19 Things We Learned from the 2016 Election

OK, we can all agree that the November election result was a shocker. According to news reports, even the Trump campaign team was stunned to come up a winner.

So now seemed like a good time to go over various theories floating around in political science and political reporting and see where they stand, now that this turbulent political year is drawing to a close. By the time I was done writing it up for Slate, I came up with 19 lessons learned. I thank my colleague Bob Erikson for help on some of these.

1. The party doesn’t decide.

We can start with the primaries, which destroyed the Party Decides theory of Marty Cohen, David Karol, Hans Noel, and John Zaller, who wrote in 2008 that “unelected insiders in both major parties have effectively selected candidates long before citizens reached the ballot box.” You can’t blame authors of a book on political history–its subtitle is “Presidential Nominations Before and After Reform”–for failing to predict the future. But it does seem that the prestige of the Party Decides model was one reason that Nate Silver, Nate Cohn, Jonathan Chait, and a bunch of other pundits not named Nate or Jonathan were so quick to dismiss Donald Trump’s chances of winning in the Republican primaries.

Indeed, I myself was tempted to dismiss Trump’s chances during primary season, but then I read that article I’d written in 2011 (http://campaignstops.blogs.nytimes.com/2011/11/29/why-are-primaries-hard-to-predict) explaining why primary elections are so difficult to predict (multiple candidates, no party cues or major ideological distinctions between them, unequal resources, unique contests, and rapidly-changing circumstances), and I decided to be careful with any predictions.

2. That trick of forecasting elections using voter predictions rather than voter intentions? Doesn’t work.

Economists David Rothschild and Justin Wolfers have argued that the best way to predict the election is not to ask people whom they’ll vote for, but rather ask whom they’ll think will win (http://www.nytimes.com/2012/11/02/us/politics/a-better-poll-question-to-predict-the-election.html). Their claim was that when you ask people whom they think will win, survey respondents will be informally tallying their social networks, hence their responses will contain valuable information for forecasting. When this idea was hyped back in 2012, I was skeptical (http://themonkeycage.org/2012/11/people-can-predict-elections-even-when-polls-cant/#comment-37868), taking the position that respondents will be doing little more than processing what they’d seen in the news media, and I remain skeptical, following a 2016 election that was a surprise to most.

3. Survey nonresponse is a thing.

It’s harder and harder to reach a representative sample of voters, and it’s been argued that much of the swing in the polls is attributable not to people changing their vote intention, but to changes in who responds or doesn’t respond. In short, when there is good news about a candidate, his or her supporters are more likely to respond to polls. Doug Rivers, David Rothschild, Sharad Goel, and I floated this theory following some analysis of opinion polls from 2012 (http://www.stat.columbia.edu/~gelman/research/published/swingers.pdf), and it seems to have held up well during the recent campaign season (http://www.slate.com/articles/news_and_politics/politics/2016/08/don_t_be_fooled_by_clinton_trump_polling_bounces.html)

The only hitch here is that the differential nonresponse story explains variation in the polls but not the level or average shift. The final polls were off by about 2 percentage points (http://statmodeling.stat.columbia.edu/2016/11/09/explanations-shocking-2-shift/), suggesting that, even at the end, Trump supporters were responding at a lower rate than Clinton supporters.

4. The election outcome was consistent with “the fundamentals.”

Various models predict the election outcome not using the polls, instead using the national economy (as measured, for example, in inflation-adjusted personal income growth during the year or two preceding the election) and various political factors. In 2016 the election was growing slowly but not booming (a mixed signal for the voters), the incumbent party was going for a third term in office (traditionally a minus, as voters tend to support alternation), and the Republicans controlled both houses of Congress (a slight benefit for the Democrats in presidential voting, for that minority of voters who prefer party balancing), and, on the left-right scale, both candidates were political centrists relative to other candidates from their parties. This information can be combined in different ways: Running a version of the model constructed by the political scientist Doug Hibbs, I gave Hillary Clinton a forecast of 52 percent of the two-party vote (http://www.slate.com/articles/news_and_politics/politics/2016/08/why_trump_clinton_won_t_be_a_landslide.html). Fitting a similar model but with slightly different parameters, political scientist Drew Linzer gave Clinton 49 percent (https://pkremp.github.io/report.html). In October the political science journal PS published several articles on forecasting the election, including one from Bob Erikson and Chris Wlezien who concluded, “the possibility of greater campaign effects than we typically observe should constrain our confidence in the predictions presented here.” (https://www.cambridge.org/core/journals/ps-political-science-and-politics/article/div-classtitleforecasting-the-presidential-vote-with-leading-economic-indicators-and-the-pollsdiv/8212BC860F9BAF1570EBE2B8CF603709/core-reader)

All these fundamentals-based models have uncertainties on the order of 3 percentage points, so what they really predicted is that the election would not be a landslide. The actual outcome was consistent with these predictions. That said, a wide range of outcomes–anything from 55-45 to 45-55–would’ve jibed with some of these forecasts. And the non-blowout can also be explained by countervailing factors: Perhaps Trump was so unpopular that anyone but Clinton would’ve destroyed him in the general election, and vice versa. That seems doubtful. But who knows.

5. Polarization is real.

Democrats vote for Democrats, Republicans vote for Republicans. It’s always been thus—what would the party labels mean, otherwise—but cross-party voting keeps declining, and members of the out-party hold the president in lower and lower esteem. Consider, for example, Donald Trump’s criticism of Barack Obama during the presidential debates. Obama is popular so this might seem to have been a mistake to stand against him—but Obama is deeply unpopular among Republicans, especially those Republicans who are likely to vote.

A corollary of polarization is that, if there aren’t many people in the middle to be persuaded, it makes sense for candidates to focus on firing up their base, and this is a key part of the story of the success of the Trump campaign. You can bet that activists of both parties will have learned this lesson when 2020 comes along.

