Trump

President Donald Trump gestures as he walks across the South Lawn of the White House in Washington, Saturday, May 30.

Will Donald Trump be reelected?

That question demands one of those complicated “on the one hand” and “on the other hand” answers that most people hate and pundits love. It’s why Harry Truman said he preferred one-armed economists — so there was no “other hand” to confuse things.

The better question is: Can Trump win an up or down referendum on himself with 46% of the vote?

That is a more precise way of framing the contest. Trump is an incumbent who has not grown his support base since his election, when he captured 46% of the vote. His base, while impenetrable, is not expanding. Over the past three years, his average job approval rating has mostly hovered between 42% and 46%. Recently, it’s slumped to the lower end.

Republican strategists point out that Trump usually does better than polls indicate — and they’re right. Pro-change voters broke for him in the last week of the 2016 election by a big margin. But, remember, the 46% he received on Election Day was him doing better than the polls.

Looking to November: The average of the twenty most recent national surveys puts Joe Biden ahead, 49% to 42%. If Trump again ends up with 46%, he’d be doing better than the polls now indicate.

Trump’s supporters will logically ask: If 46% was enough to win with last time, why won’t it be enough this time? The answer is simple. In 2016, third-party candidates pocketed 6% of the vote and this year, for a variety of reasons, that vote may fall as low as 2% or 3%. If Trump’s base doesn’t grow, then its power will shrink as the anti-Trump vote coalesces around one opponent.

That’s why it’s essential for Democrats to keep the anti-Trump vote from splintering. When Trump beat Hillary Clinton in critical Michigan by only 10,704 votes, leftist third-party candidate Jill Stein syphoned off 51,463 votes. Had Stein not been in the race, it’s likely most of her votes would have gone to Clinton.

Keep in mind, today’s political dynamics differ from 2016.

Four years ago, America was in revolt on the right, left and middle. Tea Party conservatives opposed the Republican establishment, progressive activists opposed the Democratic establishment and independent-minded voters opposed all of it, angered by a witches’ brew of partisan hypocrisy, institutional corruption and government paralysis.

The last election was ripe for protest and third-party voting. But this year, all of the nation’s anger, fears, hopes and dreams center on reelecting or defeating one man. The 2020 election will be a referendum on Donald Trump, as is usually the case with incumbent presidents.

Herbert Hoover and Jimmy Carter lost reelection because it was a referendum on them, not on their challengers.

How does all this translate into electoral votes?

As we know, Clinton led Trump 48% to 46% in the 2016 popular vote, and third-parties collected 6%. Trump was able to survive that two-point deficit by narrowly carrying key swing states, which gave him all of their electoral votes. He won Michigan by two-tenths of a point, Pennsylvania and Wisconsin by seven-tenths of a point and Florida by 1.2 points — and picked up 75 electoral votes, more than enough to win the Electoral College.

But can Trump’s narrow swing-state margins survive a wider national popular vote deficit?

For example, if Trump holds his 46% base and the third-party vote drops to 3%, that leaves 51% of the popular vote for Biden. Such a five-point Biden lead could overwhelm Trump’s vote in key states in ways that Clinton’s two-point nationwide lead couldn’t.

Between now and November, Trump’s campaign will try to expand its vote beyond 46% and, in the process, jack up Biden’s negatives. The Democrats’ dangerous swerve to the left has made that possible, they believe. They will also attempt to reframe the election as a choice between candidates, not as a referendum on the incumbent. But as long as the president’s negative ratings are so high — between 53% and 57% — dislodging the referendum dynamic will be difficult.

Joe Biden might as well change his name to “Joe Not Trump.” That’s how he needs voters to see him.

Ron Faucheux is a nonpartisan political analyst based in New Orleans. He publishes LunchtimePolitics.com, a newsletter on polls, and is author of "Running for Office."

Lanny Keller: Once the party of Lincoln, now the cult of Donald Trump