Why Trump Won, A Reminder

So we’ve got a lot of people acting as though the election of Donald J. Trump represents a seismic shift in the American electorate and–just maybe!–a prelude to the Fourth Reich. This is just a reminder that the data don’t bear that out.

It’s not exactly news–Slate Star Codex carried it right after the election–but it bears repeating. So here’s the update from Matt Bruenig: The Boring Story of the 2016 Election.

Donald Trump did not win because of a surge of white support. Indeed he got less white support than Romney got in 2012. Nor did Trump win because he got a surge from other race+gender groups. The exit polls show him doing slightly better with black men, black women, and latino women than Romney did, but basically he just hovered around Romney’s numbers with every race+gender group, doing slightly worse than Romney overall.

However, support for Hillary was way below Obama’s 2012 levels, with defectors turning to a third party. Clinton did worse with every single race+gender combo except white women, where she improved Obama’s outcome by a single point. Clinton did not lose all this support to Donald. She lost it into the abyss. Voters didn’t like her but they weren’t wooed by Trump.

Bruenig goes on to explain why this narrative is so underplayed. Which is: nobody likes it. I’ll let you read Bruenig’s analysis on this point (I basically agree with it), but here’s the point: any discussion of what’s happening in American politics should adhere to the basic facts. Trump did a little worse than Romney. Clinton did a lot worse than Obama. Ergo the defining factor was Clinton’s deep unpopularity.

Last thought: she’s been in the press a lot since the end of the election. It even sounds like she wants to try again. Will she try? Will she succeed (in getting on the ballot)? What will that look like?

White women and Trump.

Photo from “9 Women on Why They’re (Still) Voting for Trump,” New York Magazine

53% of white women voted for Trump.

In the circles I run in, there was tons of coverage and discussion about the myriad comments Trump has made over the years that many of us consider blatantly sexist. When the Hollywood Access tape came out, I took (and still take) his comments as an admission of sexual predation, a topic that means a great deal to me. I was already a #NeverTrump conservative, but the Hollywood Access tapes made it much more difficult for me to understand how people of good conscience, especially women, could vote for this man.

My feed started to include articles such as The Atlantic’s “The Revolt of the Conservative Woman” and viral tweets from conservative women feeling betrayed by their party’s defense of Trump. Between his apparent gross disrespect of women and the opportunity to elect the first female president, I thought women would vote in droves for Clinton and against Trump. Article’s like FiveThirtyEight’s “Women are Defeating Donald Trump” seemed to think so too.

But clearly I was missing some major parts of the puzzle. (Apparently a lot of us were, including the pollsters.) As Walker pointed out recently, Trump’s support among (all, not just white) women was only slightly lower than the average for Republican presidential candidates since 2000 (42% compared to an average of 44.2%). Clinton’s support among women was exactly average for the Democratic presidential candidates since 2000 (54%). Women weren’t driven to the polls to vote against Trump or for Clinton—overall turnout among women was only 1% higher than in 2012.

So what happened? What pieces of the puzzle was I missing, that women were neither particularly repelled by Trump nor particularly inspired by Clinton?


Maybe it was bigotry.

Predictably, some of my leftist friends think the missing puzzle pieces are racism and (internalized) sexism. I’ve seen repostings of LV Anderson’s piece at Slate (“White Women Sold Out the Sisterhood and the World by Voting for Trump”), which is filled with explanations like this:

What leads a woman to vote for a man who has made it very clear that he believes she is subhuman? Self-loathing. Hypocrisy. And, of course, a racist view of the world that privileges white supremacy over every other issue.

Sarah Ruiz-Grossman at Huffington Post authored a letter to white women that started with “Fellow white women, I’m done with you.” In sync with a lot of the commentary I’ve read, it showed no curiosity as to the perspectives, hopes, fears, or values of millions of women that led them to vote for Trump (or at least not vote for Clinton). Instead, and again, it simply told them what they didn’t care about, what their moral failings were, and what they must do now.

While I appreciate the frustration, I think this approach is an awful strategy. Lambasting people, especially conservatives, for bigotry has not been terribly effective at changing their minds (or votes). Berating the other side seems to mostly get them to tune out entirely when the inevitable accusations of prejudice begin. And the rampant shaming of Trump supporters clearly did nothing to dissuade them throughout the primaries (when shaming was coming from conservatives and liberals alike) or the rest of the election. Why would it work now, when they’ve won? They have less reason than ever to be concerned about the opinions of people who show no understanding of their perspectives or interest in their wellbeing.

But it’s not just that I think accusing people of bigotry is poor strategy; I think it’s poor reasoning too. In this post I go through three theories of how voting for Trump was bigoted and explain whether I think those theories make sense.


Theory 1 – Internalized sexists: women voted for Trump instead of Clinton because they are sexist against female candidates.

How Trump measured against Clinton is a major factor. The pantsuit nation adored Clinton, of course, so for them this was no contest at all. But we aren’t looking into what HRC’s biggest fans thought; we’re exploring the millions of women who disagreed.

It’s not that everyone who voted for Trump thought he was wonderful: exit polls show that 20% of people who voted for Trump had an overall unfavorable opinion of him. Nearly a quarter of Trump voters said he wasn’t qualified for or did not have the temperament to be president, and a full 17% of people who voted for Trump to be President said they would be “concerned” if he were elected!

But 28% of Trump voters said they chose him mainly because they disliked Clinton. Trump received about 60M votes, which would mean about 17M cast their votes primarily as a vote against Clinton. Along the same lines, while voter turnout for Trump was slightly lower than it had been for Romney, voter turnout for Clinton was much lower than it had been for Obama.

Some will argue that these numbers show sexism: people so rejected the idea of a female leader they either stayed home or voted for someone they despised just to stop Clinton. Actually women get accused of sexism no matter which way they vote: Women who backed Clinton are accused of bias, just “voting with their vaginas,” and the rest of us are accused of not voting for her because we’re misogynists. It’s a lose-lose.

But these theories ignore the fact that women don’t generally vote based on gender, and gender stereotypes end up being less relevant than party affiliation in voting decisions. In other words, we vote based on political positions. The reality is that most of the women voting for or against Clinton did so based on a variety of competing concerns and priorities, just as most men choose their candidates.


CNN reported that millennial women in particular “rejected the notion that gender should be a factor in their vote.” As FiveThirtyEight put it:

Clinton’s stunning loss Tuesday night showed that issues of culture and class mattered more to many American women than their gender. The sisterhood, as real sisterhood tends to be, turned out to be riddled with complications.

On average, for the last 5 presidential elections, 89% of Democrats chose the Democratic nominee and 91.4% of Republicans chose the Republican. Last week 89% of Democrats chose Clinton and 90% of Republicans chose Trump. If internalized sexism were a major factor in terms of female nominees, we’d expect 2016 to show a drop in Democrats voting for the Democrat (as internally sexist Democratic women abandoned Clinton) and perhaps even a jump in Republicans voting for the Republican (as internally sexist Republican women were motivated to stop Clinton). But there was no such change.

Similarly, if internalized sexism was a major factor we’d expect Clinton to get a lower proportion of women’s votes compared to previous Democratic nominees. Yet, as mentioned above, she got exactly the average proportion of women Democratic nominees have had in the last five presidential races. Or, if we’re operating under the idea that only conservatives can be bigots, we’d at least expect a higher proportion of women to vote for Trump in order to stop Clinton. Yet Trump got just slightly less than the average proportion of women Republican nominees have had for the last five races.