6. Demography is not destiny.

We’d been hearing a lot about how the Republican party, tied to a declining base of elderly white supporters, needs to reassess. For example, here’s Jamelle Bouie in Slate, under the heading, “It Lost Black Voters. Now It’s Losing Latinos. What’s Left Is a Broken, White GOP” (http://www.slate.com/articles/news_and_politics/cover_story/2016/10/trump_and_the_gop_are_alienating_latinos_the_way_they_once_alienated_black.html): “The latest tracking poll from Latino Decisions shows Republican nominee Donald Trump with 16 percent support, versus 74 percent for Hillary Clinton. Looking ahead to November, the group expects that electorate to cast the vast majority of its votes for Clinton, 82 percent to 15 percent for Trump, which would be the most lopsided total in history.” According to exit polls, the Latino vote ended up dividing 66%-28%, a clear Clinton lead but nothing like the forecast from Latino Decisions—a forecast that should’ve been suspect, given that it contradicted the organization’s own polls! Longer term, it may well be that the Republican party needs to change with the times, but destiny hasn’t happened yet.

7. Public opinion does not follow elite opinion.

Perhaps the most disturbing theoretical failure of political science is the general idea that voters simply follow elite opinion. This worked in 1964 to destroy Goldwater, for instance. Or so the story goes. The implication is that voters had to be told Goldwater was scary. They could not figure it out for themselves.

In 2016, Trump was opposed vigorously as dangerous, incompetent, xenophobic, tyrannical, and unhinged, by almost everybody in elite circles: most of his Republican primary opponents at one time or another, a large number of conservative intellectuals, former Republican candidates Romney and McCain, the various Bushes, the media, almost all newspaper editorialists including those that were reliable Republican supporters, all Democrats, about 10 Republican senators, and even some pundits on Fox News. Further, Trump’s breaking of all the standard niceties of politics was there for all to see for themselves. But half the voters said, we go with this guy anyway. The falcon no longer hears the falconer,” as W. B. Yeats put it.

8. There is an authoritarian dimension of politics.

Political scientists used to worry about authoritarianism within the electorate. Mainstream politicians, ranging from Republicans on the far right to lefties such as Sanders, tend not to go there. Trump did. In doing so he broke the rules of politics with extreme comments about his opponents, etc., that are hard to forget. But a significant segment of the electorate, maybe 20 percent, have always been waiting for its authoritarian champion on what we now call the alt-right dimension. There had not been one in the modern era. Trump’s absolute dominance of the political news for over a year signifies this uniqueness. There had been others with this sort of appeal—Joe McCarthy (http://statmodeling.stat.columbia.edu/2016/06/08/donald-trump-and-joe-mccarthy/), George Wallace—but they never came close to becoming our national leader.

9. Swings are national.

When you look at changes from one election to the next, the country moves together. If you plot vote swings by county, or by state, you see much more uniformity in the swing in recent years than in previous decades (http://statmodeling.stat.columbia.edu/2009/01/14/state-by-state/). The swing from 2012 to 2016 was also close to uniform. There’s been lots of talk of Pennsylvania, Michigan, and Wisconsin, and these three states did make the difference in the electoral college, but similar swings happened all over the country. To put it another way, nonuniform swings were essential to Trump’s win, but looking at public opinion more broadly, the departures from a national swing were small, and consistent with the increasing nationalization of elections in recent decades.

10. The ground game was overrated.

The Democrats were supposed to be able to win a close election using their ability to target individual voters and get them out to the polls. But it didn’t happen this way. The consensus after 2016, which should’ve been the consensus earlier: Some ground game is necessary, but it’s hard to get people to turn out and vote, if they weren’t already planning to.

11. News is siloed.

For years we’ve been hearing that liberals hear one set of news, conservatives hear another, and moderates are being exposed to an incoherent mix, so that it’s difficult for anyone to make sense of what everyone else is hearing. There have always been dramatic differences of opinion (consider, for example, attitudes toward civil rights in the 1950s and the Vietnam war in the 1970s) but research on public opinion has shown an increase in partisan polarization in recent decades. The 2016 election, with its sharp divide between traditional news organizations on one side and fake news spread by Twitter and Facebook on the other, seems like the next step in this polarization.

It’s the political version of Moore’s Law, which says that every time the semiconductor manufacturers have run out of ways to squeeze more computing power on a chip, they come up with something new. Whenever it starts to seem like there’s no more room for Americans to polarize, something new comes up—in this case saturation of social media by fake news, along with a decline of the traditional TV networks and continuing distrust of the press.

12. The election wasn’t decided by shark attacks.

Political scientists Chris Achen and Larry Bartels have argued that voters are emotional and that elections can be swayed by events such as shark attacks that should logically be irrelevant to voting decisions. Others have analyzed data and claimed to find that close elections can be decided by the outcomes of college football games (with happy voters being more likely to pull the lever for the incumbent party’s candidate). Others have reanalyzed these data and found no such effect (http://statmodeling.stat.columbia.edu/2016/10/29/no-evidence-shark-attacks-swing-elections/). What does 2016 say about all this? Not much.

You can’t prove a negative so it’s possible that irrelevant stimuli could have made all the difference. But the big stories about this election were that (a) lots of bad information about Donald Trump did not sway much of the electorate, and (b) Clinton’s narrow Electoral College loss may well be attributed to FBI leaks, which were relevant to the voting decision in reminding voters (perhaps inappropriately) of concerns about her governing style. The 2016 election was not about shark attacks or football games but rather about big stories that didn’t matter much, or canceled each other out.

13. Overconfident pundits get attention.

From one direction, neuroscientist Sam Wang gave Hillary Clinton a 99 percent chance of winning the election; from the other, cartoonist and jar opener (http://statmodeling.stat.columbia.edu/2011/04/16/dilbert_update/) Scott Adams gave 98 percent odds in favor of Trump. Looking at it one way, both Wang and Adams were correct: Clinton indisputably won the popular vote while Trump was the uncontested electoral vote winner. After the election, Wang blamed the polls, which was wrong. The polls were off by 2 percent, which from a statistical standpoint wasn’t bad. Indeed this magnitude of error was expected from a historical perspective (http://statmodeling.stat.columbia.edu/2016/11/11/david-rothschild-sharad-goel-called-probabilistically-speaking/), even if it did happen to be consequential this time. The mistake was not in the polls but in Wang’s naive interpretation of the polls which did not account for the possibility of systematic nonsampling errors shared by the mass of pollsters, even though evidence for such errors was in the historical record. Meanwhile, Adams explains Trump’s victory as being the result of powers of persuasion, which might be so but doesn’t explain why Trump received less than half the vote, rather than the landslide that Adams had predicted.

