Skip to content

Breaking news: The 2024 election isn’t tomorrow

Plethora of polling cannot predict how voters will react to events to come

Political analyst Stu Rothenberg cautions that current polls cannot predict the election year dynamics for 2024.
Political analyst Stu Rothenberg cautions that current polls cannot predict the election year dynamics for 2024. (Bill Clark/CQ Roll Call)

Turn on cable television almost any time during the day or night and you’re likely to be inundated with chatter about how terrible the polls are for President Joe Biden. You will hear the same thing on the Sunday political talk shows.

You’ll also hear that Biden is unpopular and too old — and that voters don’t want the 2024 election to be a replay of 2020, when Biden faced Republican Donald Trump. Polls will show that even Democrats think their party needs a different nominee.

The source of all these assessments invariably is a new poll that was just released by some media organization. 

I have been writing about and relying on public opinion polls for the past 43 years, ever since I arrived in Washington, D.C., to write about political campaigns and elections.

Back then, many journalists didn’t pay all that much attention to survey research. Most editors figured the average reader really wasn’t interested in getting into the political weeds, so relatively few wrote about House and Senate campaigns or spent any time on political handicapping.

Control of the House of Representatives was a forgone conclusion (Republicans had not held the House in almost three decades at that point), so House races were not regarded as all that important or interesting.

That’s how the Cook Political Report and the Rothenberg Political Report found a niche. When the GOP won the House in 1994, suddenly more people were interested in congressional elections.

Those of us who relied heavily on polling got data from media polls (and later colleges and universities that established polling institutes). We also received survey data from operatives at the campaign committees and political consultants — pollsters, media consultants, direct mail consultants, and general consultants — who were advising campaigns.

Kevin Phillips’ newsletter, The American Political Report, included some polling tidbits, but it was The American Enterprise Institute’s wonderful Public Opinion magazine and the independent newsletter The Polling Report that aggregated surveys and provided the data that we all needed to write about the voters, the parties and key issues of the day. 

Of course, we understood that some polls were released simply to help campaign fundraising or to try to build momentum for one candidate, just as we understood that some partisan pollsters and public polls were more reliable than others. 

Over time, more and more pollsters appeared, including those with few credentials and obvious partisan bias. Unfortunately, many junk polls were included in polling averages under the assumption that averaging polls will cure all ills.

The most frequently asked question probably is the presidential hypothetical ballot test — If the election were today, whom would you vote for? 

The problem, of course, is that the election is not today or next week or next month. It’s nearly 14 months away, and during that time, a lot of things can happen. 

Do you really think that a government shutdown won’t affect how voters view the parties, Biden, and Trump? 

How about the GOP’s quest to impeach Biden? Couldn’t that impact public opinion in a dramatic way?

What about strikes at major corporations, or more controversial Supreme Court decisions? Wouldn’t they change how the voters see the parties?

What about international crises, including the Russian war against Ukraine? 

Pollsters always warn that polls are a snapshot of things at a moment in time. They aren’t necessarily predictive because events will have an impact on what people think and how voters plan to behave. 

You wouldn’t know that from the way most in the media treat polling about the 2024 presidential contest. Every survey must be dissected, every subsample examined with a microscope.

Well, I have spent a lot of time over the past 40-plus years writing about polls and trying to explain what they say about the parties, presidential elections and congressional elections. I have always valued the data and used them carefully, invariably warning not to read too much into a single survey many months before voters go to the polls.

Elections create their own dynamic. As an election approaches, voters become more engaged, more energized. They sometimes change their priorities and the message they are trying to deliver when they go to the polls.

Recent Stories

Expelling bee — Congressional Hits and Misses

Appeals court rejects Trump push to dismiss Jan. 6 suits from lawmakers, police

Photos of the week ending December 1, 2023

House expels Rep. George Santos

Biden administration, states need to step up efforts to maximize CHIPS Act

Kissinger and the Capitol