Election Update: National Polls Show The Race Tightening — But State Polls Don’t

Want these election updates emailed to you right when they’re published? Sign up here.

Hillary Clinton moved into a clear polling lead over Donald Trump just after the Democratic convention, which ended on July 28. Pretty much ever since, the reporters and poll watchers that I follow have seemed eager to tell the next twist in the story. Would Trump’s numbers get even worse, possibly leading to the first double-digit victory for a Democratic presidential candidate since 1964? Or would Trump mount a comeback? As of last Tuesday, there wasn’t much evidence of an overall shift in the race. Trump was gaining ground in some polls but losing ground in a roughly equal number of them.

Since then, Trump has gotten some slightly better results, with national polls suggesting a race more in line with a 5- or 6-percentage-point lead for Clinton instead of the 7- or 8-point lead she had earlier in August. But state polls haven’t really followed suit and continue to show Clinton with some of her largest leads of the campaign. Trump received some decent numbers in Iowa and Nevada, but his polls in other swing states have been bad.

Overall, Trump has gained slightly in our forecasts: He’s up to a 15 percent chance of winning the Electoral College in our polls-only model, up from a low of 11 percent a week ago. And he’s at 25 percent in polls-plus, up from a low of 21 percent. But the evidence is conflicting enough that I don’t think we can rule out a larger swing toward Trump or, alternatively, that his position hasn’t improved at all.

Let’s start with those national polls. In the table below, I’ve listed every national poll that we’ve added to our database since Tuesday and how it compared to the previous poll from the same pollster, if there was one.¹

A number of these polls show no change. But where there have been shifts, they’ve been toward Trump, particularly in the USC Dornsife/Los Angeles Times tracking poll, which now shows a 2-point lead for Trump after having Clinton modestly ahead before, and in Pew Research’s most recent poll, which has Clinton with a 4-point lead as compared with the 9-point lead Pew showed her with before the conventions.

You can, of course, pick apart the individual polls if you like. The USC/Los Angeles Times poll makes some unorthodox methodological choices; I happen to like some of these choices and dislike others, but overall, they produce a poll that’s significantly more Trump-leaning than other pollsters. And I’m not sure anyone should be crowing about Zogby Analytics polls, which have been highly inaccurate historically. But there are ways to adjust for these things, and they don’t obscure the fact that the trend in national polls has mostly been toward Trump.

State polls tell another story, however. Here’s every state poll we’ve added since Tuesday:²

As I wrote earlier, Iowa and Nevada have been relative bright spots for Trump, with Clinton leading only narrowly even in post-convention surveys. But those states have only 6 electoral votes each, and Trump’s numbers are bad pretty much everywhere else. Since Tuesday, for instance, he’s gotten polls showing him down 6 points in Ohio, 9 points in Florida and 11 points in Virginia — and only tied with Clinton in Georgia. I suppose you can count polls showing Trump ahead by double-digits in Indiana and Mississippi as good news for him, since they’re states that could conceivably have gone to Clinton in a landslide. Then again, other polls this week showed competitive races in Missouri and Texas. Our model thinks that these polls are consistent with Clinton continuing to hold a lead in the mid- to high single digits: You probably wouldn’t get a set of results like these if she was up by only 5 percentage points nationally.

Moreover, these state polls show highly favorable trend lines for Clinton, where they’re available. Among the six polls that had previously surveyed the same state, Clinton gained ground in every one, with an average swing of 6 percentage points toward her. A caution: The average shift is inflated by a Quinnipiac poll of Colorado which found Clinton up 8 points; Quinnipiac had implausibly showed an 11-point lead for Trump when it surveyed the race in November. Even without that poll, however, Clinton’s average gain is 4 percentage points, still pretty good.

There are a couple of further nuances that explain some of the differences. Most of the recent national polls are daily or weekly tracking polls conducted online or via automated surveys, and these tracking polls have generally been a relatively friendly group for Trump. He hasn’t fared well recently in traditional telephone surveys, by contrast, with one or two exceptions like his not-so-bad result in the Pew Research poll. Also, looking at the trend lines doesn’t quite make for an apples-to-apples comparison, because most of the national polls have surveyed the race multiple times since the conventions, while the state polls haven’t. It’s plausible that Clinton is polling slightly off her post-convention peak, as the national polls suggest, but ahead of where she was for most of the pre-convention period, as the state polls suggest.

Still, our model perceives an increasing conflict between state and national polls. Polls-only calculates a national polling average, which has Clinton up by 6.2 percentage points, down from a peak of 8.0 percentage points on Aug. 15. But it also infers an estimate of the popular vote from state polls, and that continues to have Clinton ahead by 7 to 8 points. The 1- or 2-point gap between these estimates doesn’t matter much for now, since Clinton is comfortably ahead either way. But it could become pertinent if the race tightens; it was pertinent in 2012, when state polls continuously (and correctly, it turned out) showed President Obama in better shape than national polls did.

I’m not going to get too much more into the weeds for now. The past week was pretty light for polling and sometimes these differences resolve themselves as you accumulate more data. Maybe this week, we’ll get a couple of national polls showing Clinton 11 points ahead, but others showing her only tied with Trump in Pennsylvania and Wisconsin, or something.

In terms of interpreting our forecasts, though, you should know that our models mostly rely on state polls to estimate the level of the race, whereas they lean heavily on input from national polls to estimate the trend. Thus, polls-only has Clinton ahead of Trump by about 7 percentage points nationally, a result more in line with the most recent state polls than with the most recent national polls. But it also detects a modest trend toward Trump, something the national polls show but the state polls don’t yet.