An August ultimatum: No recess until redistricting reform is done
Mellman: Trump voters cling to 2020 tale
While most Americans rightly see Joe Biden as the legitimate winner of the 2020 presidential election, time, facts and evidence have done little to shift the attitudes of Trump voters who stubbornly cling to the fantasy that their candidate actually won.
In Arizona, for example, a poll my firm completed at the end of May found just 11 percent of Trump voters accepted the fact that Biden won the most votes in that state's presidential contest. Three-quarters of Trump voters remain convinced the defeated former president was the actual winner.
Only 10 percent of Trump supporters believe Biden won the state fair and square, with 83 percent maintaining Biden only appeared to win because of fraud.
How much has changed among Trump voters since our previous poll in December? Precious little. The number who said Biden prevailed fairly grew by just 3 points, while the number who saw Biden's victory as fraudulent fell by a single percentage point.
A lot happened since December. The Electoral College met; Congress counted their votes and declared Biden the winner; Biden was duly inaugurated as president; and Trump forces lost at least 86 lawsuits challenging the results.
All those facts, and yet, Trump voters in Arizona remain convinced, without an iota of evidence, that the former president actually won their state and the race nationwide.
What can we learn from this irrational stability?
First, persuasion is difficult, at least when something important is at stake on which people have developed strong views. All the events, actions and statements since November have failed to alter the beliefs of Trump voters.
In our hyper-polarized society, partisan hearts and minds are very hard to change.
Second, as I suggested last week, many conservatives are, and believe they should be, importantly impervious to evidence. Conservatives tend to discount the opinions of elites and the evidence they cite.
A survey conducted nationally, just after the November election by professors Gordon Pennycook and David Rand, found 40 percent of Trump voters admitting they would not be persuaded of Biden's victory, either by Trump losing his legal challenges or by Trump conceding. Evidence is just not terribly relevant to a sizable segment of the GOP.
Third, and most important for this piece, people can be poor predictors of their responses to such facts. Pennycook and Rand's survey found 27 percent of Trump voters saying they would be convinced that Biden was legitimately elected if Trump lost his legal challenges.
They were wrong about themselves.
As noted, Trump lost his legal challenges - over 80 of them.
If 27 percent of Arizona Trump voters were to have changed their minds as a result, we would have seen a nearly 23-point decline in the perception that Biden only won by virtue of fraud. Instead, we saw a measly 1-point drop. Trump voters thought the court losses would change their minds. But their predictions about what would alter their own attitudes proved completely inaccurate.
Of course, it is possible Arizona Trump voters are somehow unique and would have been fully cognizant of the fact that the results of the court cases wouldn't change their thinking, but that seems highly unlikely.
More probably they too failed to predict what they would react to and how.
This finding bolsters an argument I've been long making about message polling. Voters often can't tell us how and why they make the decisions they reach. We can ask questions and they will give us answers, which we turn into numbers that appear precise.
You've seen versions of those questions on nearly every message poll you've examined: "Would you be more or less likely to vote for a candidate who ____?" "If you knew candidate x did y, would you have serious concerns about voting for them?"
Often the answers to such questions are mush precisely because, just as in this case, people are poor predictors of their responses to new information.
Mellman is president of The Mellman Group and has helped elect 30 U.S. senators, 12 governors and dozens of House members. Mellman served as pollster to Senate Democratic leaders for over 20 years, as president of the American Association of Political Consultants, and is president of Democratic Majority for Israel.