And yet recent 2024 polls, which serve, at best, as snapshots of the electorate a year out, become news events unto themselves, generating reams of coverage and endless commentary. They’re not actually breaking news events, like, say, a train derailment, even if treated as such. They’re more creations of a media industrial complex that longs for easy data points, for things that feel like facts but are actually imprecise measuring mechanisms. 

For every piece that is directly about polling, like one from Politico proclaiming “the polls keep getting worse for Biden,” there are others based on the suppositions gleaned from poll results, such The Washington Post examining “Trump’s improved image.” Even pieces downplaying some headline-grabbing polls as the “wrong” ones, may seize on others to make a point.
“The odd thing about media polls is that they are reported as a newsworthy event, but this kind of event ‘happens’ only when a newsroom decides it’s time for one—and when it has the money,” NYU journalism professor Jay Rosen wrote in an email.
Polls may fall into the category of pseudo-events, a term coined in 1962 by Daniel J. Boorstin and defined as something that “planted primarily (not always exclusively) for the immediate purpose of being reported or reproduced.” Just like polls, a pseudo-event’s “relation to the underlying reality of the situation is ambiguous.” (This idea was recently discussed by on John Dickerson on Slate’s Political Gabfest episode on polling episode). In this way, a poll may be more like a press conference, something that is created to shape a narrative.

-snip-

Rosen has been out front this presidential election cycle with an “organizing principle” for journalists: “Not the odds, but the stakes.” The focus, he argues, should be “not who has what chances of winning, but the consequences for American democracy.” Placing too much emphasis on polls can shift the political conversation from critical reporting about what’s happening—such as the impact of Biden’s administration’s policies or Trump’s authoritarian plans for a second term—to predictions about what may happen a year later. 

So, what they apparently rely on now is “weighting”. Weighting is nothing new. It has been used in polling forever to cure some faults in sampling. If you find that your sample is under-or-over-supplied with people who fall into some known demographic category, you correct for that in the analysis phase. Up to a point, that can work.

Up to a point. But from what is leaking out of polling organizations now, that point has long been passed. According to an insider who made a comment here some months ago, now it’s basically ALL weighting. They try to line up the results according to a predetermined idea of what would likely be said by a truly random sample, by aligning as many demographic factors as they can with an abstract idea of how those measurements would line up in the real world.

Yet they can have less and less idea, as time goes by, of how the population actually falls according to any of these factors. There are no real base lines. They can say, “We don’t have enough Catholics”, or “We have too many people with college degrees”, and then try to align the results accordingly. Yet they are further and further from actually knowing how many Catholics there really are, or even how many people with college degrees live in given areas. On some factors they can try to align with the most recent Federal Census, but that’s extremely limited because the Census no longer collects any more than the most basic, meager information. On other possibly relevant factors there’s no data at all. In fact it’s reached the point where it’s not actually possible to know about many factors that might be relevant—new ones that have cropped up. It’s just too long now since there was any reliable base data.

So they may try to correct for a multitude of factors, without having any sort of base-line for those factors. Then on top of that, according to the person who wrote that comment, there’s a final “hand-waving” stage where they say in effect, “That just doesn’t look right”, and try to adjust the figures until it does.

This survey, which identifies key issues in the 2024 election and measures how much voters trust Biden vs. Trump on these issues and how Biden vs. Trump perform on a ballot in swing states, was conducted from Oct. 5-10, 2023, among a sample of 5,023 registered voters in Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin. The interviews were conducted online, and the data were weighted to approximate a target sample of swing state registered voters based on gender, age, race, marital status, home ownership, 2020 presidential vote and state. Results from the full survey have a margin of error of +/-1 percentage point.

Another change since 2020 is the broad public backlash, especially in Democratic-leaning and swing states, against the 2022 Supreme Court decision ending the constitutional right to abortion, which Trump has directly claimed credit for engineering through his nominations to the court. Finally, compared to 2020, the electorate in 2024 will likely include significantly more young people in Generation Z, a group that is preponderantly supporting Democrats, and fewer Whites without a college degree, now the GOP’s best group.

You’ve probably heard pundits tell you to ignore polls this far out from an election. This is not advice people are taking this week as surveys keep showing President Biden and Donald Trump tied after 91 felony charges and a string of positive headlines on jobs and inflation. Voters seem more judgmental than ever about Biden’s age, while polarized along familiar lines on the charges against Trump, and even a little nostalgic for his economic record.
But there are reasons campaign professionals tend to wave off early surveys, which have notoriously overstated threats to incumbents. Here’s why pollsters and strategists we spoke to say they’re not panicking yet.
There is no campaign. Every time you see a poll showing Biden’s approval in the 30s, mentally add an asterisk that says “before Democrats spend $1 billion.”
This isn’t so much about a prohibitive spending advantage (Republicans will have money too), but about what that money goes towards. In this case, it’s a message that so far has worked for Democrats in real-life conditions.
In the midterms, postmortems found that Democrats performed poorly in noncompetitive contests, but they won big in highly contested, swing state races where they could devote millions to ads on issues like abortion, drug prices, and entitlements while painting their opponents as “MAGA extremists.” The same formula has held up well in off-year elections since then, including a blowout judicial race in Wisconsin centered on abortion rights and gerrymandering.

In a bid to keep the Special Grand Jury’s recommendations secret in January, Fani Willis said the charging decisions were “imminent.” Since then, however, the regular Fulton County grand juries that would have to charge Trump and others have been churning out indictments for more ordinary crimes. According to Andrew Fleischman, there are 18,000 pending felony cases in Fulton County, many of them being held pre-indictment. Like some of the delays in the January 6 investigation, this backlog stems in part from COVID restrictions.

But it wasn’t just that backlog that has delayed charges against Trump. In March, Willis asked Christina Bobb for an interview (who refused). It may be that, after reading Bobb’s January 6 Committee testimony (transcripts of which were only released after the Fulton Special Grand Jury expired), Willis discovered that, while Bobb claimed to have been uninvolved in the crimes in Georgia, she testified that she and, “at least two dozen others,” over at least two rooms, sat in on Trump’s call to Brad Raffensperger, and “we all thought … it was totally fine.” On top of discovering that there were up to 24 witnesses who might be willing to misrepresent the call at trial, this may have caught Rudy Giuliani in a lie. After it became public, Rudy amended his interrogatories in Ruby Freeman’s lawsuit to reflect some involvement in the call as well. Someone recently claimed to me that Willis’ case is “open and shut.” But it’s not “open and shut” if there were 24 unknown witnesses involved.

More famously, according to a letter seeking to disqualify an attorney representing most of the fake electors, Willis has been spending recent weeks interviewing fake electors and telling them, allegedly for the first time, that they could get immunity deals if they testified against other Republicans. Friday, one of the fake electors who also accessed voting machine data on January 7, joined Trump’s effort to undercut Willis’ authority, represented by a new attorney. All of which suggests that Willis is spending time not just making charging decisions, but making sure she can win the case.