Suddenly the polls are tied: The RealClearPolitics.com average of recent polls in the presidential race shows Donald Trump with 43.4 percent and Hillary Clinton with 43.2 percent. All four of the polls sponsored or co-sponsored by television news networks show both candidates within statistically indistinguishable percentages: Fox News and ABC/Washington Post have Trump narrowly ahead, while NBC/Wall Street Journal and CBS/New York Times have Clinton narrowly ahead.
These results are not of course the final word. The contest for the Republican nomination is over, and only a (to me, anyway) surprisingly small share of self-identified Republicans are refraining from supporting Trump. The Democratic nomination race is still on, with Bernie Sanders and his supporters increasingly bitter. Clinton’s support among self-identified Democrats is already about as high as Trump’s among self-identified Republicans, but presumably it could go higher when the Sanders candidacy expires.
But it’s worth keeping in mind that many Sanders supporters self-identify as independents, which may account for her poor ratings among them. So there’s some upward potential for Clinton there too.
That said, her performance is weak. Even as Barack Obama’s job approval has risen, Clinton’s standing has eroded. In the Fox poll, for example, her favorable/unfavorable numbers are even worse than Trump’s.
Note that more than 10 percent of registered voters are still reporting themselves as undecided. Given both candidates’ high unfavorable ratings, this seems to represent negative cross-pressure. Examining the candidates’ standing among demographic groups in the NBC/WSJ and CBS/NYT polls, for example, I find that Trump is roughly equaling the numbers for Mitt Romney in the 2012 exit poll among Democratic-majority groups like women and young voters, while Clinton is still falling short of Obama’s percentages among these groups. Similarly, Clinton is roughly hitting the Obama 2012 exit poll percentages among Republican-leaning groups like whites and elderly voters, while Trump is falling short of the Romney numbers among these groups.
We shouldn’t make too much of these comparisons: Error margins among subgroups are pretty high, and in a particular poll if one subgroup comes in unusually Republican, another tends to come in unusually Democratic.
That said, there may be a clue here to the dynamic in a race between two negatively regarded candidates. One possible consequence: A lot of potential voters are reluctant to choose either one, and may not vote. In that case there is a premium on the campaign that can maximize turnout of its supporters.
