Tulsi Gabbard For President?

Add yet another name to the potential Democratic campaign field in 2020.

Congresswoman Tulsi Gabbard of Hawaii is reportedly considering a run for the Democratic nomination in 2020:

Democratic Rep. Tulsi Gabbard of Hawaii is considering running for president in 2020, a source with direct knowledge of her deliberations told POLITICO.

Rania Batrice, an adviser to the progressive congresswoman and deputy campaign manager on Bernie Sanders’ 2016 presidential campaign, has been putting out feelers for digital and speechwriting staff for Gabbard. One person approached about the positions say that 2020 wasn’t mentioned explicitly, but it was heavily implied.

Batrice denied that the staffers are being hired for a presidential campaign. She did not dispute, however, that Gabbard is considering joining what’s expected to be a crowded field of Democratic presidential contenders.

“I think everybody is focused on 2018, but we will see what happens after that,” Batrice said in an interview. “Someone like Tulsi, with her experience, is an important voice in the party and the country.”

Top aides to Gabbard did not respond to multiple requests for comment.

Amid the clamor of Trump headlines and focus on higher-profile candidates, Gabbard has been quietly making the traditional moves of a presidential candidate. She recently visited Iowa, where locals urged her to run for president, according to the Iowa City Press-Citizen. She keynoted a progressive gathering in New Hampshire in September. And she’s writing a book due out this spring titled, “Is Today the Day?: Not Another Political Memoir.”

Gabbard, an Iraq War veteran who has represented Hawaii in Congress since first being elected in 2012, previously served as a member of the Honolulu City Council and the Hawaii House of Representatives, but she did not really achieve national prominence until the 2016 Presidential election cycle when she resigned her position as a Vice Chair of the Democratic National Committee over objections to a primary debate scheduled she and others deemed to be too preferential to Hillary Clinton. Not long thereafter, Gabbard became one of the first Members of Congress to endorse Vermont Senator Bernie Sanders in the fight for the 2016 Democratic Presidential nomination. She has also been the source of controversy in recent years for actions such as meeting with Syrian leader Bashar Assad last year, after which she urged caution in the use of military force. She also raised concerns among some of her progressive supporters by meeting with Donald Trump during the transition period between Election Day 2016 and Inauguration Day.

Realistically speaking, it’s hard to see Gabbard as a top contender for the nomination. Partly because she represents Hawaii, she’s not exactly well known among the average voter or even with the rank-and-file of her own party. Additionally, Gabbard would be running against the fact that, generally speaking, sitting members of the House of Representatives have not been successful in bids for the Presidency. Americans have only elected such a person President once since 1789, and that happened in 1880 when James Garfield, who also happened to be a Civil War hero, was elected President in 1880 with barely a 2,000 vote advantage in the popular vote. Even leaving aside this history, it’s hard to see how Gabbard would differentiate herself from other progressive candidates who may run in two years, including Bernie Sanders, Elizabeth Warren, Cory Booker, and Kamala Harris.

Gabbard’s comments are not, of course, the formal start of a Presidential campaign, and she has not even established a Presidential Exploratory Committee at this point, so it’s entirely possible that nothing will ever come of this. All the same, this is yet another example of how wide open the Democratic field will be in 2020, at least in the early stages. It may not get quite as crowded as it was for the Republicans in 2016, but it’s going to be close, and given the way that Democratic primaries work that could lead to a chaotic and hard-fought race that, as in 2008 and 2016, won’t be fully resolved until the end of the process in June 2020.


FILED UNDER: Campaign 2020, The Presidency, US Politics, , , , , , , , , , , , , , , , , ,
Doug Mataconis
About Doug Mataconis
Doug Mataconis held a B.A. in Political Science from Rutgers University and J.D. from George Mason University School of Law. He joined the staff of OTB in May 2010 and contributed a staggering 16,483 posts before his retirement in January 2020. He passed far too young in July 2021.


  1. Stormy Dragon says:

    She also has a weird blind spot for Narenda Modi, to the point if becoming an apologist for his ethnic cleansing of Indian muslims, that I find disturbing.

