Joe Biden is continuing his snail-like progress toward a dead heat with Donald Trump in polling this week. The RealClearPolitics polling averages for a national head-to-head contest between the two presidents now show Trump up by a mere 0.2 percent (45.5 to 45.3 percent), his smallest lead in these averages dating back to last October. If you took a very outlierish Rasmussen Poll giving Trump an eight-point lead out of the equation, Biden would actually be ahead. As it is, he leads Trump in the most recent surveys by Reuters-Ipsos, I&I-TIPP, Data for Progress, NPR-PBS-Marist, and Quinnipiac, a pretty impressive collection of pollsters (all but I&I-TIPP are in the top-25 outfits, according to FiveThirtyEight’s ratings).
Trump is maintaining a slightly larger lead (1.9 percent) in the national five-way polls that include Robert F. Kennedy Jr., Cornel West, and Jill Stein, per RCP’s averages. RFK Jr. holds 10 percent of the 13.2 percent going to non-major-party candidates. So the larger field continues to help Trump and hurt Biden, albeit marginally.
Battleground-state polling has been sparse in recent weeks; the last public polls in Arizona, Georgia, Michigan, Nevada, and Wisconsin were from a March 24 Wall Street Journal survey. So Trump maintains his relatively robust leads in all those states. New polling in North Carolina (from High Point University and Quinnipiac) shows Trump’s lead in that state shrinking slightly to 4 percent. And fresh data from Pennsylvania via Franklin & Marshall has given Biden a slight (0.1 percent) lead in that state in the RCP averages. The trends for Biden overall are positive, albeit very slightly and slowly so.
In terms of where the numbers might go as we approach November, there are some even more positive sights for the incumbent. A fascinating new national survey from NORC published by FiveThirtyEight looked at how demonstrated propensity to vote affected presidential-candidate preferences, and the findings are potentially significant:
“When we broke out respondents by their voting history, we found dramatic differences in whom they support for president in 2024. President Joe Biden performed much better among frequent voters, while Trump had a large lead among people who haven’t voted recently. Specifically, among respondents who voted in the 2018, 2020 and 2022 general elections, Biden outpaced Trump 50 percent to 39 percent. But among respondents who were old enough to vote but voted in none of those three elections, Trump crushed Biden 44 percent to 26 percent.”
This survey reinforces evidence elsewhere that the traditional Democratic reliance on “marginal voters” has ended, and that now it’s Republicans who need an unusually high-turnout election to get Trump’s supporters to the polls. In the short term, this could mean that when pollsters begin to shift from registered-voter to likely-voter samples, Biden will probably get a boost (the sort of boost Republican candidates used to count on) in the comparative numbers. Whether that carries over to the actual results in November may depend on overall turnout levels, with Democrats holding an unusual advantage among the voters most likely to show up at the polls.
There are, of course, many other factors that will influence the direction of this contest, including the strength, wealth, and wisdom of the campaigns and of the national and state parties supporting them. But one thing to watch is whether the Kennedy candidacy, which is marginally hurting Biden right now, gets onto the ballot in all or most of the battleground states. At present, Kennedy’s campaign claims it has enough signatures to gain ballot access in Arizona, Georgia, and Michigan, and it’s in a dispute with Nevada over an early deadline for identifying a vice-presidential candidate that it missed, which may land in court. If Kennedy does gain the ballot access he needs, the big question will be whether his conspiracy-theory-drenched appeal has the sort of staying power that non-major-party candidates usually lack. If he fades, it will likely benefit Biden.
Real-world developments outside the campaign trail could matter as well. Team Biden has to worry about signs of renewed inflation. And all of Trump’s efforts to avoid a preelection criminal trial appear to have failed, at least in New York.
For now, this contest seems to be back to square one: very close and subject to a lot of cross-currents and events we can’t really predict.
Stephen,
I wish I could say. I know that they lean conservative as a site. I believe that they use a pretty consistent rule for deciding what poll gets included. I found it somewhat unsettling that the ICR poll got included only when it showed Bush leading more than any other concurrent poll. I read their site, http://www.pollingreport.com, http://www.race2004.net, http://www.electoral-vote.com, http://www.electionprojection.com, and http://www.cookpolitical.com, as well as this site. Then I try to come to a common conclusion across these sites. It’s clear that all show the race to be close. Jeff’s point on another thread about turnout seems right on target. Beyond that, I think there are others here far more expert than I on such matters.
In the meantime, enjoy a pint of someone’s best bitter for me. There’s nothing quite like it.
A question from London. Watching your polls in the Us with great interest. Can you give me an assessment of the way in which the realclearpolitics website posts polls. Are they, as they say, ‘seekers after truth’, or are they biased, conscious or not?
Best,
Steohen H.
It’s a conservative site, but realclearpolitics has a nice analysis of the state of the electoral vote.
For Kerry, it’s all about Ohio, with New Hampshire and Nevada as secondary targets. They miss Colorado, either by outright win or by referendum getting him 4 votes.