I continue to think that polling uncertainty could best be expressed not by speculative win probabilities but rather by using the traditional estimate and margin of error. Much confusion could’ve been avoided during the campaign had Clinton’s share in the polls simply been reported as 52 percent of the two-party vote, plus or minus 2 percentage points.

There’s a theory that academics such as myself are petrified of making a mistake, hence we are overcautious in our predictions; in contrast, the media (traditional news media and modern social media) reward boldness and are forgiving of failure. This theory is supported by the experiences of Sam Wang (who showed up in the New York Times explaining the polls after the election he’d so completely biffed, http://www.nytimes.com/2016/11/19/opinion/why-i-had-to-eat-a-bug-on-cnn.html) and Scott Adams (who triumphantly reported that his Twitter following had reached 100,000).

14. Red state blue state is over.

Republicans have done better among rich voters than among poor voters in every election since the dawn of polling, with the only exceptions being 1952, 1956, and 1960, which featured moderate Republican Dwight Eisenhower and then moderate Democrat John Kennedy. Typically the upper third of income votes 10 to 20 percentage points more Republican than the lower third. This was such a big deal that my colleagues and I wrote a book about it! (http://press.princeton.edu/titles/9030.html) But 2016 was different. For example, here are the exit polls (http://www.cnn.com/election/results/exit-polls): Clinton won 53 percent of the under-$30,000 vote and 47 percent of those making over $100,000, a difference of only 6 percentage points, much less than the usual income gap. And we found similar minimal income-voting gradients when looking at other surveys. Will the partisan income divide return in future years? Will it disappear? It depends on where the two parties go. Next move is yours, Paul Ryan.

15. Third parties are still treading water.

The conventional wisdom is that minor parties are doomed in the U.S. electoral system. The paradox is that the only way for a minor party to have real success is to start local, but all the press comes from presidential runs. Anyway, 2016 seems to have confirmed conventional wisdom. Both major parties were highly unpopular, but all the minor parties combined got only 5.6 percent of the vote. On the other hand, 5.6 percent is a lot better than 1.7 percent (2012), 1.4 percent (2008), 1.0 percent (2004), or 3.7 percent (2000).

Glass half full is that minor parties are starting to get serious; glass half empty is that not much bloomed even in such fertile soil.

16. A working-class pundit is something to be.

Filmmaker and political activist Michael Moore gets lots of credit for writing, over a month before the election (http://michaelmoore.com/trumpwillwin/), an article entitled “5 Reasons Why Trump Will Win,” specifically pointing to the Rust Belt, angry white men, voter turnout, and other factors that everybody else was writing about after the election was over. Moore even mentioned the Electoral College. And unlike the overconfident pundits mentioned above, Moore clearly stated this as a scenario (“As of today, as things stand now, I believe this is going to happen …”) without slapping a 98 or 99 percent on to it.

What if Hillary Clinton had won 52 percent of the two-party vote and a solid Electoral College victory? Would we now be hearing from pundits with a special insight into white suburban moms? Maybe so. Or maybe we’d still be hearing about the angry white male, since 48 percent of the two-party vote would still be a lot more Trump support than most were expecting when the campaign began.

17. Beware of stories that explain too much.

After the election, which shocked the news media, the pollsters, and even the Clinton and Trump campaigns, my colleague Thomas Basboll wrote that “social science and democracy are incompatible. The social sciences conduct an undemocratic inquiry into society. Democracy is an unscientific way of governing it.” (http://secondlanguage.blogspot.com/2016/11/the-liberal-arts-of-being-ruled.html)

Maybe so. But Basboll could’ve written this a few days before the election. Had the election gone as predicted, with Clinton getting the expected 52 percent of the two-party vote rather than the awkwardly distributed 51% that was not enough for her to win in the Electoral College, it still would’ve been true that half of American voters had refused to vote for her. So there’s something off about these sweeping election reviews: even when you agree with the sentiments, it’s not clear why it makes sense to tie it to any particular election outcome.

The Republicans have done well in political strategy, tactics, timing, and have had a bit of luck too. One party right now controls the presidency, both houses of Congress, most of the governorships, and soon the Supreme Court. But when it comes to opinions and votes, we’re a 50/50 nation. So we have to be wary of explanations of Trump’s tactical victory that explain too much.

18. Goldman Sachs rules the world.

This theory appears to still hold up. Goldman Sachs candidate Hillary Clinton managed to lose the electoral vote, but Goldman Sachs Senator Chuck Schumer may now be the most powerful Democrat in Washington, while former Goldman Sachs executive Steve Bannon will be deciding strategy inside the White House. So it looks like the banksters are doing just fine. They had things wired, no matter which way the election went.

19. The Electoral College was a ticking time bomb.

Yup.

P.S. More here.

75 thoughts on “19 Things We Learned from the 2016 Election

  1. #3 & the post it links to are now a bit out of date — NYT puts results at 48.1% Clinton 46.1% Trump, which means Clinton was at 51% of 2-party vote. So, still some persistent polling error, but a very consequential (and not as consequential as the fact that the electoral college is weird) 1% difference, not the very consequential 2% or more difference that it looked like shortly after the election.

  2. When I started this comment, the Slate article had exactly 19 comments. I started to wax lyrical about the number 19, the length in years of the Metonic cycle. Then I saw that the number of Slate comments had gone up to 21. So much for the waxing.