  2. Kylopod says:

    It may not get quite as crowded as it was for the Republicans in 2016

    Looking at the current roster of people either threatening to run or being strongly speculated about, I think it very easily could become even more crowded:

    Joe Biden
    Bernie Sanders
    Elizabeth Warren
    Michael Avenatti
    Cory Booker
    Kamala Harris
    Eric Holder
    Michael Bloomberg
    Tom Steyer
    Eric Garcetti
    Julian Castro
    John Hickenlooper
    Deval Patrick
    Kirsten Gillibrand
    Amy Klobuchar

    I got that from glancing at the list on Wikipedia, which is actually far longer; I just picked the names I’ve at least heard are quite likely looking into it, or at least being talked about. That’s 15 right there, and given how new names like Gabbard are popping up every day it seems, there could easily be more. There were a total of 17 Republican candidates in 2016.

    Of course, it’s also possible many on the above list won’t run in the end. Still, I would not be surprised to see a situation like we did in 2015 where the networks create a “kiddie debate” for the less popular batch.

  3. Just nutha ignint cracker says:

    @ Stormy Dragon: Alas, that could well be the start of a workable “Southern Strategy” for progressives;.

  4. Franklin says:

    I dunno. Was she born in Hawaii? Because some people don’t seem to accept birth certificates from that place.

  5. @Franklin:

    Probably even better for a birther conspiracy – American Samoa.

  6. Crusty Dem says:

    No. Just no. She’s awful…

  7. Neil Hudelson says:

    So we can exchange a Republican apologist for autocrats with a Democratic one?

    Hard no.

  8. Kathy says:


    At some point, a list of which Democrats are not running becomes more manageable.

  9. Jen says:

    @Franklin: There are apparently some Americans who do not think Hawaii is part of the US, or unsure about it. So, there’s that.

    I’m becoming increasingly concerned that 2020 is going to be a free-for-all with too many candidates and we’ll end up with a fractured vote mess like the Republicans had in 2016.

  10. Scott G says:

    @Neil Hudelson: Tulsi stands strong against autocratic policies, take the time to learn the truth about Syria
    Gabbard said regarding her trip to Syria: “I traveled throughout Damascus and Aleppo, listening to Syrians from different parts of the country. I met with displaced families from the eastern part of Aleppo, Raqqah, Zabadani, Latakia, and the outskirts of Damascus. I met Syrian opposition leaders who led protests in 2011, widows and children of men fighting for the government and widows of those fighting against the government. I met Lebanon’s newly-elected President Aoun and Prime Minister Hariri, U.S. Ambassador to Lebanon Elizabeth Richard, Syrian President Assad, Grand Mufti Hassoun, Archbishop Denys Antoine Chahda of Syrian Catholic Church of Aleppo, Muslim and Christian religious leaders, humanitarian workers, academics, college students, small business owners, and more.

    Their message to the American people was powerful and consistent: There is no difference between “moderate” rebels and al-Qaeda (al-Nusra) or ISIS — they are all the same. This is a war between terrorists under the command of groups like ISIS and al-Qaeda and the Syrian government. They cry out for the U.S. and other countries to stop supporting those who are destroying Syria and her people.

    I heard this message over and over again from those who have suffered and survived unspeakable horrors. They asked that I share their voice with the world; frustrated voices which have not been heard due to the false, one-sided biased reports pushing a narrative that supports this regime change war at the expense of Syrian lives.

    I heard testimony about how peaceful protests against the government that began in 2011 were quickly overtaken by Wahhabi jihadist groups like al-Qaeda (al-Nusra) who were funded and supported by Saudi Arabia, Turkey, Qatar, the United States, and others. They exploited the peaceful protesters, occupied their communities, and killed and tortured Syrians who would not cooperate with them in their fight to overthrow the government.”
    The Syrian People Desperately Want Peace

  11. Stormy Dragon says:


    Considering that free-for-all left the Republicans in charge of the entire government, perhaps the Dems could use a free-for-all with too many candidates.

  12. Matt says:

    Rep Gabbard stepped down from the DNC, not over the debate schedule, but because she intended to endorse Bernie Sanders. According to DNC rules, members are required to maintain neutrality during the primary. Interesting how you chose to frame that decision.

    Also failed to note that she’s won each of her House elections by substantial margins. Or that, after choosing to endorse Sanders, she was informed by establishment insiders that they would no longer assist her in fundraising, because “Hillary WILL be the nominee”.

