Trump Could Win Re-Election

Larry Sabato lays out a quite plausible path.

Donald Trump is exceedingly unpopular and has been throughout his presidency. His staff and supporters treat him like a toddler. Yet it’s quite plausible that he’ll serve a second term. Indeed, as I’ve frequently argued, that’s the safe bet, absent an economic collapse, because re-electing our Presidents is the default position in American politics.

Larry J. Sabato and Kyle Kondik agree and lay out their case in the Washington Post.

It’s too early to handicap 2020, but Trump may try to capitalize on some of the same factors that helped three modern Republican presidents, Richard M. Nixon, Ronald Reagan and George W. Bush, win reelection.

The reelections of all three men were not always certain. Around this time in the 1972 election cycle, Nixon held only a modest lead over the early Democratic front-runner, Edmund Muskie, who in 1968 had been the vice-presidential running mate of Hubert Humphrey. In late January 1983, pollster Lou Harris found former vice president Walter Mondale leading Reagan 53 percent to 44 percent. John Kerry’s challenge to Bush was nip-and-tuck throughout 2004. Fast-forward to 2019, and Trump often trails some Democrats in presidential trial heats, but with his large, solid base and a continuing good economy, it isn’t hard to see how Trump could win again. 

Considering that Nixon and Reagan won re-election in landslides, it’s easy to remember that they weren’t considered shoo-ins from the start. But, indeed, they both made comebacks. Bush’s victory wasn’t ensured even on Election Day.

That is not to suggest that Trump is destined to win, much less that he would rebound to a gigantic victory like Nixon’s and Reagan’s. For one thing, the landslides that one finds at regular intervals throughout much of the 20th century don’t even seem possible in this highly partisan, polarized era. America is in a stretch of eight consecutive presidential elections where neither side has won the popular vote by double digits, the longest such streak of close, competitive elections in U.S. history. 

The streak itself is remarkable. Even more amazing is that it began with the 1988 election, which most of us consider an absolute landslide because of the Electoral College outcome (426-111) but was in fact much closer, 53.4% to 45.6% , in the popular vote.

Another caveat: Trump’s approval rating has been upside down for essentially his entire presidency, and he has shown no inclination to broaden his base of support by changing his policies or softening his sharp rhetoric. From that perspective, even matching Bush’s 50.7 percent in 2004 seems like a major reach. Yet Trump could again win the presidency without winning the popular vote because of the strength of his coalition in the crucial Midwest battlegrounds. 

Given that he won several of those battlegrounds by incredibly narrow margins last time, I’m skeptical of that. But there’s not much in the way of state-level polling available at this juncture, so it’s hard to make even an educated guess as to his current popularity there.

Trump is in the process of jumping one major hurdle: He lacks a major primary challenger. (Bill Weld, the 2016 Libertarian vice-presidential candidate who recently declared a GOP primary challenge, does not count as “major.”) With approval ratings among Republicans usually exceeding 80 percent, and with his allies firmly in control of the party apparatus almost everywhere, Trump has thus far boxed out major intraparty opposition. The last three reelected GOP presidents all waltzed to renomination. 

This might seem like a throw-in but it isn’t. In the last half-century, only two American Presidents have failed to win re-election: Jimmy Carter and George H.W. Bush. Both faced rather significant primary challenges: Teddy Kennedy in 1980 and Pat Buchanan in 1992. (Both of those races also featured significant third-party candidates: John Anderson in 1980 and Ross Perot in 1992.)

Trump is also going to be in a much better financial position than he was in 2016, when Hillary Clinton vastly outspent him. Trump already has $40 million in the bank for his reelection bid, and he should be able to raise hundreds of millions more now that his party is more completely behind him than in 2016. Money is not everything, as Trump himself showed in 2016, but any campaign would prefer having more, not less.

Of course, we have no idea who the Democratic challenger is going to be, much less how much money they’ll have to spend. Regardless, they’re right that Trump will have more than enough to get his message out.