    I am intrigued by the seventh lesson, “Public opinion does not follow elite opinion.” I see that it’s true to some extent. I see a complication here, though; by some standards, Trump counts as a member of the extreme elite, yet he won people over with his anti-elite rhetoric. So what matters more: being elite or seeming elite? Or is it some combination of eliteness and non-eliteness that wins people over?

    • Trump is perceived many many as an elite defector. No one not insane would expect him to give up all his elite advantages though, particularly since he claims to be using them in opposition.

  3. I think this captures a lot of good and important stuff about 2016, but it’s so focused on the Presidential race than it misses some key pertinent downballot stuff. Straight ticket voting soared in this election in the Senate races, though not the governor’s races.
    That strongly jives with 5 and 4. I think the fact that Senate results looked so much like the Presidential race argues against the idea that anyone but Clinton could have beaten Trump. It seems like people have taken sides on a basic level, and it’s hard for a person to alter that.

    One important idea may be that social identity is a powerful driver of party identification and voting. There’s been some debate on the importance of issues and ideology versus social groups. This cycle is a powerful testament to the significance of social identity in voting.

  4. Just to point out this: the idea you can’t prove a negative is a trope that has no basis in logic. For those who think so, just think of the logic of the claim itself.

  5. 1. I have had open for a few days in my browser an article that argues Trump got closer to 18% of the Latino vote. I’m not sure we know well enough yet.
    2. I’d say it’s only partly that news is “siloed”, but rather that news which you believe is objective is actually slanted factually. That is, to pick on the NYT (which deserves it), it’s one thing that they were pro-Clinton and anti-Trump but another that their coverage excluded the details that a thinking person would need to make up his or her own mind about what was going on. This I think led to the whole “elites missed it” idea. I’d say that’s more a spoon feeding than a siloing but then maybe you also mean this.
    3. To harp on my first point, the talk about Latino vote seems to have generally not looked at their locations in depth. As in, a lot of additional Latino votes in Texas are wasted for the Democrats unless you consider getting somewhat closer a victory. The same issue comes up with black turnout, which one would expect would decline after the large increases when Obama ran but which is only truly important in a handful of states (which Hillary lost). In that regard, I can only remember a few discussions at all of the way a Mormon candidate depressed white vote by a big number (2+ million, right?). It seems like people thought that vote was all in the deep red states but even a relatively small change in Ohio or Michigan, etc. would matter a lot.

  6. Very nice post, Andrew. I was a little surprised that you avoided direct commentary on the Electoral College, other than (to my ears) pejoratively calling it a “Ticking Time Bomb”. Given all the less-fraught metaphors one might imagine using– “Sleeping Giant”, “Final Judge”, etc– would you mind please sharing a few more thoughts on #19?

    • Brad:

      I’ve written about the electoral college in the past, and my colleagues and I have found that, on average, it is not biased against either of the two major political parties. But concerns have been raised, and continue to be raised, about the massive differences in voting power depending on whether you live in a swing state, and about the crisis of legitimacy that can arise when one candidate wins the popular vote and the other wins the electoral vote. The ticking time bomb was that this would happen, and it did. The “bomb” also would’ve blown up had Trump won the popular vote and lost in the electoral college.

      • Andrew, in its defense, it seems that the Electoral College — excepting the part about actual electors voting and being able to be “unfaithful” — mainly boils down to winner-take-all per state. Maybe there’s more to it than I understand, but if it’s the by-state winner-take-all, that doesn’t seem much different from winner-take-all per congressional district. At that lower level, we can blame gerrymandering for winning the statewide vote but not the popular vote, which is a bad thing, but does it rise to the level of a ticking time bomb?

        While direct election of the president is very appealing, it would be quite different from how government works at these other levels — as I understand it. And I think we have to acknowledge that part of this year’s disconnect between winner-take-all-by-state and the popular vote is pretty much California, which votes later than the rest of the country and provided 4 million of Clinton’s 2 million popular vote advantage. I’ve heard, though I haven’t researched it, that one of the targets of the Electoral College when it was designed was Virginia, the California of the colonies, so perhaps the winner-take-all-by-state concept is enduring.

        Part of democracy and feeling represented is spatial in nature — who are your neighbors and how closely related are you to those making decisions for you.To some degree that’s addressed by a hierarchy of layers of government: city, county, state, national. But with the transcendence of the national, my guess is we still need some kind of spatial/neighbor/local aspect to our democracy, and what we currently have is by-state national elections.

        • Wayne:

          No, winner-take-all by congressional district has much different consequences than winner-take-all by state. My colleagues and I published a paper about this a few years ago.

  7. Seriously, comparing Scott Adams to Sam Wang is unfair.

    Sam got completely wrong. Predicting the popular vote means nothing in a electoral college. Popular vote does not win elections.

    Scott got it right. It was a landslide. Trump knew exactly where he needed to win and what he needed to say to persuade those people in those areas.

    • Jack:

      Wang got it wrong. So did Adams. The vote was extremely close, not anything like a landslide. And I have no idea if Trump knew exactly etc. as you claim, but according to the news reports the Trump campaign was surprised by their victory too.

        • Jack:

          Both Wang and Adams were much too sure of themselves. Wang said 99% which came from a naive reading of the polls—a naive reading that was particularly inexcusable given that others were out there explaining that poll averages can be off by a couple percent. Adams said that Trump would win by a landslide, which was never going happen.

          Had Wang simply said, “Clinton’s at 53% of the 2-party vote in the polls with a standard error of 2%” or “Clinton is leading in the state polls in key swing states and these leads are larger than the stated margin of error but the actual historical margin of error in state polls is approx twice the stated margin of error so we can’t be so certain,” I’d have no problem with him.

          And had Adams simply said, “My subjective reading of the situation is that Trump has more support than is shown in the polls, and I think he’ll win,” I’d have no problem with that, either.

          My problem is that both of these guys expressed way too much certainty. A slight shift in votes and the election could’ve gone the other way. If so, I’d still say that both Wang and Adams were foolish in their certainty. That small hypothetical shift wouldn’t suddenly make Wang’s 99% claim reasonable and Adams’s landslide claim unreasonable. They were both unreasonable in the first place.