  13. Kathy says:

    @Stormy Dragon:

    Correlation is not causation (someone should trademark that slogan).

    I wouldn’t mind if the Democrats took the trifecta in 2020, so long as they don’t elect someone like Avenatti to any kind of office (mush less the White House).

  14. Kylopod says:

    @Stormy Dragon:

    Considering that free-for-all left the Republicans in charge of the entire government, perhaps the Dems could use a free-for-all with too many candidates.

    Kathy said “Correlation isn’t causation,” but I think the real fallacy here is Post hoc ergo propter hoc: the fact that A is followed by B doesn’t prove that A caused B.

    There are two elements to this. First, the GOP won control of the entire government, but with an unqualified clown at the top. Most Republicans may not mind that fact, but that doesn’t mean it’s the sort of thing the Dems ought to emulate.

    The second element is a little trickier. I don’t believe the free-for-all nomination contest that ended with Trump as the nominee helped the GOP win the White House. They did win the White House, but I think they would have won it more easily with a more traditional candidate such as Rubio or Kasich. That’s impossible to prove one way or the other, and I know there are people who disagree with me on this. But I think people are making a huge mistake when they talk of Trump as some formidable behemoth of a candidate who will steamroll over just about anyone in his path. I find it interesting that people make that assumption at the same time they focus on Hillary’s weaknesses. You can’t have it both ways. If Hillary was as weak a candidate as she’s reputed to be, then the fact that Trump came within a hair of losing to her (and actually did lose the popular vote to her) is not exactly a testament to his strengths.

    Of course the Dems are a different party, and there’s no reason to think a free-for-all would necessarily lead to the nomination of a Democratic equivalent to Trump (whatever that would mean). Even the rules of the Democratic primaries (which tend to allocate their delegates proportionally, as opposed to being winner-take-all like most GOP primaries) might ensure a different result were a Democratic Trump to emerge–and that’s assuming such a figure would hold any appeal to Democratic voters.

    Still, a crowded field with no clear front-runner could potentially lead to the nomination of an unlikely dark horse, and I especially think with the Democratic primary rules (and the recent evisceration of superdelegates), there’d be a nontrivial chance of a contested convention.

  15. WeatherRays says:

    @Stormy Dragon: The attempt to link Tulsi to centuries’ old political problems in India is ignorant at best, malicious bigotry at worst. Tulsi exemplifies and constantly pushes for religious tolerance and mutual respect (aloha) between all people. What political leaders do in India is something they must address in their own democratic system, being the largest democracy in the world. As far as Tulsi’s interactions with Modi, she is on the House Foreign Affairs Committee and has met with Prime Ministers and leaders from all over the world. When she visited India, she also met with leaders of the opposing political party. The facts don’t support some people’s efforts to smear her with innuendo.

  16. aloha says:

    @Stormy Dragon: Tulsi is an advocate for all people to have love and respect for one another, regardless of religion, nationality, gender, sexual orientation or anything else. She was invited to speak at The Reason Rally and also to be they keynote speaker for Muslims for Peace Prophet Mohammed Day. She embodies aloha and that is something this country greatly needs more of in our political system.

  17. Andre Kenji de Sousa says:

    If Assad and Putin could vote on Democratic Primaries she would be certain to win.

  18. Anonne says:

    I’d get behind Tulsi Gabbard. If Tulsi ran, I don’t think Bernie would run because he could be reasonably assured that a strong progressive would be in the mix and he wouldn’t have to run – he’d be out there working for her, though. I also don’t think that Michael Avenatti would run if Tulsi runs, because she’s not a mealy-mouthed politician. Avenatti’s stated interest is in making sure someone with a spine, who can think quickly and isn’t afraid to get a little dirty with negative ads will run against Trump.

    Bloomberg and Biden are probably a lock for running. Everyone below Bloomberg should not bother.

  19. @Kylopod:

    “correlation is not causation” and “post hoc erho propter hoc falacy” are little more than two different expression for saying the same thing.

  20. Kathy says:

    @Miguel Madeira:

    Not quite.

    To take the classic example, you notice that the crime rate goes up in the summer, and that ice cream sales also go up in the summer. That’s correlation. But ice cream sales aren’t driving up crime, nor crime ice cream. So there’s no causation.