The Internet will be a campaign wild card again. Trump will probably reprise his 2016 digital advertising strategy to dissuade specific demographic groups, such as African Americans and young women, from voting for the Democratic candidate. His army of domestic online trolls no doubt will also turn out in force, and foreign actors, particularly Russians tied to the Kremlin, will almost certainly try to influence the election. Don’t expect the Trump administration to devote a lot of energy to frustrating those efforts.

That’s . . . nefarious. But likely true.

The Democratic Party may inadvertently boost Trump if it gets carried away with an impeachment frenzy that prompts a voter backlash. Opposition to Trump will help unify the Democrats and fund the eventual nominee after a standard-bearer emerges from what is a giant and growing field of about 20 candidates. But one or more factions of the Democratic Party may emerge from the primary season disappointed and angry. Trump’s well-funded digital strategy will work to widen these fissures.

Nancy Pelosi has already figured this out, so it’s exceedingly unlikely that there will be a serious impeachment attempt. Of course, that could backfire in the opposite direction, making Progressives in the party angry.

Ultimately, Trump may turn out to be at the mercy of conventional factors. In 2016, academic predictive models based on fundamentals such as the state of the economy suggested that Trump, or any other Republican candidate, was in position to win the election or come very close. This time, such models(once they become operative next year) could make Trump the early favorite despite his poor approval ratings. 

Credit the powers of incumbency and a strong economy, the state of which may matter more to Trump’s odds than nearly anything else. Incumbency and the economy, among other matters, ended up being more than enough for Nixon, Reagan and Bush. Despite Trump’s unprecedented outlandishness, that same combination might work for him, too.  

This is the bottom line; the rest is just noise. If we go into another recession in time for voters to feel it, Trump is likely to lose. If we don’t, it’s going to be very hard, indeed, for his Democratic challenger to gin up the momentum to defeat him.

FILED UNDER: *FEATURED, Campaign 2020, Donald Trump, Public Opinion Polls, US Politics
James Joyner
About James Joyner
James Joyner is a Security Studies professor at Marine Corps University's Command and Staff College and a nonresident senior fellow at the Scowcroft Center for Strategy and Security at the Atlantic Council. He's a former Army officer and Desert Storm vet. Views expressed here are his own. Follow James on Twitter @DrJJoyner.

Comments

  1. There’s a middle ground, Madame Speaker: Censure. The House can, alone, pass a Censure Resolution that doesn’t invoke impeachment but makes the same case an impeachment resolution would make. The Senate could follow suit, as a simple majority is all that’s required. These are stand alone resolutions with no legal effect. If the House and the Senate could agree on language and each pass the SAME resolution, there would be no presentment issue, no opportunity to veto — only the scathing political rebuke of both houses of Congress against the Presidence corruption, incompetence, and intransigence.

    This path yields almost the political benefits of passing an impeachment resolution in the House, but avoids the embarrassing failure to remove the President in the Sentate — and because it’s purely the prerogative of each House, there’s nothing the President could do about it but Tweet and fume — and maybe overstep.

    19
  2. Daryl and his brother Darryl says:

    Depressingly true.
    This nation will not survive another 4 years in the same form it has survived the previous 243 years. If Dennison is allowed to stay in office until 2024; America will be a completely different place than it is today.

    10
    1
  3. Kathy says:

    One factor that may have led to Dennison’s victory in the EC, was the conviction that he couldn’t possibly win. This negated Heinlein’s Electoral Principle (there may not be a candidate worth voting for, but there sure is one worth voting against) on the Democratic side, which led to fewer crucial votes.

    So if scare stories of reelection motivate the Democratic electorate in battleground states, then I’m all for them.

    10
    2
  4. Tim d says:

    Sadly, I agree 45 may win the EC again, but it’s funny to note that a Dem incumbent with polling numbers in the low 40s would be cause for outright panic and despair among those of us on the left.

  5. gVOR08 says:

    @Kathy: I agree. I think there was a big “538 effect” in ’16. That’s an unfair name, as Nate Silver had it only about two to one against Trump. That aside, there was a widespread belief Hillary had it in the bag, driven I think mostly by a belief that nobody could really vote for that swine. It did suppress D turnout. And it allowed goddam James fracking Comey to suck up to the Rs in congress at Hillary’s expense.