      • I agree with you that it was not a landslide if we compare the results in absolute terms with other elections. But the margin of victory was large if we compare it relatively to what the pundits were expecting.

  8. > Democrats vote for Democrats, Republicans vote for Republicans. It’s always been thus—what would the party labels mean

    This sentence is really quite creepy to me. Is there any other Western nation and/or democracy where their supportive voters are classified by a proper adjective/noun
    (e.g. “Democrats”). In Australia, e.g., it make sense to say “he always has voted Labor
    and always will” but to to then to say that thus “he is a ” seems really weird. The political party one currently supports: != one’s race, gender, age, or other (quasi-)essential characteristic. But in the US, it seems it can be? “What else would party labels mean?” – one would somewhat wish that that such labels (with the implications you impose) would “mean” nothing for 99% of the population; even those of very firm political convictions. My guess is that this is generally true for most other non-US countries.

    • >The political party one currently supports: != one’s race, gender, age, or other (quasi-)essential characteristic. But in the US, it seems it can be?

      You can usually predict what party someone supports by their demography. The prediction is not always correct (some Latinos voted for Trump, for instance), but it’s usually a good enough heuristic (most Latinos voted against Trump). And it makes some logical sense in a way – people don’t just vote randomly. If a person belongs to a certain demographic group, then he/she might want to support a party that helps his/her demographic group (and then rationalize his/her choice in a way to argue that his/her party is actually good for the whole country).

    • The term “Tory” arose as something to call individuals, as did its counterpart “Whig” (which was once a US political party). I believe that they are officially called the Conservative Party, but it seems more common to call them the Tories.

    • In Canada, people identify and are identified as Liberals or as Tories (voters for the Conservative party). Only the NDP (New Democratic Party, sort of socialist) has no proper noun. You’d say that someone votes NDP.

      • At the time of the American Revolution, people who were loyal to the king (against “Amexit”, if you will) were called Tories. When the rebels won, the Tories moved to Canada, and were called United Empire Loyalists (a name still sometimes claimed by their descendants.)

        • Correction:

          Should be “During or after the revolution, many Tories moved …” rather than “When the rebels won, the Tories moved …”

  9. >There’s a theory that academics such as myself are petrified of making a mistake, hence we are overcautious in our predictions; in contrast, the media (traditional news media and modern social media) reward boldness and are forgiving of failure

    Caution and being honest about uncertainty isn’t even rewarded within academia from what I can see

  10. >So there’s something off about these sweeping election reviews: even when you agree with the sentiments, it’s not clear why it makes sense to tie it to any particular election outcome.

    I wish this was more widely appreciated. Every think-piece writer is now coming up with great pronouncements on what Americans think and feel, but let’s think about the counterfactual where we give Clinton a few more votes in key states and she barely wins. The same think-piece writers would now be proclaiming the opposite, despite the views of Americans being almost the same in this counterfactual than they are now.

    We knew Trump would get a huge amount of votes months before the election. The election just gives a bit more precision and it determines the outcome, but it’s an extremely crude discretisation of the signal, not the signal itself.

  11. Andrew:

    You’re pretty hard on Bartels here. On one level, that’s fair. The shark-attack theory of voting is questionable. But on another level, you’re overlooking the main point that Bartels’ and others have been making for decades now: that most voters are massively uninformed about politics, policy, and governing, and that many of them, consequently, vote against their own economic self-interest.

    The 2016 presidential election definitely confirms this larger point.

    • WB:

      Good point. A key message of Achen and Bartels is that the facts don’t matter much to people, that voters live in their own versions of reality. The success of fake news in the recent campaign supports this point.

      It’s too bad that Bartels got sucked into the whole shark-attack, subliminal-smiley-face thing, as his errors there are a distraction from his larger points.

  12. It’s worth highlighting your comment about couching predictions in terms of probabilities. Consumers of predictions want probabilities, but in general models predict margins. But in interesting areas of the distribution, probabilities have high sensitivity to small changes in margin prediction and also to small changes in variance assumptions. This is exacerbated by the Electoral College, as well, because predicting the margin and variance in each individual state (and their covariances) is much harder than predicting the overall margin and variance (fewer, less reliable polls, more space for idiosyncratic stuff happening and so on).

    • Yes, I keep on thinking of the phrase “speculation dressed up as statistics”.

      All estimates of uncertainty are conditional on a model and in this sense they are giving us “partial”, conditional estimates of uncertainty, not the total uncertainty that we are interested in.

      In this context, model transparency — as in the ability to understand what is being conditioned upon — needs to become an important criterion by which we judge forecasts. We don’t just have a bias/variance tradeoff in our modeling; we also have a competition between simple, transparent models (the shortcomings of which we know) and more complicated models (the shortcomings of which are very hard to understand).

      • Joshua:

        I agree with you on transparency but I disagree with your alignment of simplicity with transparency. A simple method can be opaque (for example, if it relies on data that are hard to find or data coding and analysis choices that are not clearly justified), whereas a complicated method can be transparent (if all data are accessible and all model choices are open). This general point came up, in the context of election forecasts, a couple months ago in my discussion of Pierre-Antoine Kremp’s open-source poll aggregator. Kremp’s model (which in turn was based on Drew Linzer’s model) didn’t end up doing so wonderfully, and in retrospect I don’t think it fully accounted for systematic polling errors. Transparency is no guarantee of correctness. But the transparency makes it easier for us to see the assumptions and find problems.

  13. Thank you for the interesting article. Writing from the UK (where we have also had two major national votes in the last 18 months go against polling in tight races), one of my wonders is how / whether it would be possible to give polling less prominence in pre-election coverage. The media love the numbers, but are pretty good at giving them spurious certainty, whether intentionally or not. This leaves less space for discussing a whole range of other things, like why one candidate / party might or might not be better than the other.

    For an analogy, in anticipating an important sporting event, we might have a brief look at the betting odds, but we don’t obsess endlessly about them. We obsess endlessly about coaches, players, track records, tactics, etc.

