Trump Polls Don’t Deserve This Much Attention. By Jonathan Bernstein

www.washingtonpost.com

 | Bloomberg

5 - 6 minutes

About half of all Republicans say they won’t vote for Donald Trump if he’s convicted of a felony, according to a Reuters/Ipsos poll taken last week after his third indictment was announced. Don’t take those results too seriously.

To begin with, at this point most voters haven’t given much thought — any thought, really — to elections scheduled for November 2024. Because Trump is treated as a celebrity by the mass media, and as the Republican leader by the partisan media, even those paying only mild attention to politics have probably heard about the charges. But that doesn’t mean they’ve considered the news as carefully as they would a voting decision.

Then there is the uncertainty around any poll that asks about hypotheticals — that is, how you would vote if such-and-such situation holds. First, people tend to be bad at predicting how they would vote because they tend to underrate how important partisanship is to their choices. In real life, most people tend to be partisan, and it takes quite a jolt to get them to veer from the party line. Even if 15 months earlier they said otherwise.

Second, if a poll presents a hypothetical, then respondents are likely to overrate its importance, especially this far from an election. Why? Because the question asks people to think about one aspect of a candidate. But one of the main jobs of a political campaign is to get voters to consider the candidate whole.

If Trump is the nominee next year, for example, he’ll work hard to remind potential supporters about all the things they like about him — and all the things they dislike about President Joe Biden. Trump’s legal problems will be only part of the context of the election. Meanwhile, Trump’s campaign will also work hard to convince hard-core supporters that his legal troubles are merely evidence that his enemies are out to get him. Indeed, it’s already trying to do that.

All of this is basically standard stuff about survey research that most pollsters and academics would find familiar. But there are also issues particular to Trump and this election that make interpretation of this poll difficult.

Usually, it’s fairly easy to set expectations by referring to polling from earlier election cycles. Sure, there are always new wrinkles — analysts predicting the 2020 election from previous polls had to cope with how unusually unpopular then-President Trump was even during good economic times, and had to factor in the impact of a worldwide pandemic.

But there is now the possibility of an election with a former president as a major-party nominee, which hasn’t happened since well before survey research began.(1) Nor do previous examples offer insight about how indictments — and perhaps convictions, at least pending appeals — would affect a major-party nominee. And aside from the three (likely soon to be four) criminal cases he is facing, Trump has also been held liable for a sexual assault since the last time he faced voters.

All that said, there are some basic principles about polling that still apply: Early polling before an election year isn’t predictive. Elections with an incumbent are mainly about the incumbent’s record. Challengers are much less important than incumbents. Ideological extremism is at least somewhat costly to candidates. All of these factors argue for viewing these results with some skepticism. 

So what’s the the best way to think about all this? If he’s the party’s presidential nominee in 2024, Trump will almost certainly cost Republicans some number of votes compared to a generic Republican candidate. Whatever the polls show now, however, it’s difficult to say what that number will be — 1%? 2%? 5% or more? He could still win anyway, especially if Biden remains unpopular or becomes more so. The question for Republicans, both the party’s leaders and its voters, is whether that is a chance they want to take.

(1) Democrat Grover Cleveland won in 1892; Republican Teddy Roosevelt ran as a third-party candidate in 1912 (as Martin Van Buren had done in 1848).

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Jonathan Bernstein is a Bloomberg Opinion columnist covering politics and policy. A former professor of political science at the University of Texas at San Antonio and DePauw University, he wrote A Plain Blog About Politics.

More stories like this are available on bloomberg.com/opinion

©2023 Bloomberg L.P.