    Went to get a FL drivers license and license plates yesterday. On my morning visit, there were signature collectors on county property for a state constitutional amendment to require ID to vote. On the afternoon visit, it was different people with a constitutional amendment to change one word in an utterly meaningless way that sounds like it’s opposing non-citizen voting. Looked it up online. Apparently it’s a well funded effort to put what sound like anti-immigrant measures on the ballot in 2020 to drive R turnout. Back to the days of Karl the rat bastard Rove’s anti gay initiatives.

    Trump aside, the Republican Party must be destroyed, and I don’t see how that happens. They have their own media and the Billionaire Boys Club money.

    I don’t know if the revulsion against Trump that drove the midterms will last another two years. At some point outrage fatigue may set in. Vote blue, no matter who.

    12
    2
  6. Franklin says:

    I’m truly worried we’re going to mess this up. Impeachment is a losing card for Dems (but feel free to indict him after he’s been voted out of office).

    We’ve also got to avoid being painted as “open borders” (even if that’s not actually what our policies propose), because that’s also a losing issue. The wall, as a symbol of the effort to stop illegal immigration, is Trump’s strong issue behind the temporarily-goosed economy. The wall brings out Trump voters. Separating children from their parents is *not* popular, however (at least as far as I can tell from conservative friends and family on Facebook), so hammer that.

    I think healthcare ideas are a winning issue, because the Republicans have nothing.

    And damn, never ever think you’ve won it until the votes are counted.

    3
    1
  7. @Butch Bracknell: I think that the only real “censure” is impeachment. Being impeached has stigma and is historic in scope–and it at least forces a trial.

    I do think that impeachment will be perceived as a unipartisan act, but censure by just the House would be even moreso. (and easily dismissable by the GOP and their media allies).

  8. James Pearce says:

    Yet it’s quite plausible that he’ll serve a second term.

    I’ve found the plausibility of this depends on who you speak to. There’s still a lot of BS you have to cut through on the anti-Trump side that allows the pro-Trump BS to flourish.

    2
    7
  9. @gVOR08:

    I agree. I think there was a big “538 effect” in ’16. That’s an unfair name, as Nate Silver had it only about two to one against Trump.

    People just don’t understand basic probabilities…

    And, really, while 538 gets the blame (unfairly) it was other prognosticators who had HRC at 90%+ (I think the NYT and HuffPo both, IIRC).

    6
    1
  10. Teve says:

    @gVOR08: Jim Wright was complaining about this a few months ago, he was accosted somewhere in the panhandle by some dumb Fox asshole wanting him to sign a petition or amendment and claiming that illegal immigrants were allowed to vote.

  11. mattbernius says:

    @Steven L. Taylor:

    It was other prognosticators who had HRC at 90%+ (I think the NYT and HuffPo both, IIRC).

    Sam Wang was among the worst for this (if not the worst). A lot of people, myself included, used his model as a sort of talisman to hand-wave away what turned out to be real concerns.

    It will be interesting to see how he handles predictions around 2020.

  12. Teve says:

    @Steven L. Taylor:

    People just don’t understand basic probabilities…

    back in the day I would try to explain to people things like, when the weatherman says there’s a 20% chance of rain, if it doesn’t rain one out of five times he says that, then he’s wrong, and… eventually I just gave up

  13. James Pearce says:

    @Steven L. Taylor:

    People just don’t understand basic probabilities…

    If only there were a publication, perhaps a blog affiliated with ABC News or something, that uses statistical analysis to help people understand basic probabilities.

  14. Kylopod says:

    To avoid getting repetitive from stuff I’ve written in other threads, here is my basic summary of why I believe Trump is at least 50-50 to winning reelection:

    1) His approval ratings likely understate his level of support, because there’s still a chunk of the populace who find Trump repugnant but who will still probably vote for him.