If anything, these stats suggest women weren’t influenced by gender at all.


Theory 2 – Indifference to sexism: women cared more about party lines than taking a stand against Trump’s misogyny.

There are several assumptions embedded in this line of thinking: (A) The women who voted for Trump accessed the same information we did about him. (B) When they assessed that information, they came to the same conclusions we did about the degree of Trump’s misogyny. (C) There was nothing else in the balance for them in this election that could have meant more to them than Trump’s misogyny.

2a. Trump voters were likely accessing different information.
Hopefully it’s not a secret that conservatives and liberals consume different media. I wish I had time to do an entire blog post on how drastically this impacts our views of each other and of our political landscape. But the main point is we should be very careful when assuming that everyone else—especially people that run in different social circles and already hold different perspectives—“knows” the same “truths” we know. Which stories get reported and how they’re described varies a lot, and sadly, at least in my experience, most people don’t look for sources from worldviews they don’t hold. Or, if they do, it’s not in an attempt to observe and understand, but to feel outraged and argue.

So when John Oliver does a witty, biting piece on “making Donald Drumpf again” and you see it reposted over and over, that doesn’t mean everyone saw it. The people who already hated Trump were a lot more likely to see it than anyone else. Late night comedy is, after all, a bastion of liberal derision.

Pictured: Echo Chamber with John Oliver
Pictured: Echo Chamber with John Oliver

2b. Trump voters were likely interpreting information in different ways.
That’s not to suggest Trump supporters were wholly unaware of criticisms against him. I think it’s unlikely, for example, that many Trump supporters didn’t at least hear about the Hollywood Access recording. But the context in which conservatives in general, and enthusiastic Trump supporters specifically, interpreted that was often quite different than how leftists saw it.

Many people (including me) were disgusted and horrified by Trump laughingly talking about getting away with kissing and groping women without their consent. But many others mostly heard politically-motivated faux outrage. The same people so focused on Trump’s comments and the sexual assault allegations against him remained dismissive or defensive about the long history of sexual misconduct and assault allegations against Bill Clinton—and Hillary Clinton’s role in silencing Bill’s accusers. Clinton fans retorted that Hillary isn’t responsible for Bill’s behavior, but that misses the point. She’s responsible for her behavior: she referred to these women as “floozy,” “bimbo,” and “stalker,” and put great effort into “destroying” their stories.

Yeah, okay.
Yeah, okay.

Good luck telling conservatives they must take a principled stand against sexual assault while refusing to acknowledge that the Clintons basically embodied rape culture.


Of course the Hollywood Access tapes are only one example of Trump’s sexism, but the pattern remains the same. Whatever example you point to, if outraged accusations of bigotry are coming from leftists or the media, conservatives are extremely skeptical. In fact, getting back to point 2a, conservative circles are more likely to have articles about people who made up stories of hate crimes – stories which, before being shown to be false, often caused viral online outrage (as well as extensive donations to the alleged victim).

I find this problem very frustrating. I do believe the left is too quick to claim bigotry, but I believe the right is therefore too quick to dismiss actual bigotry. Robby Soave of Reason.com summarized this view well:

[It’s] the boy-who-cried-wolf situation. I was happy to see a few liberals, like Bill Maher, owning up to it. Maher admitted during a recent show that he was wrong to treat George Bush, Mitt Romney, and John McCain like they were apocalyptic threats to the nation: it robbed him of the ability to treat Trump more seriously. The left said McCain was a racist supported by racists, it said Romney was a racist supported by racists, but when an actually racist Republican came along—and racists cheered him—it had lost its ability to credibly make that accusation.

Kirsten Powers explained the same idea in terms of misogyny:

After all, these same voters have watched as every Republican candidate in recent memory has been accused of waging a “War on Women.” If Democrats are going to claim that Mitt Romney and John McCain hate women (and they did), then they shouldn’t be surprised when voters ignore them when they say Donald Trump hates women. If every Republican is a misogynist, then no Republican is.

I don’t believe the right’s resistance to recognizing bigotry is all the left’s fault. I think that’s a factor, but ultimately we’re all responsible for assessing each situation and trying to be fair-minded about it.

Even so, I think many conservatives viewed the outrage over Trump as nothing more than yet another chapter in a long history of selective and manufactured leftist outrage, and so they discounted it. So even if they had watched John Oliver, they probably would have just rolled their eyes at another leftist show mocking conservatives again.

2c. Trump voters were weighing a lot of additional concerns apart from bigotry.
But there were a lot of conservatives who heard about the problems with Trump and were seriously concerned. Many of them became the #NeverTrump crowd, but others still voted for Trump. Why? Because they weren’t balancing the problems of racism and sexism against nothing. They were taking those issues and factoring them in with a lot of other issues, weighing each one, and coming to a decision. Even women who voted against Trump had other concerns they considered more important than sexism.

Many reject as ridiculous this concept of weighing multiple factors, saying it’s a weak excuse to try to cover up bigotry. They assert nothing could outweigh the civil rights threats Trump represents, and therefore the people who came down on Trump’s side, by definition, just didn’t care enough about civil rights.

Keep lecturing people about what they don’t care about while showing no understanding of what they do care about. That’s been so effective so far.

Interestingly, I saw the same reductive thinking from conservatives trying to berate #NeverTrump people into voting for him. If you didn’t vote for Trump—if you voted for Clinton, or even if you voted third party—you must not care about massive government abuse and corruption, our country’s impending economic collapse under an overregulated welfare state, and, possibly above all, the killing of tens of thousands of babies.

Does that last part sound hyperbolic to you? Because, for a huge portion of the pro-life movement, that was the assertion. Many pro-lifers view abortion as morally equivalent to any other unlawful human death. If you want to imagine the abortion debate from a pro-life perspective, just replace the concept of “fetus” with “toddler,” and listen to how the arguments sound. So when Hillary Clinton campaigned on a platform of no restrictions through all three trimesters and requiring Medicaid to cover abortions, that was an absolute deal breaker for many people. Abortion happens in this country roughly 1 million times a year. Imagine for a moment you were choosing between (1) a candidate who stirs racial animosity and blatantly disrespects women and (2) a candidate who unapologetically embraces policies making it legal to murder a million toddlers a year. Who would you pick?

If your first response is to explain why that second description is false, you’re missing the point. Yes, I understand that for many, abortion is nothing at all like killing a toddler and even the comparison is offensive. I’m not trying to convince anyone here how to feel about abortion. I’m trying to convince people that you can’t sincerely talk about what motivates others if you refuse to acknowledge their actual perspectives. People who voted for Trump could (a) recognize Trump’s racism and sexism, (b) care greatly about those issues, and (c) still believe the threats Clinton represented were more dire. The only way you can genuinely believe that every single vote for Trump represented at minimum a callous disregard for civil rights is if you ignore or dismiss the circumstances and value systems of millions of people.