    • I think one of the problems with moving away from polls is that the media organizations are usually the ones paying for the polls, so there are incentives for both pollsters and media organizations to promote media reliance on polling numbers.

  14. “sharp divide between traditional news organizations on one side and fake news spread by Twitter and Facebook on the other” Why does this blog so naively or sinisterly buy into the “fake news” narrative that the establishment media and Clinton machine desperately constructed?

  15. Viewing things from the perspective of Occam’s Razor and H.L. Mencken, consider the following statistics instantiation regarding a typical athletic/higher-ed scandal

    http://www.kansascity.com/sports/college/sec/university-of-missouri/article116694258.html

    where we find
    ##################
    When the athlete needed additional help to stay eligible this fall, he was again placed with Kumar for help with an online [!] statistics [!]class.

    “It’s taken almost four weeks for him to even understand it. … I looked at him,” Kumar said as she burst into tears, “and he was so depressed that he couldn’t do it.”

    Her voice now cracking, she repeated, “He couldn’t do it.

    “No matter how many times I told him, how many examples I gave him, he couldn’t do it. It was just (expletive) adding. That was all he had to do was (expletive) add it up, and he couldn’t do it.
    #############################
    Moral: statistics, online or not, is (are?) not straightforward either for statisticians or for the voting public and ergo, Trump triumphed.

  16. Few things never get discussed, so!
    The R’s recognized years back H would probably be the candidate and acted accordingly, in fact brilliants in terms of long range strategy. For years, almost since O took office, the right attacked and labeled her as no one to trust, a liar, way to insider, professional pol, elitist, despite many hearings, responsible for Benghazi and other international “Failures”., and never let up. Hearing after hearing kept her in the spotlight(s) as well as a very good propaganda campaign to fully discredit, assign her upper 3%-elites of NE, take your guns, supports illegals-lawlessness-more Obama,fear-loathing, anti-American Etc. I would guess this anti-slam-propaganda campaign, very well conducted-organized by R’s-Right- targeted lots of scary stuff by alt-right did more to instill “evil” image then pres campaign. Estimate 4-6 years of this, volume-less then mainstream media turned way up in last three-four years/PRIOR to the campaigns.Seeds of fear, distrust, NE elite-(evil) media supported, right wing nut cases on extreme fed rumors, gun control-take out Second Amd, no lie to small, nor fact checking allowed as it was tool of the elite, not the commoners. Actually a very well run strategic long range plan, while D’s basked in the glow of O’s programs, the right attacked for years, going after state legislators, the House-Congress who openly stated, we do not care, our job is block O, and there was more then a bit of dabbling in racism, subtlety, entertain and arouse the R’s racists while maintaining image of “not so, we ain’t racists (but)”, many of whom chuckled when at State of Union, to alleged, horror(of both sides) O was absolutely run down when shouted at as a liar, fatally for his image to many racists-etc, as often said in our history, “Sure put this uppity //// (you fill in)) in his place etc”. O/D Chair/D’s should have, but did not did not respond directly, publicly degraded “O (D’s) put in his place”. The stage was then set to assign H-D Party as various versions of O the wimpy, for any number of reasons. Core issue while O was popular,he was not respected, by association nor was H/D’s by the commoners.

    The D Party for past 15-20 years battered by Roving’s, lost local and state elections time after time, and did nothing but gather what might be called, NE Establishment with assist from celebs from CA. at fund raisers, big city politics and DNC Chair simply directed more of same, despite Rove types strategy starting to take the states, as that was where further right wins could be started. DNC Chair Deb, found it better for photo ops in WH, powerful D pol’s, Celebs, urban NE fund raiser’s, then note the tide was changing and D’s were already up to necks in strategic loss’s. The D Party seemed determined to ignore voters outside of any city of less then half million or more, and rural USA did not exist. The commoner of cities less then 100K was only addressed if D had sure winner, but less and less really contested as “Big urban-minority-educated(R’s organizer implies upper 3%-“elites”) were the true D Party. This was very strategically well used against D’s and H for years before this election, but none, more so D’s seemed to care or worse understand the frustrations that were building, across the nation beyond the NE Golden Triangle: NYC-Boston-DC, and CA, plus/minus a few miles. The triangle feed D’s, seemingly shaped policy for national items, and simply told the rest of USA, we know what is good for you, look at jobs coming back (at reduced wages-no benefits) as it good because the data we say so. Images of welfare programs,money, for inner cities dominated stories, but never went to areas outside where poverty was a bad or worse. Never saw much D activity on real jobs, really affordable HC for ALL. Pastern’s of votes in DC went “Bill passed when few D’s crossed aisle” (for well off) but “Party line vote-lobbyists killed Bill” when orientated for commoners wages-HC-jobs-education, or the disgusting response by D’s, “We got best we could get blah blah” which often was not close to “best” for “other 97%”. The latter’s became D Party/Chair policy, process and procedures as the commoners got more angry and frustrated, for years, seemingly totally ignored, desperate. D’s Chair-Party at highest levels simply ignored the reality of rural areas, who along with workers seethed with anger as dollars-programs-“justice” etc poured into the big cities (who would supposedly then vote D) while poverty, low end jobs, wide spread corruptions, decent broadband-cell, lack of HC, poor education, etc grew for most.. and not to well reported was the real welfare of same folks-areas, which was bigger then most rural s (we are proud folks etc) admitted or thought, or pol’s mentioned,largely unrecognized or reported.

    Fact is H did not loose the race, DNC Chair-Party lost it over past decade or perhaps decades while R’s simply lied cheated and spread rumors, fear mongering with dash of targeted hate about “them” (a moving target, but a history provides, needed for hate, fear, bigotry, intolerance etc), racism and propaganda as needed, Goebbels would have been proud. So now we have dozens of experts etc , same ole predicable crowd pseudo-intellectually explaining another miss by experts. Now appears that various types of Gobble/Trump/Right and D”isms” will rules the day(s) and in a kind of perverted justice, the commoners will once again cheer as their wallets are lifted, the price of their worship at the alter of decades of a failed governments of the people, whose lives may get worse. Guess best actual measure of integrity of DC was Bill passed requiring “disclaimers-etc” AKA “small print” to protect/inform the citizens, but along with this glorious protection, Bill also read small print need not be readable..so goes it for the commoners

    I guess summation is H did not loose election, D Chair-Party did, and had for years, they wore not political clothes, rather lots of make up, but could not, did not, get out to where the fabled children of Emperor, could tell them they were politically naked.

