In the history of presidential polling, there have been six occasions when an incumbent faced an electorate that favored the challenger. How many polls have you seen since the 2016 election? I can’t count them all. But now, months after the election, there seems to be a consensus among political scientists that Trump’s getting clobbered in 2020 could be problematic for Republicans.
Pollsters who have conducted surveys before the Trump presidency usually face obstacles in figuring out whether people feel strongly enough about an issue to matter in a presidential election. Surveys in the Trump era have been largely positive. (The issues in every example include his strong approval ratings — though few dispute that more people approve of him than dislike him.)
Pollsters typically only release surveys when they say they are pretty confident they will find an answer. Most presidential polls have also been conducted before the release of the presidential debate — then, as now, it’s difficult to find a solid answer. (I should add, however, that Trump regularly names different polls and claims that they are so wrong and consistently over-count the level of disapproval that they are corrupt, dishonest and useless.)
Yet what if all of this simply reflects the fact that the political electorate has become so hyper-partisan, so deeply polarized, that people are more likely to say one thing on television than they are to say it on a survey? How different would polls have been a year ago, or six months ago, or a month ago? (I suggest you take a look at the superb and detailed Times analysis by FiveThirtyEight’s Nate Silver, which relies heavily on polls.)
It is possible that many of the surveys have found a tight race, with turnout likely to be low. (The Trump administration’s contorted claims about voter fraud may have already helped that outcome.) Or it’s possible — I think — that the polls, if examined in this light, would show Trump doing well.
One thing to keep in mind is that polls are only a snapshot of public opinion. Another is that early presidential primaries are usually free from Trump’s effect. It’s true that Arizona Sen. John McCain ran in 2008 and lost to Obama. But that was long before Obama went on to win Arizona by 27 points. Romney did badly in Florida, but Obama was up eight points there months before the 2012 election. Romney also did poorly in Ohio, but Obama lost that state by seven points.
No matter how the 2020 presidential election turns out, history would be less favorable for Trump than the nine instances in which President Clinton has faced an electorate that favored the challenger. (“Chad” Clark ran against Clinton in 1992, and he ran into trouble.)
I also don’t think it is unfair to note that in all those cases, despite their national popularity — and despite the conventional wisdom of the moment — the challengers lost. But in hindsight, none of them look terrible.