Passion about abortion likely affected many of women who voted for Trump. LV Anderson was aghast that more than half of white women would vote for a man who said he’d appoint Supreme Court justices to overturn Roe v. Wade. It is amazing to me that so many people are still taken off guard when women are antiabortion. Half of American women are against abortion, and that has been true since long before Trump entered the primaries. Yet each time thousands or millions of women don’t go for the pro-choice position, pro-choice people are just so surprised. This is another example of the same pattern: be totally unaware of what people have repeatedly said they care about, and then be surprised and angry when they vote for the things they said they cared about all along.

I’ve used abortion as an example of competing values, but it’s only one of many. A recent New York Times article profiled women who voted for Trump. While 22-year-old Nicole Been mentioned her deep opposition to abortion as part of her stance, other women discussed Trump’s approach to veterans, their own dire financial situations, and their disillusionment with Democratic efforts to improve their lives. Another New York Times article profiled college-educated women who voted for Trump; they, too, opposed abortion, but focused more on economic security and job and college prospects for their children.

Article after article about the parts of the nation that went wholeheartedly for Trump (including many counties that had previously voted for Obama twice) describe recurring themes of economic and cultural despair, resentment at being derided by the rest of us, and the way economic and racial anxiety intertwine. (Here’s one or two more.)

Note that racial anxiety is one of the recurring themes. The left seems to want to reduce this narrative to bigotry and nothing more, and I’ve spent a lot of time here explaining why I think that’s inaccurate. But the right seems to want to reflexively deny bigotry had any part to play, and I don’t think that’s true either. At minimum there was certainly a racial component to Trump’s candidacy. Looming large in the support of Trump were concerns about minority groups getting unfair preferential treatment and resources, immigrants taking resources and increasing criminal activity, and terrorists threatening our safety.


Theory 3 – Institutionalized sexism and racism – regardless of personal motivation, women who voted for Trump supported a platform that would disproportionately harm minority groups.

A major hurdle with discussions of racism and sexism is the use of the same words to mean very different things. In my right-leaning circles, “racism” generally means an individual’s disdain or animosity towards others based on race. Same thing with “sexism,” but based on sex. In my left-leaning circles, “racism” and “sexism” often mean individual disdain or animosity, but can also mean cultural norms and systemic and institutionalized systems that disproportionately negatively impact minority groups.

So when someone claims that a vote for Trump was racist, they could either mean (a) the person casting the vote has disdain or animosity toward people of other races or (b) the person casting the vote, regardless of his or her motivations, helped to uphold systems that have major negative impacts on women and nonwhite people.

The interesting thing about the “effects not intentions” version of racism is that it can be empirically verified. Motivations can be pretty complicated, multifaceted, and irrational. Effects can be objectively measured. So if “racism” (or sexism or Islamophobia or homophobia) is defined as “policies and practices that hurt these groups,” and if electing Trump ends up hurting these groups, then it follows that electing Trump was racism, by this definition.

3a. It’s reasonable to believe electing Trump will end up hurting these groups.
Trump campaigned on ending sanctuary cities, suspending visas, and deportation. If implemented, those policies would disproportionately affect undocumented immigrants (about half of which are Hispanic or Latino, followed by Asian) as well as American citizens from families with mixed citizenship statuses. Whether you agree with these policies or not, and whether you personally care how these policies affect others or not, it’s hard to deny that they will negatively impact immigrants and their family members who are American citizens.

Trump has talked about requiring immigrants from predominantly Muslim countries to register upon entering the U.S.  Alternatively he called for a ban on Muslims immigrating to the country; while he clarified this would not apply to American citizens, it’s hard to believe such policies and related rhetoric about Muslims won’t affect public perception of and reaction to Muslims already living here. The FBI has released data showing a 67% increase in hate crimes against Muslims in 2015. Some are worried this trend is related to Trump’s rise and campaign rhetoric; others say there are other factors–such as 2015 terrorist attacks–that likely played a part. But I think either of those reasons underscores the central point: when the public increasingly perceives a group as dangerous, violence against innocent people in that group becomes more likely.

It seems like Trump’s potential effects on African American communities have been less of a focus, but there are reasons for concern. (This link includes some reasons I think conservatives will dismiss as more faux outrage, but, for what it’s worth, I believe some of these points are pretty valid.)

While Trump didn’t propose specific policies against LGBT folk, the 2016 Republican Party platform did object to legalized gay marriage and take other positions seen as anti-gay. Because Trump was the Republican nominee and can now nominate SCOTUS judges, many believe he will work to adopt those Republican positions. I think it’s unlikely gay marriage will get overturned, but I don’t think it’s a certainty, and I see why many people are worried their marital status could be threatened.

Trump has a history that suggests a pretty disrespectful view of women, not to mention (again) his statements in the Access Hollywood recordings. To the extent women believe support for Trump signals societal dismissal of sexual assault, that belief could have another chilling effect on women reporting assaults and seeking help. I watched this play out on both the national level and with women I know personally after the Access Hollywood firestorm. Women (and men, for that matter) who have experienced sexual assault listened as friends and family who were Trump supporters minimized, dismissed, and, in my opinion, very generously interpreted Trump’s statements. That was difficult. Victims of sexual assault hear those reactions and believe the reactions would be the same if they came forward with their own stories. I can understand why people fear this kind of dismissal of sexual misconduct will only get worse now that Trump will be president.

A vote for Trump lent support to these policy proposals and attitudes, even if the person voting didn’t personally support one or any of the above. In this sense I think Theory 3 is truer than the other theories—I think a Trump administration will very likely make life harder for these groups.

3b. First problem: negative effects count as racism regardless of what they’re being weighed against.
Consider Trump’s campaign regarding Islamic terrorism. I do believe requiring (mostly) Muslim immigrants to register upon entering the country, or refusing to let them enter at all, will negatively affect the public’s views and behavior toward Muslim Americans and Muslim immigrants already here.

But I also recognize that the people who support these measures believe they will significantly increase our national security and safety. Based on my understanding of Theory 3, what Trump voters believe about these measures (and how those beliefs speak to their motivations) is irrelevant, because Theory 3 is all about effects on minority groups, not the intentions of the people pushing these policies. Whether they sincerely believe these measures will save American lives doesn’t change whether or not this approach is defined as racist.

And we’ve only talked about what they believe, not what is objectively true. Apparently NSEERS, the similar Bush-era program that required immigrant registration, was ineffective at preventing terrorism; it sounds like it was just more security theater, but in this case directed at specific groups. But suppose, hypothetically, immigrant registration made a huge difference in national security. Suppose—as I suspect is the belief of some who support this idea—that without immigrant registration we’d have more San Bernadino and Pulse nightclub shootings. Or another World Trade Center.

If these policies actually prevented terrorism deaths in our own country, does that change whether they are racist? If I understand Theory 3 correctly, it does not. In this way Theory 3 rings a bit hollow for me, because while it is at least technically accurate and objectively measurable (does X policy negatively impact Y community or not?), if it ignores all other factors I still consider it misleading.

3c. Second problem: conflating Theory 3 with Theories 1 & 2.
In my experience, the left frequently blurs the line between “negative impacts” and “personal animosity.” A great example is the Slate article “There’s No Such Thing as a Good Trump Supporter.” Chief political correspondent Jamelle Bouie argues that Trump supporters do not merit empathy because they “voted for a racist who promised racist outcomes.” He cites other authors who have claimed Trump’s victory does not reveal an “inherent malice” in the populace (referring to the “personal animosity” definition of racism). Bouie counters with the “negative impacts” definition:

Whether Trump’s election reveals an “inherent malice” in his voters is irrelevant. What is relevant are the practical outcomes of a Trump presidency…If you voted for Trump, you voted for this, regardless of what you believe about the groups in question.