  17. “I came up with 19 lessons learned.”
    To me the real lesson is the suspicion of social science methodologies. Usually estimates can’t be validated/invalidated so vividly as election polls. But the issues with the methodology like differential nonresponse are likely to crop up in all kinds of research.

    • Arnott:

      Fake news and rumors are not new. But, as I wrote, the siloing of news consumption seems to have reached a new level in the past few years. If it was just the National Enquirer running that ridiculous story this summer that Hillary Clinton’s weight had ballooned to 289 pounds, I’d say, sure, nobody takes that rag seriously. But there seems to be much more of this than there used to be.

  18. Andrew,

    I’m confused by number 6 on the Latino vote.

    1. When you say that our forecast at Latino Decisions “should’ve been suspect, given that it contradicted the organization’s own polls,” to what polls are you referring? Our forecasts relied almost entirely on our polling, and I used a simple model to estimate how the undecided (or unsure) would vote. Our final forecast, made several days before Election Day was 79-18 for Hillary vs. Trump among Latinos:
    http://www.latinodecisions.com/blog/2016/11/03/latino-electorate-on-track-for-historic-turnout-in-2016/

    The results of our Election Eve poll (a combination entrance/exit poll of those who voted early and those who claimed they would definitely vote the next day) was came out exactly at 79-18 (more accurate than we did in 2012):
    http://www.latinovote2016.com/app/

    We can certainly argue over the virtues of our election eve poll and the Edison exit polls with respect to capturing Latino voting, but I’m not sure where you got the idea that our forecast was inconsistent with our polling. Speaking of the exit polls…

    2. If you haven’t already, I’d encourage you consider the merits of our challenges to the Edison-Mitofsky exit polls. I think we could have a nice conversation about the merits and disadvantages of our two approaches, but I’m baffled by the persistent assumption that the exit polls are somehow a gold standard with respect to Latino votes. The last time Edison publicly addressed this, over a decade ago, they conceded that their polls are “not designed to yield very reliable estimates of the characteristics of small, geographically clustered demographic groups. These groups have much large design effects and thus larger sampling errors.” Their current strategy seems to be to not engage in a debate at all, not release information on the precincts they poll nor explain how or whether they weight their results to Census demographic information on the nationwide Latino population, and just coast on their reputation.

    http://www.latinodecisions.com/blog/2016/11/10/lies-damn-lies-and-exit-polls/
    https://www.washingtonpost.com/news/monkey-cage/wp/2016/12/02/donald-trump-did-not-win-34-of-latino-vote-in-texas-he-won-much-less/?utm_term=.df27f831870c
    https://www.washingtonpost.com/news/monkey-cage/wp/2016/11/11/in-record-numbers-latinos-voted-overwhelmingly-against-trump-we-did-the-research/?utm_term=.76040b78f88f
    http://www.mattbarreto.com/papers/exitpoll.pdf

    Coverage by people not working for LD:
    https://www.washingtonpost.com/blogs/plum-line/wp/2016/11/15/why-the-exit-polls-are-wrong-on-latino-votes/?utm_term=.94419df9e65c
    http://www.politico.com/story/2016/11/latino-groups-exit-polls-donald-trump-231268

    I look forward to your thoughts.

    Best,
    Justin H. Gross
    Latino Decisions and UMass Amherst

    • Justin:

      I did not read the original documents on this one. I was just quoting the news report that stated, “The latest tracking poll from Latino Decisions shows Republican nominee Donald Trump with 16 percent support, versus 74 percent for Hillary Clinton. Looking ahead to November, the group expects that electorate to cast the vast majority of its votes for Clinton, 82 percent to 15 percent for Trump, which would be the most lopsided total in history.” I said that the 82-15 ratio contradicted the organization’s own polls, because in that quote it said that the organization’s poll gave a 74-16 ratio, which corresponds to a Trump share of the two-party vote of 16/(16+74) = 18%, not the 15% that you get by taking 15/(15+82). It just seemed odd that the forecast was more extreme than the poll that was reported.

      I do agree with you, though, that exit polls have problems, and I’d guess that Trump’s actual share of the two-party vote among Latinos vote was somewhere between the 15% from that forecast and the 30% reported by the exit polls. I agree that exit polls are no gold standard.

  19. A comprehensive list. Thanks!

    One thing, though. “Fake news” is not new, and there’s no evidence that there was more of it in this election than in the past.

    Before the internet, fake news was passed over the back yard fence, at bowling, around the water cooler at work, in church groups, at the hairdresser’s, in bars, and so on and on.

    Fake news is a non-story.

    • Greg:

      One possible novelty is the way that fake news is siloed by party and ideology. That’s why I discussed it as part of political polarization. Backyard fence, bowling, water cooler, etc., you get a mix of people in the audience. Less so perhaps with partisan sources. But I’m not sure; it’s worth further study.

    • Well, today there are more ways to disguise fake news as real news. That must have something to do with the situation. Also, the weight of numbers plays a larger role. On Twitter, a piece of news might seem real merely because it comes from someone with thousands of followers, or because it has been retweeted hundreds of times. At the hairdresser’s, you just hear a few people (at most) talking. Their words are no weightier than anyone else’s, and you do not confuse them with the NYT.