For Bush, the targets of opportunity are Wisconsin, Iowa, New Mexico, and Maine’s more conservative district.
They given Florida to Bush and Pennsylvania to Kerry, for now.
Per this analysis, for Kerry, losing Ohio means he has to get New Hampshire, Nevada, and some Colorado votes, while holding New Mexico, Iowa, and Wisconsin.
If he wins Ohio and New Hampshire, even if he loses Wisconsin and the Maine district, Kerry wins, if he can hold Iowa and New Mexico.
So…it seems time to focus on Ohio, New Hampshire, Nevada, Colorado, Wisconsin, Iowa, New Mexico, and Maine.
In the back of my mind, I see some opportunity for Kerry in Missouri, Arkansas, West Virginia, and Florida, and some danger in Minnesota, Oregon, Pennsylvania, and Michigan.
All I am going to say is that in doorknocking and registering voters. I have reached and registered people without phones who are adamantly opposed to Bush.
If the dems turn those voters out to vote GWB should be very very afraid on Nov 2……
Gallup would never try to develop a model to sample the phoneless…
Post Debate 2: We did well enough to keep the Big Mo’ so let’s keep at it. GOTV is crucial. So enjoy the post debate glow, and if you have not done so already, sign up for GOTV drives.
SJ-
Sorry, I fail to see the problem with choosing a sampling from people who watched the first debate. Anymore than I fail to see the problem with sampling only people who watched either one of the debates, no matter what the outcome of either one. It’s just as valid an assumption that people who watch the debates are likely voters than any criteria Gallup is using. What seems to be your main gripe is not the methodology, it’s that it shows Bush doing worse than all of the other polls (except those you arbitrarily dismiss as “partisan”).
But all of the polling shows that even a majority of Republicans felt Kerry did better in the first debate, so it does not defy reason that Kerry would score somewhat higher with the subset of people who watched that debate. Likewise, if Bush did better in that debate it would likely show him doing better overall than other polls that did not only sample that subset. I doubt that Ipsos knew going into their polling that Kerry would “win” the debate, but conspiracy theorists might believe that.
ARG and Zogby are Dem polls? I must’ve missed that. BTW, go ahead and ignore the AP poll if you want. Every other poll shows the race basicallly tied and Kerry with all the momentum. Tracking polls have him gaining, and even the Time poll that had Bush up 12 a month ago has it tied.
As for state polls, that poll for Colorado you dismiss is by Gallup, which is hardly pro-Dem. And most of the states Mr. Abramowitz cites have multiple polls, which he averages, hardly a Dem conspiracy tactic to boost Kerry.
The race is far from over, but Kerry’s debate performance has changed the dynamic, at least in the short term. Anyone who can’t see that is in denial.
Smooth: I suppose you also wouldnt change a thing about Iraq and that the economy is turning a corner. Are all of you in complete denial?
What happened to MO, NV, OR, & WV? These four states also appear to be swinging.
Jazz’s comments are somewhat valid, but let’s limit the reposting of an entire link.
-DS
There is no question, despite the protestations of Smooth Jazz to the contrary, that Kerry is on a roll. Smooth Jazz discounts as partisan polls that give him bad news like Zogby. Smooth Jazz should be aware that Zogby called both the 96 and 2000 elections accurately. While showing Kerry ahead in the battleground states Bush has a 1 point lead in the popular vote. Instead of trying to go to great lenghths to debunk polls that he doesn’t like he should go back to the right wing web sites and figure out what to do about it. It seems to me there is only one valid point to make about these polls and it comes out on the Democratic side: If the reports about new registrations are accurate in Democratic areas of battleground states and a reasonable number of these new voters turn out, pollsters just may be UNDERESTIMATING support for Kerry. Whatever the argument we will find out for sure on Nov. 2nd providing ALL legitimately cast ballots are counted this time.
Jazz, there’s some food for thought here. However, one word: hyperlink. Learn it. Live it. Embrace it.
Smooth Jazz-
That’s potentially interesting news on Ipsos. I went to the website, but would have to pay to actually use it.
As you point out, there report comes from a Bush partisan, so perhaps should be taken with of salt, though I’m sure he worked very hard at it.
It’s particularly interesting to me that he only selected 6 polls to look at as the non-Ipsos. At the RealClearPolitics site, I see 10 polls conducted post-survey, using a 3-way race. From these, I get a Bush lead of 48.0-45.9, which is a bit closer than the author’s 49.2-45.5. He wouldn’t by any chance be inclined toward cherry-picking would he? I’m perfectly willing to believe that Bush is up by 2.1% at this point. In fact, if I add in the two Ipsos polls (using two-way numbers from the AP one, since they seem not to have gone three-way), only narrows it to 47.6-46.3. If he’s interested in fighting over a difference of less than a percent in the Bush lead, he should be my guess. Seems like a sign of desparation to me, though.