But, as the title of the piece suggests, Bouie is not condemning only the effects of voting for Trump; he’s condemning the Trump voters themselves. He asserts that it is myopic and even “morally grotesque” to suggest Trump supporters are good people. He compares Trump voters to the men in the early 20th century who organized lynchings (they “weren’t ghouls or monsters. They were ordinary.”) and the people who gawked and smiled at those lynchings (“the very model of decent, law-abiding Americana.”) He sums up: “Hate and racism have always been the province of ‘good people.’”

Note the switch here. Bouie is no longer talking about practical outcomes; he’s talking about hate. He has switched from the “negative impacts” definition of racism back to the “personal animosity” definition. So is he saying that most Trump supporters did not have inherent malice but should be condemned for the policies they supported? Or is he saying that anyone who can support Trump has to be, at least in part, motivated by hate?

And this is often how I see the conversation going. To (heavily) paraphrase:

Person A: If you voted for Trump, you’re racist.
Person B: I’m really not. I disagreed with a lot of what he said but I thought Clinton would do more damage in XYZ ways.
Person A: Yeah, you may not personally feel racist but you supported racist policies. It just shows you think the concerns of white people are more important than the actual human rights of everyone else.
Person B: That’s not what I think at all!
Person A: It’s not about what you personally think! It’s about what you supported!

And repeat.

In other words, in principle motivation is supposed to be irrelevant because racism is about effects, but in practice accusations of racism nearly always boil down to motivation—at best a selfish indifference and at worst outright malice. So, in principle, I think Theory 3 has some merit and is worth talking about. In practice, I find I just end up repeating the arguments I made for Theories 1 & 2.


Clinton “winning” the popular vote doesn’t tell us much.

According to Snopes, as of yesterday (November 13) Clinton is currently ahead of Trump by about 630,000 votes, and it looks likely that number will increase as the ballots continue to be counted. This situation has lead to familiar calls to end the electoral college, and already I’ve seen plenty of angry posts from HRC supporters feeling they’ve been robbed. So a few points here:

1. At least as of now, statistically Trump and Clinton are actually tied.

So far, about 121M people voted for either Trump or Clinton. Previous research has suggested the error rate for uncounted votes in a presidential election is roughly between 1.1% – 2.0% (here’s the study and its tables). If those numbers hold, one of the candidates would need to beat the other by somewhere between 1.3M – 2.4M votes for it to be statistically significant. At least as of now, Clinton falls short of that, meaning the difference between them is in the margin of error. They’re tied.

2. Even if Clinton ends up above the margin of error, it doesn’t mean the electoral college has robbed anyone.

Politicians campaign and people vote with the electoral college in mind. Think of how many times you heard people discuss how everyone in swing states ought to vote, whereas if you were in a solid red or blue state, people didn’t care as much. Why? Because they knew how the electoral college works. Party hardliners were relatively forgiving of third party votes in the states already decided. I live in California, and no one seemed very worried about third party votes here. We all assumed (correctly) that Clinton would get all of California’s electoral votes either way. But if you went third party in Florida this year? Apparently the end of the world might be a bit your fault.

If the electoral college hadn’t existed, you don’t know how people would’ve voted differently. How many more conservatives would have bothered to vote in California? How many more liberals would have showed up in Alabama?

The same goes for the way the politicians campaign. They choose which states to pour their ads into, to get people on the ground, to hit hard, based on their knowledge of the electoral college–and those processes greatly influence people’s votes.

If the electoral college didn’t exist, campaigning and voting would have looked different, and we don’t know how different. We don’t know how the landscape would have changed, and so we don’t know which candidate would have received the “mandate” from the American people.


So it’s possible HRC will end up above the margin of error. And it’s possible if she does that would happen to represent what Americans would have wanted if they had voting absent an electoral college. It’s also possible it would have looked completely different. Especially when considering how unprecedented this election was and how surprised so many of us were by the results, it’s pretty hard to predict. It’s certainly not clear that the electoral college foiled the (overall) will of the people.

How Atypical Was This Election?

Image result for donald trump speech

Donald J. Trump is President Elect of the United States of America.

The predictions were off. Way off (The polls themselves may not have been.)*****

Numerous commentators–from the professional to the everyday Facebook warrior–have put forth various explanations based on the exit polls, including change-seeking voters, gender gaps in the voting blocs, the white working class, wealthy whites, anti-Clinton voters, anti-PC voters, sexist voters, racist voters, etc.

There’s probably a grain of truth in all of these explanations (some more than others). But if we just look at the general numbers, what can we say for sure? Despite early claims of high voter turnout, that just wasn’t the case.**

2016 Election
Trump Clinton Total Turnout
59,698,506 59,926,386 119,624,892
2012 Election
Romney Obama
60,933,504 65,915,795 126,849,299
2008 Election
McCain Obama
59,948,323 69,498,516 129,446,839
2004 Election
Bush Kerry
62,040,610 59,028,444 121,069,054

Not only was voter turnout low in general, but even Republican turnout was slightly lower* than in past years. It was significantly lower for Democrats.

Given the racially-charged rhetoric and sexual misconduct this election season, how did Trump do among minorities compared to other Republican presidential candidates?

% Trump Romney McCain Bush 04 Bush 00 Average
Whites 58 59 55 58 55 57
Blacks 8 6 4 11 9 7.6
Hispanics 29 27 31 44 35 33.2
Asians 29 26 35 43 41 34.8
Other 37 38 31 40  No data 36.5
Women 42 44 43 48 44 44.2

A few things that jump out at me:

  • Trump had higher support among Hispanics than Romney, though both fell woefully short of McCain and Bush. Nonetheless, the bump in Hispanic support is surprising.
  • Trump had higher support among blacks than both Romney and McCain and ended up above average for the last 5 elections.
  • Trump had higher support among Asians than Romney, though still short of McCain and especially Bush.
  • Trump had the lowest support of women of the last five elections (can’t imagine why…). Nonetheless, it’s not drastically lower than most of the other candidates, which may astonish some.
  • Trump actually had lower support among whites than Romney and ended up only slightly above average when considering the last 5 elections. In other words, the white support for Republicans has been pretty consistent.

There is much more that could and should be considered when parsing the numbers. Groups aren’t monolithic and factors like income, education, etc. are necessary in order to paint a more accurate picture. However, pundits pointing out that, say, white people–rich/poor, urban/rural, college/no college, whatever–supported the Republican nominee is hardly a shocker. As conservative writer David French put it, “White voters responded mainly by voting in the same or lesser numbers as the last three presidential elections. That’s not a “whitelash,” it’s consistency.”**** It increasingly looks like what some outlets have been saying is correct: Democrats stayed home.*** While further analysis may tell a different story, a lot of this looks like pretty normal partisanship. One party just didn’t show up.**

*Update (11/14): Trump currently sits at 60,265,858, while Clinton is still ahead with 60,839,922. This puts Trump’s turnout slightly above McCain’s numbers, while Clinton falls woefully short of Obama. Some are projecting that overall turnout may actually surpass 2012, though others are claiming a 20-year low. We won’t know for several more weeks. Check out Monica’s post today on the popular vote.