    • Couldn’t disagree more. Social media if it doesn’t provide anonymity at least provides distance and makes it easier for people to share seriously whacked out stuff. Then the fact that others have shared it and liked it lends credence to it for some people. Before the internet there was much more of a reputational risk to spreading fake news, which helped keep a lid on it. Also, the audience was limited, if crazy Aunt Martha said Hillary was a pedophile there was little chance of it going ‘viral’ as everyone Martha could share her info with probably already viewed her as nuts. Those barriers are gone with the internet. All of the crazy Aunt Martha’s can find each other, like each others whacked out stories, and once they get started these internet memes, like wildfires, can create their own weather.

    • Greg,

      Fake news is a non-story. MSM put out more Fake news stories than any other source, whether given by Hillary’s Campaign for the News of the day to distract from her many damaging Real News stories the MSM refused to vet.

      So who is the arbiter of what is fake and what is Real? MSM, no! The American people are smart enough to know the truth and that is the problem. People skipped the MSM to get their news from other sources to find out the truth. It’s simple. The Government is trying to control News and therefore Control Thought.

      This is Not America. We The People, have already spoken about Fake News. MSM need to change, and they never will, their ideology will not allow it. FB, Twitter, run by Liberal’s are going to deside what is Fake News to keep it off their sites. Nothing new, they do it already!

  20. AG:

    Sure, conventional wisdom says news is “siloed.” But the best evidence (from passive metering data) doesn’t support the idea: https://dl.dropboxusercontent.com/u/663930/GuessJMP.pdf

    On social media, see: http://pablobarbera.com/static/barbera_polarization_APSA.pdf

    And even if there were evidence of selective exposure/polarized media consumption habits, that wouldn’t prove an *effect* on attitudes, voting behavior, etc.

    Of course, there’s more to be said about fake news during 2016. Stay tuned….

    -AG

    • what worries me at first read is that aol.com, finance.yahoo.com, and (especially) msn.com visits outnumber all others combined (Table 2). these are all (appropriately)scored as moderate. but are people getting news from these websites when they visit, or are these like default home pages that pop up every time a browser is opened? (I’m pretty sure msn was default for explorer for a while.) do consumption habits look as moderate w/o these sites? sorry if I missed a footnote or something.

      • dl:

        Good question. I’ve also (but not in this draft) built a classifier via supervised learning to separate “hard news/politics” articles from everything else, including home pages. Same pattern holds.

  21. The really, really scary thing to me, as if fake news wasn’t scary enough, is the effect it’s having on the perception of “real news” (aka news). Just today I was reading the comments section on the WaPo article on the CIA report that Russia was indeed trying to rig the election for Trump. If you read the comments, you’ll see how many people now have a free license to simply not accept what is clearly a “real news” (aka news) story. They just label it fake as if it is coming from some backstreet blogger (you know, or Breitbart). It’s really scary for our democracy.

  22. I will Scare you Evan. What makes this story about Russia, Fact? Who or what makes it true, the fact that a Secret CIA Report said so? Maybe not all, but most of you speak from one ideology, all you do is speak to others just like you so nothing is questioned anymore. There is No investigative reporting anymore. There are No Pure Journalist left. It’s hard work. Most News outlet just Copy and Paste each other. That makes it Real News? Repeat it enough and it must be true and anyone that questions it is propagating Fake News?

    This Fake News, is a Non-Story is it just a distraction? For what? It’s not clear what the distraction is or what it’s for? It can’t be as simple as Liberal’s need something to do? They don’t want to say one word that could possibly make President Elect Donald J Trump look like he may be doing anything right so they need something for their “Real News” Reports, right? So, Hillary surfaces to give a speech. How sad was that? Hillary and her campaign were the Sources of Fake News stories, Damn, so many I won’t even start on that garbage. Right, you think I’m crazy, listening to lies! That’s News? The Media needs self reflection, but that won’t happen. Hillary continues to what? Lead her party? Speak out for the American People? Obstruct? And Obama, said how Trump shows him respect but Obama acts like a Child. These are Very Embarrassing Times, and you don’t have a clue what is happening around the block from you. So Sad!

    Evan, it was the Comment Section, People HAVE Free License to simply not accept the context of the News Story. We woke up this election and are Simply asking question. That’s how a Democracy works, or have you forgotten? There was so much BS put out as Real News, no thanks, I’ll investigate all “Real News” Stories myself. I don’t blindly follow. Only fool’s do that, in this day and time.

  23. Minor correction: Bannon was not an “executive” at Goldman, he was a junior employee for a few years. His title of “vice president” is held by roughly a third of Goldman employees. Of course, since Trump has brought some actual Goldman executives onboard, your broader point stands.

  24. And one thought re: elite influence on opinion. All you have to do is look at post-nomination opinion among Republicans on trade & favorability toward Russia to see a pretty clear Lenz/”Follow the Leader”-type story playing out.

  25. “#6: Demography is not destiny.”

    NO! The ONE thing this election was about, and you STILL get it wrong! Demographic ARE destiny. They always have been, and they always will be. Blacks, Latinos, and Muslims vote Democrat by HUGE margins because they party pushes their ethnic interests – take from Whites and give to “my people.”

    The Republican establishment – the Washington Generals of American politics – have been insisting, coached by the Democrat media heads offering their “honest advice,” that “the GOP isn’t going to win unless we moderate and appeal to the Black & Latino (& Muslim, gay, who knows what next) vote.” And, golly gee willikers, their brilliant strategy lost over and over and over again, and each time, they said “Wul, clearly the problem was that we didn’t go left enough!”

    Demographics ARE destiny. Americans – White people – just finally decided to wake up and DO something about it before it was too late.

  26. I did a search on the word “email” and found no hits. Hillary’s private email system and Podesta’s hacked DNC email got a lot of play and it is surprising these things, which bore on the honesty of the candidate and the party, are not mentioned. L

  27. The Clinton campaign’s ground game was pretty awful. Look at the number of her field offices in swing states compared to Obama’s and then look at the voter turn out in 2016 compared to 2008 and 2012. Although Trump’s field office numbers were even lower, I don’t think you can conclude that ground game is not important. Ground game may be more important for Dems to get out the vote than for Republicans.

Leave a Reply to Corey Cancel reply

Your email address will not be published. Required fields are marked *