    2) I think it’s a big mistake to assume his 2016 numbers represented some kind of absolute ceiling on his potential support. The GOP is much more united behind him than they were last time around. Conservatives who were skeptical of him in 2016 have warmed up to him, and his status as the incumbent gives him a strong advantage. (Indeed, I think unity may be one of the biggest reasons why incumbents usually win reelection–which is also why incumbents who lose tend to be those who receive a strong primary challenge.) And just as Clinton lost potential voters who thought she had the election in the bag, the same was probably true of Trump: I’m sure at least some stayed home thinking their vote wouldn’t matter since Clinton was certain to win.

    3) A Generic Republican would be a favorite to win in this economy. Maybe Trump will do worse than a Generic Republican, but betting on Trump’s Trumpiness to doom him does not have a good track record, and besides, he could underperform and still win–which is basically what happened in 2016.

    4) At the state level, the Big Three (WI, MI, and PA) are probably all must-wins for Dems: winning just two of the three would still lead to a Trump victory unless Dems pick up 10 or more EVs from at least one other 2016 Trump state. The next-best candidates are probably AZ or FL, and neither will be easy.

    5) Um, Howard Schultz.

  15. Kathy says:

    @Steven L. Taylor:

    People just don’t understand basic probabilities…

    Witness the fortunes made by non-Trump casinos.

    To be fair, probability and statistics are really counter-intuitive. Take roulette, a simple game. It has two basic variants usually called single-zero and double-zero, meaning the number of zeroes, marked in green, in the roulette and the betting area.

    Both have the same payoffs, but single-zero roulette is a better bet. Why? because the best bets, from a house edge perspective, are even money bets on odd/even or red/black. But the double-zero wheel gives these bets a slightly smaller probability (also to the various number bets, seeing as there is one more number on the wheel), as well as a higher house edge.

    People also don’t understand house edge. They think a low edge means the bet is more likely to win. This is not so. the edge is how far from true odds the payoff strays. If we were to bet on a regular 6-sided die, each number has a 1 in 6 chance, so a true odds payoff would be six to one. A game offering, say, 3 to 1, has a huge house edge (50%), while one paying 5 to 1 has a merely high edge (16.6%). The odds of winning any given bet are the same, 1 in 6.

    Then there’s hedging of bets. This gives the house a bigger edge on any bet, sort of, but I don’t preach against it. In the single die game, betting on two numbers doubles your odds of winning, 2 in 6, but lowers your payoff from 5 to 1 to 5 to 2.

  16. Kylopod says:

    @mattbernius:

    Sam Wang was among the worst for this (if not the worst).

    It’s amazing people took Wang seriously to begin with. Wang had a history of wildly wrong forecasts. In 2004 his model gave John Kerry a 98% chance of winning–a conclusion I find truly bizarre since unlike Clinton in 2016, Kerry was pretty consistently trailing in most pre-election polls. In the 2010 Nevada Senate race, Wang gave Sharron Angle a 99.997% chance (I kid you not) of winning–then she went on to lose by more than 5 points. Now granted, the polls were wildly off in that election (538 gave Angle about an 83% chance of winning), but polling upsets like that happen often enough that any competent probability forecaster must take them into account; the idea that Harry Reid had only a 1 in 10,000th chance of winning a race where he was in single digits is insane.

  17. steve says:

    We usually re-elect presidents, and the incumbent usually wins if the economy is going OK. Trump also has favorability ratings among Republicans that have rarely been seen, and they are enthusiastic about him. Republicans LOVE it when he mocks the handicapped, declare the Mexicans are rapists and lies about everything while declaring himself the best at everything. Trump is giving Republicans exactly what they wanted. In a turnout election he is guaranteed that his voters will turn out.

    On the Dem side, while there is uniform disgust with Trump, there is not uniform support for any candidate. There is a good chance that just like in 2016 if Bernie does not win that his voters may stay home or vote third party. I think that Trump should be considered the favorite until the election returns are in.

    Steve

  18. Daryl and his brother Darryl says:

    @steve:

    There is a good chance that just like in 2016 if Bernie does not win that his voters may stay home or vote third party.

    This.
    Bernie’s ego is going to get a good stroking…but he’s going to give us another Dennison WH.
    Just imagine what a lame duck Dennison would be freed to do.