**Update (11/15): From The Washington Post: “It’s a very odd result. Turnout up slightly in terms of raw numbers, but down as a percentage of those eligible. A likely drop in votes for the Democrat and a spike in votes for third party candidates, with the Republican holding steady. More votes for the Democrat, but the Republican becoming president.” Looks like my claims about low voter turnout were incorrect, but my main point about low turnout among Democrats seems to be holding up.

***Update (11/18): Looks like the case for low Democrat turnout is getting stronger and stronger.

****Update (12/02): Trump didn’t flip the Rust Belt. Democrats lost it.

*****Update (12/21): Vox has an excellent write-up now that the votes have been counted: “Voter turnout in 2016 was actually closer to 58.9 percent, slightly higher than 2012, according to data from the US Elections Project. Clinton won the popular vote by more than 2.8 million votes — more than Al Gore won the popular vote in 2000, John F. Kennedy in 1960, and Richard Nixon in 1968. She lost the election by 107,000 votes in three states…Clinton, the “unpopular” candidate, was able to turn out voters — just not where it mattered. Democrats lost the presidential election in three states; Clinton lost in Wisconsin by fewer than 23,000 votes and in Michigan by fewer than 11,000 votes. Votes afforded to Green Party candidate Jill Stein in either state were more than Trump’s margin of victory. Trump won Pennsylvania by a slightly bigger margin, surpassing Clinton by a bit more than 44,000 votes…Clinton’s ability to mobilize voters was in already-blue states — states like California and New York.”

Update (1/9): The title of this new FiveThirtyEight post–“Registered Voters Who Stayed Home Probably Cost Clinton the Election“–says it all.

2016 Presidential Endorsements

For the first time in its 40-year existence, Foreign Policy endorsed a presidential candidate:

Image result for never trumpIn the nearly half-century history of Foreign Policy, the editors of this publication have never endorsed a candidate for political office. We cherish and fiercely protect this publication’s independence and its reputation for objectivity, and we deeply value our relationship with all of our readers, regardless of political orientation.

It is for all these reasons that FP’s editors are now breaking with tradition to endorse Hillary Clinton for the next president of the United States.

Our readers depend on FP for insight and analysis into issues of national security and foreign policy. We feel that our obligation to our readers thus extends now to making clear the great magnitude of the threat that a Donald Trump presidency would pose to the United States. The dangers Trump presents as president stretch beyond the United States to the international economy, to global security, to America’s allies, as well as to countless innocents everywhere who would be the victims of his inexperience, his perverse policy views, and the profound unsuitability of his temperament for the office he seeks.

The litany of reasons Trump poses such a threat is so long that it is, in fact, shocking that he is a major party’s candidate for the presidency. The recent furor over his vile behavior with women illustrates the extraordinary nature of his unsuitability, as does his repudiation by so many members of his own party — who have so many reasons to reflexively support their nominee.

This endorsement made me want to look at the state of newspaper and magazine endorsements thus far. As of now, there is not a single newspaper or magazine that has endorsed Trump. None. Zero. And the following have more-or-less broken tradition:

  • New York Daily News: pro-Clinton. Typically centrist; endorsed George Bush in 2004 and Mitt Romney in 2012 (Barack Obama in 2008).
  • Houston Chronicle: pro-Clinton. Third time in 64 years to endorse a Democratic candidate.
  • Tulsa World: No endorsement. Franklin Roosevelt in 1940 was the last Democratic candidate they endorsed.
  • Richmond Times-Dispatch: pro-Johnson. Have endorsed every Republican nominee since 1980. First endorsement of a Libertarian.
  • Dallas Morning News: pro-Clinton. First endorsement of a Democratic nominee since Roosevelt in 1944.
  • New Hampshire Union Leader: pro-Johnson. First Libertarian endorsement and non-Republican in over 100 years.
  • Cincinnati Enquirer: pro-Clinton. First Democratic endorsement since Woodrow Wilson in 1916.
  • Arizona Republic: pro-Clinton. First Democratic endorsement in its 126-year history.
  • Detroit News: pro-Johnson. First non-Republican endorsement in its 143-year history.
  • USA Today: not Trump. First time to take a position on the presidential election in its 34-year history.
  • San Diego Union-Tribune: pro-Clinton. First Democratic endorsement in its 148-year history.
  • Desert Sun: pro-Clinton. First Democratic endorsement in 90-year history.
  • Foreign Policy: pro-Clinton. See above.
  • Wired: pro-Clinton. Magazine’s first presidential endorsement.
  • Deseret News: not Trump. Has not endorsed a candidate in 80 years, but wrote an editorial saying Trump should drop out.

And much more. Check out the list here.

Protest Votes Exist

There's a reason for using a pic of an iceberg that you'll see below. But if it made you think of the Titanic well, this election year, that fits, too. (Photo by AWeith, CC BY-SA 4.0)
There’s a reason for using a pic of an iceberg that you’ll see below. But if it made you think of the Titanic well, this election year, that fits, too. (Photo by AWeith, CC BY-SA 4.0)

A Facebook friend shared a link to Clay Shirky’s Medium post (There’s No Such Thing As A Protest Vote) and tagged me. I had thoughts, and so I thought I’d share them.

I’m not sure if Shirky’s slapshot case is meant as a serious argument or merely a pretext for a rant. This paragraph, which comes towards the end when the tone shifts abruptly from reasoned to strident, highlights my confusion:

Throwing away your vote on a message no one will hear, and which will change no outcome, is sometimes presented as ‘voting your conscience’, but that’s got it exactly backwards; your conscience is what keeps you from doing things that feel good to you but hurt other people. Citizens who vote for third-party candidates, write-in candidates, or nobody aren’t voting their conscience, they are voting their ego, unable to accept that a system they find personally disheartening actually applies to them.

Nevertheless, the first 1,000 words present a case, and so we’ll take it on the merits. But first, let us simply observe that accusations of disloyalty are Plan A when it comes to browbeating recalcitrant idealists into conformity.That’s not to say there’s never truth to the idea that we should sometimes put our personal preferences aside for the sake of a group’s welfare. It is to say that deploying the exact logic under which despotic regimes have justified silencing voices of protest throughout history ought to be treated as a red flag. Besides which, there’s a big difference between issues of personal taste and issues of conscience. That’s something we’ll return to in the end.

Shirky’s argument about protest votes boil down to two claims. First, it’s a “message no one will hear” and second, it “will change no outcome.” Both these claims are false.

When it comes to “no one will hear,” Shirky argues that since it doesn’t change the outcome of the vote, not voting and voting for a third-party candidate are equivalent. His argument seems to be that, since we can’t guess why a voter would pick Gary Johnson or Jill Stein or to abstain altogether, no information is transmitted. But if no information is transmitted, then we ought to be able to say absolutely nothing whatsoever about the differences in political preference between a group of Gary Johnson voters, a group of Jill Stein voters, and a group of non-voters. But we can infer all kinds of things about the preferences of these groups from the votes they cast. What’s more, if we can’t derive why they voted from the who they voted for, then that applies to voters who pick Hillary Clinton or Donald Trump as well. If we can’t say anything about third-party voters or non-voters based on their votes (or lack thereof), then we can’t say anything about anybody based on voting behavior.

