the outcome wasn’t a surprise by the time the election happened.) And finally, there’s the fact that polling of key states at the end of the 2016 cycle was sparse, which is more an indictment of the failure to conduct enough polls than it is of the accuracy of the polls per se.
@Kylopod:
Sometimes you get the shit that makes you paranoid.
My apologies.
Indeed…it is no mystery why one of their top reporters is badmouthing the Biden campaign…Biden’s team should give her the cold shoulder once he wins…
Geez, the biggest difference and why Joe is up by 12 is because HRC is not on the ballot!!!!! It helps when the most disliked candidate in your party ‘s history is not running. This election is a slam dunk winner for Joe. Why? Because he’s up by 22-26 in polls with women across the board. He is crushing the gender gap which Hillary won by only 12 against a misogynist! Women, particularly black women will give Joe the election. The youth vote is a myth and because of Obama, African Americans will turn out, but the main driver is the disgust white women now have with Trump, especially non college educated white women who went for Trump with 64%. That ain’t happening with Biden.
@Scott F.: “Weak as he is, his skill with attack is still pertinent which is why I’m taking so little comfort from the polls right now.”
Sure. Unless people come to the conclusion — hopefully egged on by the Democrats — that his skill with attacks is the only thing he’s got, and that he’s doubling down on attacks because he has no idea how to deal with the real problems that are hurting and killing Americans. Once that idea is set — and I think we’re getting pretty close to that point — then every new attack gets the same response: “There you go again.” And the attacks stop working.
There is a NYTimes/Sienna College Poll out of the 9 key battleground states…and Biden is ahead in everyone of them. Some of them by large margins.
And 538 has Biden up in polling averages of every battleground state (tied in Iowa).
I’d give one of my testicles for Biden to win 50 states.
@Scott F.:
I missed your comment before. I’m not sure what you mean by “skill[ed] with attack,” but there is simply no evidence that Trump’s attacks on Democrats have ever been politically effective. Hillary was already unpopular before he set his sights on her, and his attacks on her were never followed by any worsening in her poll numbers. Indeed, his poll numbers seemed to get worse the more attention he got, and in the immediate aftermath of the first debate, when millions of viewers got to see him lay into her on live TV, his poll numbers crashed and hers rose. He was the beneficiary of her demise, not the cause.