    The Post Hoc example is, you wash your car and it rains later the same day, so you conclude it rained because your washed your car. But you don’t keep a detailed log of how often it rains the days you wash your car. So there’s no correlation at all.

  21. Kathy says:


    Thanks for the correction.

  22. Just nutha ignint cracker says:


    Bloomberg and Biden are probably a lock for running. Everyone below Bloomberg should not bother.

    Statements like this are what give me pause about the prospects for the Democratic Party’s near term success.

  23. Kylopod says:

    @Kathy: The two are closely related, of course. I think the main difference is that Post hoc is usually talking about a single event–“A was followed by B, therefore A caused B”–whereas correlation-causation has to do with a sequence of events–“Every time A happens, B happens; therefore, A is the cause of B.”

    It’s even worse than that. People tend to think “Whenever A happens, B often happens; therefore, A is the cause of B”–where the connection does not in fact happen more often than chance would dictate (that’s especially true when they’re analyzing a relatively small number of events). So not only do people confuse correlation with causation, typically what they think is a correlation turns out to be nothing more than a coincidence.

    I see these fallacies popping up all the time when it comes to election analysis. Part of the problem is that there just aren’t that many elections that have happened, especially because people tend to put some arbitrary limit on which range of elections they’re deriving their “rule” from by focusing only on those in “modern” history. For example, throughout 2012 a talking point I kept hearing was that “No president since WWII has won reelection with unemployment as high as it is now.” Cutting it off at WWII was convenient, because FDR won reelection in a landslide when unemployment was in double digits. (Unemployment was also pretty high in 1984, though not quite as high as in 2012.)

    It’s hard to disentangle cause and effect when it comes to election outcomes because there are so many possible things to look at, some of the elements are unprovable (how do you prove someone voted for a particular reason and not another?), and people allow their own assumptions and prejudices to color their interpretation of whatever happened. Was Obama really helped by the drawn-out, contentious primary battle with Hillary Clinton in 2008? I’ve often heard people make that argument, but it seems based on nothing more than the fact that it happened, and he won, and then they construct a narrative about why he supposedly benefited from it–and they’re unwilling to revise their narrative because they think it makes sense to them.

    Similarly, the shock of Trump’s election–which destroyed a lot of previously held notions about how elections are won–has caused many people to construct new narratives that really have no more grounding than the theories they replaced. They typically revolve around the idea that Trump’s particular traits–his brashness, his bullying, his Twitter tantrums–are at least part of why he won, and therefore those are traits worth emulating, to some degree, if Dems wish to beat him. This goes a long way in explaining the behavior and potential presidential bid of Avenatti, as well as Eric Holder’s recent remark “When they go low, we kick ’em.” Some Dems have gotten the idea in their head that the only way to beat Trump is to become as “tough” as he pretends to be–in other words, to engage in crude machismo and stoop to his level of juvenile mudslinging and trolling. People who take this position never seem to consider the possibility that Trump wasn’t actually helped by these traits–that he won the election in spite of how he behaved, not because of it.

  24. al Ameda says:

    Interesting list, my current take …
    Joe Biden …. nope, stick to campaign support
    Bernie Sanders ….. nope, go away
    Elizabeth Warren … see above ‘Sanders’
    Michael Avenatti … see above, ‘Sanders’ and ‘Warren’
    Cory Booker … nope, performance on Judiciary Committee was embarrassing
    Kamala Harris … interested, but wait and see
    Eric Holder … okay but … who knows?
    Michael Bloomberg … okay, I’m interested
    Tom Steyer … I’m tired of Tom
    Eric Garcetti … who knows, need more information
    Julian Castro … know very little about him
    John Hickenlooper … see above ‘Castro’
    Deval Patrick … see above ‘Hickenlooper’
    Kirsten Gillibrand … see above ‘Patrick’
    Amy Klobuchar … definitely interested, strong performance on Judiciary Committee

  25. Kathy says:


    I see these fallacies popping up all the time when it comes to election analysis. Part of the problem is that there just aren’t that many elections that have happened, especially because people tend to put some arbitrary limit on which range of elections they’re deriving their “rule” from by focusing only on those in “modern” history.