On the other hand, maybe Shirky isn’t saying that it’s impossible to derive the why from the who, but just that nobody will take the trouble to do the derivation: “But it doesn’t matter what message you think you are sending, because no one will receive it. No one is listening.” This approach doesn’t fly either. As Shirky points out, in a parliamentary system the coalition-building happens after the votes are cast when various small, relatively ideologically pure parties have to form a coalition to govern. In a 2-party, winner-take-all system the coalition-building happens before the votes are cast, with the Republican and Democratic parties pulling together various constituencies to form coalitions. But how does he think that this happens if the respective parties don’t pay very careful attention to what they can infer about voters from every source available, including third-party votes? It is emphatically not the case that “no one is listening.” We have an entire industry of pollsters, analysts, and consultants who make a living by listening to the signals that voters send, and a protest vote is a pretty clear signal.

It’s easy to see how the argument that protest votes “will change no outcome,” falls immediately after the argument that they are a “message no one will hear.” The Democratic and Republican parties are not going to spend millions and millions of dollars every year on small armies of pollsters, analysts, and consultants to infer voter preferences for the purpose of crafting their coalition and then just ignore the results.

If literally the only thing that you are willing to consider is the result of one, particular election then–and only then–does it make sense to say that protest votes are indistinguishable from non-voting and therefore don’t exist. But pretending those future elections don’t exist doesn’t actually mean that they don’t.


If you want to change the behavior of one of the two major parties, then the best way to do it is not to stay home, but rather to vote for someone else. If you would like to change one of the two major parties to be more like the other one, than by all means switch from R to D or from D to R. In that case, protest votes don’t enter into it. But if you wan to move either (or both) of the major parties in a direction neither is amenable to, then the best and clearest way to send that signal is through a third party.

Of course, there is a cost associated with that. The protest vote is going to have no impact on today’s election and only a possible impact on future elections. And so the most reasonable theory of protest voting is to attempt to weigh the benefit of sending a corrective signal for the future against the cost of not influencing an election in the present. This is a very, very difficult calculation to make and the stakes are high. That is why reasonable people can come to differing conclusions, even when their political views are quite similar.

Notably, however, the simplest and most straight-forward explanation of protest voting is omitted from Shirky’s piece, which posits only three options: boycott, defection, or “step to third-party victory.” Each of these options has some validity to it, but none of them are as potent or as simple as the one given here. Most tellingly: none of them incorporate Shirky’s own analysis of the incentives of coalition-building in a 2-party system. Defection is closest to what I have in mind, but Shirky explains it only in terms of simplistic: “voters believe they can force a loss on either the Democrats or the Republicans, and thus make that party adopt their preferred policies, rather than face another such loss in the future.”

It is neither necessary nor possible for protest votes to “force a loss.” It is not possible for the simple reason that–like any complex event–there is never one, singular explanation for the outcome of a vote. It is not necessary because the two major parties are in constant competition with each other to build the bigger coalition. Protest voters do not need to threaten or coerce them–although that might help–but only to clearly communicate what they want.

One major thing to keep in mind: the entire point of having an election is that we don’t know ahead of time what people want. If we had perfect knowledge of preferences, we wouldn’t need to vote. Ergo, the parties don’t actually know–with perfect precision–what constituencies exist out there and how best to appeal to them. Protest votes are an extreme form of conveying that information, and that is a much lower bar than the idea of having to “force a loss.”

Finally, the reactions to protest vote are going to be subtle, temporally distant, and often intentionally muted. They will be subtle because a major party is a coalition: a delicate balance of overlapping constituencies. They do not, with rare and historical exceptions, make abrupt changes in any direction because it threatens the cohesiveness of the overall balance. They will be temporally distant because the very earliest that a party can visibly react is the next election, a minimum of two years away, and in practical terms the lag will be even greater. And they will often be intentionally muted because part of the narrative every party tries to present is that it’s always been right, and so changes–especially in reaction to third-party protest votes–will be deliberately downplayed in front of most audiences.

But the fact that the impact of protest votes is not easy to spot doesn’t mean that they don’t exist. To believe that, you’d have to believe either that protest votes convey no useful information about voter preferences or that Democrats and Republicans ignore useful information about voter preferences, neither of which is tenable.

Shirky’s argument is very hard to take seriously on its merits, since it requires us to ignore obviously relevant factors (like future elections) and accept flagrantly false premises (like the idea that protest votes either convey no useful information or that major political parties don’t care about that information). However, it does provide an approximately thousand-word pretext for the real payload of this article, which consists of claims like these:

  1. Advocates of wasted votes don’t bring up this record of universal failure, because their votes aren’t about changing political results. They’re about salving wounded pride.
  2. Citizens who vote for third-party candidates, write-in candidates, or nobody aren’t voting their conscience, they are voting their ego…
  3. The people advocating protest votes believe they deserve a choice that aligns closely with their political preferences.

This makes this whole piece an example of bulverism, a coin terms by C. S. Lewis.

You must show that a man is wrong before you start explaining why he is wrong. The modern method is to assume without discussion that he is wrong and then distract his attention from this (the only real issue) by busily explaining how he became so silly.

Another explanation of how bulverism works shows how it relates in this case:

Suppose I think, after doing my accounts, that I have a large balance at the bank. And suppose you want to find out whether this belief of mine is “wishful thinking.” You can never come to any conclusion by examining my psychological condition. Your only chance of finding out is to sit down and work through the sum yourself. When you have checked my figures, then, and then only, will you know whether I have that balance or not. If you find my arithmetic correct, then no amount of vapouring about my psychological condition can be anything but a waste of time. If you find my arithmetic wrong, then it may be relevant to explain psychologically how I came to be so bad at my arithmetic, and the doctrine of the concealed wish will become relevant — but only after you have yourself done the sum and discovered me to be wrong on purely arithmetical grounds. It is the same with all thinking and all systems of thought. If you try to find out which are tainted by speculating about the wishes of the thinkers, you are merely making a fool of yourself. You must first find out on purely logical grounds which of them do, in fact, break down as arguments. Afterwards, if you like, go on and discover the psychological causes of the error.

I’ve also seen this general tactic referred to as “the heuristic of suspicion.” The general idea is that we give very short shrift to what our opponents actually think–to the objective case they are making–and instead rush quickly past it to psychological analysis that takes their error for granted and indulges in self-satisfied dissection of their inferiority. I’m not saying we can indulge in zero time spent on analyzing the motives or intentions of our interlocutors, but I do think we should try to shift the balance onto the arguments at hand and take them seriously. And, on that basis, Shirky’s argument that protest votes are indistinguishable from non-voting, simply do not hold up to any level of scrutiny.

Now, at the very end, I want to return to the first comment I made about personal taste and conscience. Here is the beginning of Shirky’s concluding paragraph:

None of this creates an obligation to vote, or to vote for one of the two viable candidates. It is, famously, a free country, and you can vote for anyone you like, or for no one.

If Shirky really believes that there is no obligation to vote for a major party candidate, than his entire essay collapses into nonsense. The whole point–from start to finish–is that protest voters are selfish egotists who are abdicating their duty and “making the rest of us do the work of deciding.” If that isn’t a violation of an obligation, then what on Earth could be?

