Freaking Out

Dear Media, Stop Freaking Out About Donald Trump’s Polls 

By Nate Silver

Lately, pundits and punters seem bullish on Donald Trump, whose chances of winning the Republican presidential nomination recently inched above 20 percent for the first time at the betting market Betfair. Perhaps the conventional wisdom assumes that the aftermath of the terrorist attacks in Paris will play into Trump’s hands, or that Republicans really might be in disarray. If so, I can see where the case for Trump is coming from, although I’d still say a 20 percent chance is substantially too high.

Quite often, however, the Trump’s-really-got-a-chance! case is rooted almost entirely in polls. If nothing Trump has said so far has harmed his standing with Republicans, the argument goes, why should we expect him to fade later on?

One problem with this is that it’s not enough for Trump to merely avoid fading. Right now, he has 25 to 30 percent of the vote in polls among the roughly 25 percent of Americans who identify as Republican. (That’s something like 6 to 8 percent of the electorate overall, or about the same share of people who think the Apollo moon landings were faked.) As the rest of the field consolidates around him, Trump will need to gain additional support to win the nomination. That might not be easy, since some Trump actions that appeal to a faction of the Republican electorate may alienate the rest of it. Trump’s favorability ratings are middling among Republicans (and awful among the broader electorate).

Trump will also have to get that 25 or 30 percent to go to the polls. For now, most surveys cover Republican-leaning adults or registered voters, rather than likely voters. Combine that with the poor response rates to polls and the fact that an increasing number of polls use nontraditional sampling methods, and it’s not clear how much overlap there is between the people included in these surveys and the relatively small share of Republicans who will turn up to vote in primaries and caucuses.

But there’s another, more fundamental problem. That 25 or 30 percent of the vote isn’t really Donald Trump’s for the keeping. In fact, it doesn’t belong to any candidate. If past nomination races are any guide, the vast majority of eventual Republican voters haven’t made up their minds yet.

It can be easy to forget it if you cover politics for a living, but most people aren’t paying all that much attention to the campaign right now. Certainly, voters are consuming some campaign-related news. Debate ratings are way up, and Google searches for topics related to the primaries1 have been running slightly ahead of where they were at a comparable point of the 2008 campaign, the last time both parties had open races. But most voters have a lot of competing priorities. Developments that can dominate a political news cycle, like Trump’s frenzied 90-minute speech in Iowa earlier this month, may reach only 20 percent or so of Americans.

We can look deeper into the Google search data for some evidence of this. In the chart below, I’ve tracked the aggregate share of primary-related searches in the 2008 and 2012 presidential cycles, based on the number of weeks before or after the Iowa caucuses.2 As you can see, public attention to the race starts out quite slow and only gradually accelerates — until just a week or two before Iowa, when it begins to boom. Interest continues to accelerate as Iowa, New Hampshire and the Super Tuesday states vote, before slowing down again once the outcome of the race has become clear.

To repeat: This burst of attention occurs quite late — usually when voters are days or weeks away from their primary or caucus. At this point in the 2012 nomination cycle, 10 weeks before the Iowa caucuses, only 16 percent of the eventual total of Google searches had been conducted. At this point in the 2008 cycle, only 8 percent had been. Voters are still in the early stages of their information-gathering process.

When should you start paying attention to the polls?

But maybe you don’t trust the Google search data. That’s OK; exit polls like this one have historically asked voters in Iowa and New Hampshire when they made their final decision on how to vote. These exit polls find that voters take their sweet time. In Iowa, on average, only 35 percent of voters had come to a final decision before the final month of the campaign. And in New Hampshire, only 29 percent had. (Why is the fraction lower in New Hampshire than in Iowa? Probably because voters there are waiting for the Iowa results before locking in their choice. In fact, about half of New Hampshire voters make up their minds in the final week of the campaign.)

SHARE OF IOWA WHO DECIDED ELECTION
                           > 1 MONTH OUT 1       WEEK TO 1 MONTH OUT       FINAL WEEK
2004 Democrats                    30%                      27%                  42% 
2008 Republicans                 28                          31                      40 
2008 Democrats                    49                          24                      27 
2012 Republicans                 32                          21                      46 
Iowa Average                        35                          26                      39 

SHARE OF N.H. VOTERS WHO DECIDED ELECTION 
                   > 1 MONTH OUT 1       WEEK TO 1 MONTH OUT       FINAL WEEK
2004 Democrats                    26%                      19%                     54% 
2008 Republicans                  29                         22                         50 
2008 Democrats                     34                         17                         48 
2012 Republicans                  28                         26                         46 
New Hampshire Average       29                         21                         50 

By comparison, voters decide much earlier in general elections. In Ohio in 2012, for example, 76 percent of voters had settled on Mitt Romney or Barack Obama by the end of September. This is why it’s common to see last-minute surges or busts in nomination races (think Rick Santorum or Howard Dean), but not in general elections.

If even by New Year’s Day (a month before the Iowa caucuses, which are scheduled for Feb. 1) only about one-third of Iowa voters will have come to their final decision, the percentage must be even lower now — perhaps something like 20 percent of voters are locked in. When you see an Iowa poll, you should keep in mind that the real situation looks something more like this:3

CANDIDATE SUPPORT IN IOWA
                             Undecided 80% 
Donald Trump      5 
Ben Carson          4 
Ted Cruz              3 
Marco Rubio        2 
Jeb Bush              1 
Carly Fiorina         1 
Mike Huckabee     1 
Chris Christie       1 

So, could Trump win? We confront two stubborn facts: first, that nobody remotely like Trump has won a major-party nomination in the modern era.4 And second, as is always a problem in analysis of presidential campaigns, we don’t have all that many data points, so unprecedented events can occur with some regularity. For my money, that adds up to Trump’s chances being higher than 0 but (considerably) less than 20 percent. Your mileage may vary. But you probably shouldn’t rely solely on the polls to make your case; it’s still too soon for that.