    The numbers are problematic, IMO.

    Pre my quick calculation, there have been 57 presidential elections in the US from 1798 (the first presidential election under the current Constitution) until 2016. Is that a large enough number? is it too large? Besides, the rules have changed from time to time. I’m hazy on this, but I think some early elections didn’t even have a popular vote. Not to mention the country changes a lot. Norms change, customs change, the culture changes. There have been elections with more than one candidate per party, there have been third parties, etc.

    What I heard plenty in 2016 is that the candidate with the highest disapproval ratings has never won. By that “rule,” El Cheeto should have lost.

  26. Kylopod says:


    Pre my quick calculation, there have been 57 presidential elections in the US from 1798 (the first presidential election under the current Constitution) until 2016.

    58 elections, and the first was in 1789 (was that a typo?).

    I’m hazy on this, but I think some early elections didn’t even have a popular vote.

    I know not all states used a popular vote initially; they just had the state legislatures choose electors. I’m not sure how many states voted in this way, and for how long. I remember reading an article about the 1824 election, the first one in which the popular-vote winner lost the election, and it was pointed out that even by then there were still two states that did not have a popular vote. So the claim that Jackson won the popular vote that year has an asterisk; he got the most votes of those states where individual citizens got to vote for president. (And putting aside for the moment the fact that only a small minority of the public had the franchise.)

    What I heard plenty in 2016 is that the candidate with the highest disapproval ratings has never won. By that “rule,” El Cheeto should have lost.

    Technically, only sitting presidents get “approval” ratings, what you’re talking about is “favorability” ratings. Gallup has been taking presidential approval ratings since Truman, but the first favorability ratings are more recent (I don’t know exactly when they started, I know there were at least some being taken in the ’70s).

    The most obvious answer to what you describe is that Hillary won the popular vote, and so the “rule” should be that the candidate with the better favorability ratings always wins the popular vote (or the one with the worse favorability ratings always loses the popular vote).

    Unfortunately, that runs into problems too. Even within most of the individual states that were crucial to Trump’s electoral victory, exit polls showed his favorability ratings were worse than hers. For example, in Iowa, which Trump won by 10 points and with an absolute majority of the vote, Trump’s favorability ratings were at 39% positive, 59% negative, whereas Clinton’s were the not quite as awful 41% positive, 57% negative. You find the same thing in other crucial swing states like WI, MI, FL, and so on, even in some hard-red states like Utah.

    What happened? Why did more people vote for him if they had a more negative opinion of him? I think the key to unraveling this paradox lies in another question on the exit polls: voters who gave negative favorability to both candidates voted overwhelmingly for Trump over Clinton. Nationally, this bloc voted for Trump 47-30. In Iowa, it was 45-24. In Wisconsin, it was 60-23.

    In other words, even though Clinton’s favorability ratings were relatively better than Trump’s, Trump was much likelier to be seen as the lesser of the two evils among those voters who disliked both candidates. So Clinton was likelier to be punished by voters who disliked her, and therefore her bad favorability numbers ended up being more devastating, even though they technically weren’t quite as low as his.

  27. Kathy says:


    58 elections, and the first was in 1789 (was that a typo?).

    For some reason, the French Revolution has been recurring a lot in my recent modern history reading.

    What you say about Hillary, though, would indicate she was a far worse candidate than Trump (which really boggles the mind). IMO, there are two factors at play. One, Hillary was more or less a known quantity, and people who dislike her have been disliking her since 1991. Two, El Cheeto wasn’t known, not politically, and people expected him to “pivot” just like every other candidate has done in the past.

    Another factor, I think, is the assumption on the part of most people, even Trump’s supporters, that a Clinton win was unavoidable. This might have had the effect of keeping a few Clinton voters home, or allowed them to vote for Stein or Johnson, who had as much chance of winning in the EC as I did.

    One thing I wonder, is how much aversion there is to voting for the wife of a former president for president. It seems almost like nepotism. Granted, of the recent candidates to have won their party’s nomination for president, Clinton was the most qualified. She had legislative experience, executive experience, foreign policy experience, and has been a policy wonk for a very long time.