That fact that we are allowed under the law to behave in a certain way is not the only nor the final word on what our obligations may or may not be. The set of things that are obligatory (in any sense) and the set of things that are legally required are not the same. So–far from celebrating a genuine sense of freedom in which government refrains from attempting to demarcate the boundaries of the permissible–Shirky is engaging in a kind of totalitarian thinking in which what we must do and what the law requires are assumed–despite all common sense–to be identical.

This is not an incidental misstep. It’s integral to Shirky’s case. After all, if a protest vote is really just a matter of personal preference–if I prefer Candidate Alice to Candidate Bob in the same way in which I prefer rocky road to mint chocolate chip or blue to red–then it would be the height or selfishness to become an absolute stickler on that point to the detriment of the group. This is a world of moral relativism, where all moral decisions are reflections of personal preference and can pretend to no greater validity.

It’s not a very coherent world. In it, Shirky first argues that protest votes are immoral because they place one’s personal preferences ahead of the common good. But, in this incoherent world, Shirky has to immediately repudiate his own argument. You are an arrogant, hypocritical egoist if you vote third party! Not that that means you can’t do it, of course. You can do whatever you’d like. Who am I to judge? I’m not saying. I’m just saying. The whole thing collapses into irrelevant, incomprehensible muttering.

But if there is such a thing as an objective moral reality, then when a person refuses to vote the way you’d like them to out of conscience, you can’t simply browbeat them for being selfish. Perhaps they are! But perhaps they are acting out of an earnestly held believe in a universally applicable moral stand, one that does not suddenly become irrelevant or disposable merely because it is unpopular or inconvenient. And so, in this world, the specifics actually matter. In this world, some people who vote third party are irresponsible and selfish and some are responsible and selfless. It’s frustrating that we can’t always line up the good guys and the bad guys based on how they vote. It’s downright dangerous when we try to do so anyway.

Shirky offers a thin pretense of not telling you how to vote, but in fact he not only tells you how to vote, but specifically who to vote for. His post has a form of non-judgmentalism but denies the power thereof. I’ll do you the courtesy of not pretending to be neutral. Instead, I’ll just go ahead and tell you what I think you should do without any caveats or qualifications: you should vote, and you should vote informed, and you should vote your conscience. Consider the costs and the benefits of voting for one of the two candidates who will almost certainly win vs. the costs and benefits of voting for a candidate who will almost certainly not win. Take it it seriously, do your homework, and if you’re religious pray for guidance. Then vote accordingly.

And hey, if you’re looking for a silver lining during this awful election, here’s one. In the past, I always felt that there was a clear-cut candidate that should win. There was always a little strain and tension when family or close friends felt the same, but about the other guy. This year, I don’t have that strain. It’s all a mess, and I can see compelling cases for voting in a lot of different ways. I have people I respect voting for Clinton, voting for Trump, voting for Johnson, and voting for McMullin. And–for the first time–I actually have absolutely zero reservations about their voting differently than I am. Not to say I agree with all of them–I don’t! I can’t!–but I can see where they are coming from. So that, at least, is a nice side-effect of this ongoing train wreck.


Assessing the Presidential Candidates’ Economic Proposals

A brand new study analyzes the economic proposals of both Trump and Clinton and the results aren’t encouraging. As reported by Financial Times,

US Republican presidential candidate Donald Trump’s protectionist trade policies would send the US into recession, result in the loss of almost 4.8m private sector jobs and lead to shortages of consumer goods such as iPhones, according to the most detailed study yet of his plan.

…The study also offers a sceptical view of Democratic candidate Hillary Clinton’s trade policies, and particularly her opposition to the Trans-Pacific Partnership, a vast new Pacific Rim trade pact the US has negotiated with Japan and 10 other economies. 

But Mr Trump’s threats to rip up existing US trade agreements and impose punitive 45 per cent tariffs on goods from China and a similar 35 per cent levy on products imported from Mexico would probably set off a trade war and wreak huge damage on the US economy, the study found. 

“While [Mrs] Clinton’s stated trade policy would be harmful, [Mr] Trump’s stated trade policy would be horribly destructive,” said Adam Posen, the institute’s president. “His stated approach to the global economy of waging trade war and protecting uncompetitive special interests would be disastrous for American economic wellbeing and national security.” 

While the “biggest trade-related employment impact…would be in manufacturing and on states such as Washington,” it turns out “the biggest impact on jobs would come as the consequences of a trade war reverberate out across the economy hitting retail distribution hubs, grocery stores, restaurants and even hospitals, the study found. It also would likely lead to shortages and higher prices of consumer goods — including popular products such as smartphones — and potentially even have an impact on US retirement savings.”

That’s exciting.

Image result for we're screwed gif

There’s a lot more to Trump support than racism.

hillbilly elegy

I’m a conservative. Over the last 9 months or so the Trump campaign has sent me from incredulous to enraged to dejectedly resigned, and I admit I’m still trying to figure out what just happened.

To hear some of my leftist friends talk, the Trump phenomenon is the inevitable climax of an increasingly out-of-touch, racist, backwards party, a predictable extension of the “extremism” of Bush.

Those theories makes no sense to me. The conservatives I know who supported Bush hate Trump. In fact, Pew Research recently published a study showing the most religious and “very conservative” Republicans have been most resistant to Trump. This study echoes a Gallup poll last year that found the highly religious ranked Trump 12th out of 17 GOP candidates. But if the most religious and conservative Republicans dislike him, who has been supporting Trump from the outset?

In his article “Trump: Tribune of Poor White People,” Rod Dreher interviews J.D. Vance, author of Hillbilly Elegy: A Memoir of a Family and Culture in Crisis, on the interplay between our current political parties and hillbilly culture.

The interview is a fascinating read. Vance discusses a variety of sometimes opposing ideas in an empathetic but honest way. For example, he talks about how the anti-elitist streak in hillbilly culture helps those who achieve financial success stay grounded, but it also pressures people not to become successful in the first place. Vance talks about  the problems with how, when it comes to the poor, there are opposing narratives suggesting either (a) the poor are helpless, with no power to affect their own lives or (b) the poor must simply bootstrap their way out of poverty. Both of these views fail to address important factors.

Vance uses his understanding of hillbilly culture to explain Trump’s popularity:

The two political parties have offered essentially nothing to these people for a few decades.  From the Left, they get some smug condescension, an exasperation that the white working class votes against their economic interests because of social issues, a la Thomas Frank (more on that below).  Maybe they get a few handouts, but many don’t want handouts to begin with.  

From the Right, they’ve gotten the basic Republican policy platform of tax cuts, free trade, deregulation, and paeans to the noble businessman and economic growth.  Whatever the merits of better tax policy and growth (and I believe there are many), the simple fact is that these policies have done little to address a very real social crisis.  More importantly, these policies are culturally tone deaf: nobody from southern Ohio wants to hear about the nobility of the factory owner who just fired their brother.

Trump’s candidacy is music to their ears.  He criticizes the factories shipping jobs overseas.  His apocalyptic tone matches their lived experiences on the ground.  He seems to love to annoy the elites, which is something a lot of people wish they could do but can’t because they lack a platform.  