  19. Daryl and his brother Darryl says:

    Did anyone else notice that Dennison told us ISIS was defeated, but they just killed 359 in Sri Lanka?
    Yet he will likely be re-elected.
    I don’t know why I allow myself to get worked up over this shit…I’m a white male with maybe 20-30 years left…what do I have to worry about?

  20. mattbernius says:

    @Kylopod:

    It’s amazing people took Wang seriously to begin with.

    Confirmation bias is a hell of a drug.

    Not to mention a number of “serious people” seemed to be taking him seriously at the time. I didn’t have any history with him and didn’t do any due diligence.

  21. Andre Kenji de Sousa says:

    @Kylopod:

    At the state level, the Big Three (WI, MI, and PA) are probably all must-wins for Dems: winning just two of the three would still lead to a Trump victory unless Dems pick up 10 or more EVs from at least one other 2016 Trump state. The next-best candidates are probably AZ or FL, and neither will be easy.

    More or less. Democrats could win only PA and MI to reach 268 votes. You’d just have to win a single state to do the trick. It’s not that difficult, the problem is that so many people on the Democratic Party, specially among the extremely online, think that beating Trump is a piece of cake – that’s why so many ideas that are toxic among rural voters are being seriously debated(Or even there is a debate about having a Woman or a P.o.C as candidate instead of having a candidate that could win).

  22. @James Pearce:

    If only there were a publication, perhaps a blog affiliated with ABC News or something, that uses statistical analysis to help people understand basic probabilities.

    But, of course, this is the point. Silver is constantly trying to explain how odds work, and many people refuse to learn.

    538 had the final odds of a Hillary win at 71.4% and a lot of people (educated ones) thought this meant that HRC was a slam-dunk.

    Further, the model was a prediction of a popular vote win, and, well, I think you know where I am going with that.

    12
  23. @Teve:

    back in the day I would try to explain to people things like, when the weatherman says there’s a 20% chance of rain, if it doesn’t rain one out of five times he says that, then he’s wrong, and… eventually I just gave up

    Yes, I have a relative who likes to vocally criticize the rain predictions when there is a 50% or more chance of rain and it doesn’t rain. Sigh.

  24. Teve says:

    @Steven L. Taylor: then there’s the whole remember the hits and forget the misses thing, or in the weather example, the opposite.

    It’s a shame because short term (~3-5 day) forecasts have gotten amazingly accurate in the last 2 decades, thanks to sophisticated computer modeling.

  25. Scott F. says:

    @Daryl and his brother Darryl:

    …what do I have to worry about?

    The end of the republic as we know it?

  26. Kathy says:

    @Steven L. Taylor:

    Silver is constantly trying to explain how odds work, and many people refuse to learn.

    Yup.

    Near election day last year, 538 gave a 75% chance for the democrats to take the House by X number of seats. Some criticized this as a hedge, saying Silver wouldn’t commit to declaring which party would win the House .

  27. Gustopher says:

    From the quoted article, and echoed by James and Butch:

    The Democratic Party may inadvertently boost Trump if it gets carried away with an impeachment frenzy that prompts a voter backlash.

    I think this is dead wrong. In order to defeat Trump we have to do two things:
    1. Present a case why he shouldn’t be re-elected
    2. Present a case why we should be elected.

    Impeachment goes to number one, and focuses the media on Trump’s high crimes and misdemeanors in a way that a censure or investigations cannot. We need to reach the low-information voters who just skim the news, hear the Republicans supporting, and say “ugh, more partisan squabbling.”

    Impeachment is about educating that segment of the American people. An educated and angry populace might put enough pressure on the Senate to vote to remove him but short of the two thirds required — probably the best we could hope for, and utterly devastating.

    Otherwise, we have to make the case for the voters to dump the president based on he economy, white-nationalism, or moral revulsion. None of these are winning issues with most of the electorate right now.

    We also need to make a case for our candidate, with a positive vision for America. But that kind of comes second, after making the case that Trump shouldn’t be re-elected. People have gotten used to Trump.