    But had her husband not been president, how far would she have gotten? No way to tell, especially as I know little of her life pre-1992

  28. Kylopod says:


    What you say about Hillary, though, would indicate she was a far worse candidate than Trump (which really boggles the mind).

    Whether she was a “worse” candidate is, of course, subjective to some degree, unlike the empirical question of how popular or unpopular she was.

    Speaking for myself, I always felt she was a weak candidate. I held this belief going all the way back to 2000, when she first ran for Senate. She struck me as not having a natural presence on stage; she came off artificial, stilted, almost robotic. Next to her husband, who had plenty of weaknesses but projected warmth and charm, this made her seem the lesser of the two. So it was the worst of both worlds, since people tended to associate her–unfairly–with her husband’s moral lapses, but found her less likeable, with a good dollop of sexism on top.

    In 2008, I remember thinking that if she became the Democratic nominee, the Democrats would lose. In retrospect, I was probably wrong about that; after the collapse of the banks pretty much any Democrat would have defeated any Republican. (Well, okay, maybe John Edwards, whose sex scandal broke in August of that year, would have blown it.) My views on elections have changed since then, as I’ve learned a lot more about the poli-sci research on the subject, suggesting economic fundamentals are more important in determining an election’s outcome than the identity of the candidates. If it had been Hillary vs. Trump in 2008, I think she’d have whipped his butt in the election. In 2016, she already had the task of winning a third straight term for her party. Most of the poli-sci models suggested that based on the fundamentals (as opposed to the candidates), the election was either a toss-up or lean-Republican. Given that Trump lost the popular vote, a case could be made that he underperformed relative to the fundamentals.

    Even with my long-time reservations about Hillary, it may surprise some people to learn that she was never really unpopular until 2015. She was always polarizing–there was always (ever since her days as First Lady) a significant segment of the country that strongly despised her. But that segment was never in the majority–until late 2015. As Sec. of State, her favorability ratings were astronomically high, and even the first time she ran for president, back in 2008, her numbers tended to be around 55% positive, 45% negative–hardly ideal but still above-water. If you look at a chart of her favorability ratings from when she was First Lady, with the exception of one very brief period in 1996, consistently more Americans had a positive than negative view of her.

    That’s why a lot of these theories about how terrible a candidate she was seem to be missing something, in my view. Ever since her days as First Lady she was the subject of one quasi-scandal after the next and a vicious hate campaign, but none of that made her toxically unpopular–not Whitewater, not Vince Foster, not Benghazi. Until 2015, with emailgate and Clinton Foundation-gate. That’s when these smears finally, at long last, penetrated the right-wing bubble and started to seep into the general population. Personally, I think that had a lot more to do with the Russian infiltration of social media than with Hillary’s particular weaknesses.

  29. Kathy says:


    I didn’t see her campaign in 2000. I thought at the time her run for the Senate was a cynical ploy to stay in the public eye, had no desire to represent NY, and would dump her seat and run for president at the first available opportunity. I can’t tell how right I am, except she ran twice, eventually leaving her seat and taking a cabinet post under Obama.

    I did see her campaign in 2016 on TV. I don’t think she could generate much enthusiasm, past that which any candidate of a major party would. and of course among her base.

    I’ve noticed getting a third term for the party in the White House, at least in recent times, is rather hard. But, as you point out, what constitutes “recent”? It’s hard to say, too, given the split between popular and electoral votes.

    Bush the elder won a third term for the GOP in 88, against a rather weak opponent. Gore won the popular vote, though just barely, while losing the electoral vote as barely. Clinton won the popular vote by a fair amount, while losing the electoral vote. In just about any other country in the world, Gore and Clinton would have won (and it would have been Bush the younger clamoring for recounts).

    The economy has a lot to do with reelection, but less so with the election of a new candidate. Bush the elder was elected in good times, but Gore and Clinton lost in good times as well.

    So it’s hard to say.

  30. Ratufa says:

    @Scott G:

    Tulsi seems like a sincere, nice person. But, I’d take her impressions of what Syrians really want want with a grain of salt. An American politician travelling in a repressive country, particularly one undergoing internal and external conflict, will be monitored. Access will be controlled. Anyone who talks with them should (and will if they value their lives and their family’s lives) assume that their discussion is monitored, and speak accordingly. A dog & pony show of people praising the government is expected. People who talk to her without government permission are taking a major personal risk.