This part especially resonated with me:

The other big problem I have with Trump is that he has dragged down our entire political conversation.  It’s not just that he inflames the tribalism of the Right; it’s that he encourages the worst impulses of the Left.  In the past few weeks, I’ve heard from so many of my elite friends some version of, “Trump is the racist leader all of these racist white people deserve.” These comments almost always come from white progressives who know literally zero culturally working class Americans.  And I’m always left thinking: if this is the quality of thought of a Harvard Law graduate, then our society is truly doomed.

Read the full interview here.


Quick Thoughts on Trump and Predictions

Now that Donald Trump has wrapped up the GOP nomination, there are two kinds of articles flooding the national conversation:

  1. Surely this is a sign of the End Times.
  2. Pundits are dumb!

I can’t really contest #1 because, in a nutshell, I agree.

But let me talk about #2 for a second. First: yes, articles like this one and this one and this one that are full of  screenshots of pontifications about how Trump has no chance are funny now that he is the winner of the primary after all. But hold on a second. Does the fact that somebody was wrong about a prediction actually mean that they were dumb? That their prediction was a stupid one? Not necessarily.

If you bet your life savings on being able to flip a coin 5 times in a row and get heads every time, I am going to say you have made a dumb decision. You’ve got about a 97% chance of losing everything and only a 3% chance of winning. But lets say you ignore me, place your bet, and you manage to get lucky and win. This doesn’t make you a genius. It makes you an idiot who got lucky.

People are terrible at statistical reasoning. We like reality simple and we like it deterministic. That means we like to believe that if Y happened, then it’s because of X and obviously X had to cause Y and couldn’t have caused W or Z or coconut soup.

But reality is messy. Sometimes good predictions fail because there was systematic information that nobody had access to. Sometimes good predictions fail because of just plain old dumb luck (like in my example with coin flips). The point is just that even someone who is right 95% of the time is going to get it wrong 5% of the time, and that doesn’t mean that in those 5% of cases where they get it wrong there was actually anything they could or should have predicted differently.

Take a guy like Nate Silver. Unlike your average pundit, he’s got a track record of getting things right by using objective, transparent, quantitative methods. He was largely right in 2008, 2010, and 2012. And this year, he was one of the leading voices telling people to calm down and not worry too much about Trump. As a result, he is now being singled out by name by, for example, Business Insider: NATE SILVER: ‘We basically got the Republican race wrong’ as one of those silly propeller heads we never should have listened to in the first place.

Not so fast.

First, this is actually a meaningful question. Journalistic credibility is already at historical lows, and that has dire consequences for society. A strong, free press is a bedrock institution of democracy, and if the press is perceived to be hopelessly biased or suborned or just plain old incompetent, that’s bad for our nation. And it matters whether the perception is real or false. In reality, it’s probably a little of both. My read on Nate Silver (who fulfills the role of a journalist even if he doesn’t fit the old, pre-21st century conception) is that he’s an honest guy doing good work. On the other hand, there is a legitimate argument that journalism is fracturing into a thousand click-bait seeking echo chambers. If we, the reading public, don’t discern between episodic mistakes and systematic corruption, then we can’t provide the necessary incentives for the good ship Free Press to right herself.

It’s also meaningful because we need to know what happened with Trump. Yes, obviously Silver (and most everyone else) predicted the wrong outcome. But the question is why were they wrong? Was it incompetence? Was it information nobody had access to? Or did Trump keep getting heads every time he flipped the coin? Almost none of the articles I’ve read address this question. They just assume that if your prediction fails it was a bad prediction. In other words, they are assuming incompetence. And this means we’re likely to draw all the wrong conclusions about what Trump’s victory means.

So–even though he got it wrong–I strongly recommend reading Silver’s take. In it, he raises this important question explicitly:

What’s much harder to say is whether Trump is a one-off — someone who defied the odds because a lot of things broke in his favor and whose success will be hard to repeat — or if he signifies a fundamental change in American politics.

In answer, he notes the three ways his prediction was systematically off:

  1. Voters are more tribal than I thought.
  2. GOP is weaker than I thought.
  3. Media is worse than I thought.

It’s not clear if these are things Silver should have known or if these are new facts that we only know now because of Trump’s victory, but the point is that–because they are systematica rather than merely random–these are new facts that can help us understand and predict the world going forward.

But there is also room for randomness. Take a look at this chart:



Prior to the New York state primary, Trump had never taken a majority in a GOP primary. Since then, he’s never not taken the majority. What happened? Silver’s guess is that voter rationally and deliberately evaluated Trump’s argument that whoever gets the most votes should get the nomination (even if they fall short of a majority) and so decided to make sure Trump got the majority to avoid a contested election. To me, this makes very little sense, because “rationally and deliberately” are not adverbs I usually apply to voter action. And I’m not making a dig at Trump’s voters, I just mean that group action seldom has that kind of rational, deliberate nature without some kind of elaborate infrastructure to support it.

My guess? Well, two of the big stories leading up to New York were Ted Cruz’s statements about “New York values” and his machinations to win delegates without winning elections in Colorado and Maine. So, what do you think happens when a populist demagogue leading a movement of voters who feel bitter, dispossessed, and disenfranchised by a corrupt, elitist system gets to tar his top challenger as a conniving insider using questionable technicalities to steal votes? Oh, and he insulted you, too. Cruz walked face-first into Trump’s zeitgeist, and that–I am guessing–fueled Trump’s unprecedented win in New York state which, in turn, snowballed into his further victories after that.

Who knows if I’m right or not, though, and that’s my point. This is basically the blind luck category. If I’m right, then it took the coincidence of Cruz and Trump being pretty much right where they were leading up to New York in the polls, plus the rules being what they were in Maine and Colorado, etc. You can find explanations for all of that, but at a certain point we’re just making stories up after the fact to explain what is, in reality, an essentially chaotic reality. Or hey, maybe I’m wrong, in which case nobody knows.

Either way, this matters a lot for the big question: is Trump a one-off? Or a sign of the new reality? I don’t know the answer definitively, but I do know the shallow and vapid articles where some pundits make fun of other pundits for not punditting well enough are just a distracting waste of time.

Have Mormons Checked Out of the Political Process?

On the very large sample size of my Facebook friends and my own personal feelings (so clearly, sound and publishable data), I’ve noticed that Mormons (read: my Facebook feed and myself. Except, I guess, not me at this moment.) are kind of quiet about the election this year. I’ve never seen it happen (on my Facebook feed).

I’m going to extrapolate from this great sample of data: I wonder if this silence has to do with the following.

This year we have a super rich guy with multiple marriages, multiple failed businesses, no government experience, no experience turning around a failed business, no (previously) conservative values, who is vulgar, rude, hates immigrants, hates women, speaks without learning, and is doted upon by the media (even when they are making fun of him its with a gentle glee that their (read: Democratic) nominee will surely beat this buffoon).

Mitt-MissMeYetOnce upon a time we had an amazing anti-Trump candidate (other than he was also rich, but only 1/18th as rich as Trump). But this candidate had this YUGE personal problem. He was a Mormon. And that just won’t do. The media, Republican establishment, and core Republican (conservative? evangelical? INSANE?) voters just could not allow a stand-up guy like this to be the nominee, unless of course it was against a sitting president, a set-up for failure.

This is what you have done to us media (Democrats?) and Republicans. But they haven’t just done it to Mormons. They’ve done it to America.