  28. Franklin says:

    @Daryl and his brother Darryl:

    Did anyone else notice that Dennison told us ISIS was defeated, but they just killed 359 in Sri Lanka?

    Sure, a lot of us noticed. But that’s because we notice things that happen outside of America, unlike the typical enthusiastic Trump voter.

  29. James Pearce says:

    @Steven L. Taylor:

    Silver is constantly trying to explain how odds work, and many people refuse to learn.

    Is that what he’s doing, explaining how odds work?

    This is how 538 describes what they do: “Nate Silver’s FiveThirtyEight uses statistical analysis — hard numbers — to tell compelling stories about elections, etc.”

    That’s why I personally think it’s a little rich when they claim “you don’t understand statistics” when you understood their “compelling story” just fine.

  30. Kylopod says:

    @Steven L. Taylor:

    538 had the final odds of a Hillary win at 71.4% and a lot of people (educated ones) thought this meant that HRC was a slam-dunk.

    Not only that, but 538 repeatedly warned readers that Hillary was not a slam-dunk–and they were attacked by other forecasters (including Wang and HuffPost) for giving Trump such a high chance of victory. A few days before the election, they ran an article titled “Trump is just a normal polling error behind Clinton.” They also were among the earliest to notice that Trump had an EC advantage: that the chances of him winning the EC while losing the popular vote were substantially higher than the other way around, because a lot of Clinton’s support was concentrated in states that were less electorally crucial, and so was being effectively wasted. In other words, his votes were more efficiently distributed in states that mattered. 538 figured this out as early as the summer of 2016.

  31. The abyss that is the soul of cracker says:

    @Daryl and his brother Darryl: Alas, that horse jumped the fence in 2016.

  32. @James Pearce: Sigh.

    I am not even sure what you are trying to say save this is your typical contrarian BS.

    I knew that I would regret engaging.

    13
    1
  33. al Ameda says:

    @James Pearce:

    This is how 538 describes what they do: “Nate Silver’s FiveThirtyEight uses statistical analysis — hard numbers — to tell compelling stories about elections, etc.”

    That’s why I personally think it’s a little rich when they claim “you don’t understand statistics” when you understood their “compelling story” just fine.

    I understand statistics pretty well, and to me Nate Silver’s final pre-election numbers told me that Trump basically had to run the table (electorally) in about 7 states in order to win – and he did.
    I don’t see that Silver was wrong at all.

    And, yes, a lot of people don’t understand statistics very well. That’s not an aspersion, that’s true.

  34. James Pearce says:

    @Steven L. Taylor:

    I knew that I would regret engaging.

    You regret engaging?

    @al Ameda:

    And, yes, a lot of people don’t understand statistics very well. That’s not an aspersion, that’s true.

    And these people come away from reading Silver’s work with a very poor understanding of probabilities apparently. That’s not an aspersion on 538.

    It’s just true.

  35. al Ameda says:

    And these people come away from reading Silver’s work with a very poor understanding of probabilities apparently. That’s not an aspersion on 538.

    That’s right, if people don’t get it, it may not be the fault of Nate Silver.

  36. @James Pearce:

    And these people come away from reading Silver’s work with a very poor understanding of probabilities apparently. That’s not an aspersion on 538.

    Which is essentially what I said above.

  37. Kit says:

    @Teve:

    back in the day I would try to explain to people things like, when the weatherman says there’s a 20% chance of rain, if it doesn’t rain one out of five times he says that, then he’s wrong, and… eventually I just gave up

    It’s good that you gave up because what you said wasn’t correct. The chance of a coin landing on heads is 50%. The chance of a coin landing on heads once after two throws is… also 50% (not 100%): 25% of the time, it wouldn’t land on heads at all. It could rain all five days, and that doesn’t mean the odds were incorrect.

  38. wdp says:

    @Daryl and his brother Darryl: Yes. Same. It isn’t my social security President Poopsniffer and the Trumpicans target. It’s the kids.

  39. wdp says:

    @Kylopod: Agree. Contrast that to the Obama 538 odds. Given the margin of error, the Comey announcements, the russion farms, and the tiny wins in 3 critical states, I would say the 538 model was pretty good.