    That’s not to say that she’s wrong about whether or not we should be trying to remove Assad from power. But, a realistic evaluation of the pros and cons of that policy is different from the one-side view of the Syrian government that she seems to have bought into.

  31. Kylopod says:


    I’ve noticed getting a third term for the party in the White House, at least in recent times, is rather hard. But, as you point out, what constitutes “recent”?

    It’s definitely tricky, for all the reasons we discussed. For one thing, before WWII it was common for a party to maintain control of the White House for longer than 8 years. The Dems did it for 20 years with FDR and Truman. Before that, the GOP did it for 12 years (1921-1933), 16 years (1897-1913), and 24 years (1861-1885).

    Even when you look at the post-WWII period, the “rule” is not as obvious as it seems. 1960, 1968, 1976, 2000 and 2016 were basically statistical ties, and in the last two cases the candidate running for their party’s third term actually won the popular vote. To put it another way: in 3 of the last 4 elections in which a candidate was running for their party’s third straight term in the White House, that candidate won the most votes. Doesn’t sound as formidable when put that way, huh?

    Nevertheless, for various reasons political scientists do generally think it is harder for a party to win a third straight term than a second. Alan Abramowitz’s model (which is one of the best regarded) is partly based on that theory: it is called the “time for change” model. In 2016, his model suggested that a generic Republican would have an advantage over a generic Democrat. But Abramowitz publicly admitted he believed his model would fail that year, due to Trump’s weaknesses as a candidate. In a sense, he was right: his model predicted that the Republican would win the popular vote, something Trump did not do.


    Of course one can make a case that it was the Comey Letter that cost Clinton the election, but political scientists base their models on ignoring the impact of “events”–in fact ignoring the candidates entirely–and focusing only on broad fundamentals: how long the incumbent party has been in power, how well the economy is doing, whether the country is at war, and so on.

    The economy has a lot to do with reelection, but less so with the election of a new candidate. Bush the elder was elected in good times, but Gore and Clinton lost in good times as well.

    The poli-sci models generally focus not on the state of the economy in an absolute sense, but the direction of the economy. Abramowitz’s model was bullish on the GOP’s chances in 2016 not just because it was a non-incumbent election but also because GDP growth had been slowing down for over a year.

    The 2000 election was a different matter. Most poli-sci models–including Abramowitz’s–projected that Gore would win the popular vote by a comfortable margin. That was one of the biggest misses for poli-sci election models–though technically 2016 was the first time the Abramowitz model (since its inception in 1988) incorrectly predicted the winner of the popular vote.

  32. Kathy says:


    Of course one can make a case that it was the Comey Letter that cost Clinton the election

    I’ve mentioned air crash investigations before. One fundamental aspects of these investigations is the idea that there’s never one single cause for an accident; and even when there is just one thing, you’ll find a lot of other things behind it.

    Many other aspects of life are like that, but people tend to focus on the one obvious thing.

    Maybe the Comey letter did cost Clinton the election. But only because her campaign had other failings that led to a close race that shouldn’t have been close, and because some people won’t ever vote for a woman even if she’s running against a hybrid of Hitler and Stalin, and because some people won’t vote for the other party regardless of any considerations, and plenty else.

  33. Kylopod says:


    But only because her campaign had other failings that led to a close race that shouldn’t have been close

    That’s the thing, though: the poli-sci models–which ignore the candidates entirely–absolutely did predict a close race. One of the political scientists who creates these models (Abramowitz) refused to believe it would actually happen that way, given Trump’s unique awfulness. But it did. We can talk all day about all the things Clinton did wrong, but in the end the result was almost exactly what the models predicted, and if anything Clinton did slightly better than some of those models predicted (as mentioned, it’s the first election where Abramowitz’s model incorrectly predicted the popular-vote winner).

    This is not the first time we’ve seen that kind of weird convergence. The 1988 election outcome also was almost exactly what the poli-sci models predicted. Yet the common narrative about that election is that it was all based on the trainwreck of Dukakis’s campaign, from wearing a funny hat to awkwardly answering a debate question to being assailed by the Willie Horton matter–the sorts of things the poli-sci models paid no attention to.