Why Her Poll Decline Could Be Good News for Clinton
To say Hillary Clinton had a rough week last week would be an understatement. After declaring that half of Donald Trump’s supporters – that is, some 30 million Americans – fall into a “basket of deplorables,” she was seen collapsing at a 9/11 memorial ceremony. After a day’s worth of conflicting information, her staff finally declared that she was suffering from pneumonia and would take the following few days off.
It’s unsurprising, then, that her poll standing took a hit. What was a three-point lead in the RealClearPolitics average as of September 11 was a 0.9 point lead one week later. Her solid 340-198 electoral vote lead had subsided to a narrow 293-245 advantage, with her leads in North Carolina and Nevada – worth a combined 21 electoral votes – being less than a point.
On its face, this looked like atrocious news. I’m not sure that assessment is correct. To be sure, Clinton did not want the polls to tighten. At the same time, this was a particularly awful series of news cycles for her, while Trump had managed to go over a month without reprising some of his more polarizing statements, such as his flap with the Khan family, who lost a son in Iraq. We would expect a big swing in the polls, and there was one.
But it did not put Trump over the top. A week in, she still leads by 0.7 points in the four-way RCP average, and 0.9 points in the two-way average. She maintains a lead in the Electoral College, and while North Carolina and Nevada appear to be close, her lead in the next-most-Republican state, Virginia (which would put Trump over the 270 mark), is 3.5 points.
In other words, a truly terrible news cycle was still not enough to put Trump ahead. In a strange way, that’s good news for Clinton. The rhythm of the campaign is such that news cycles are almost guaranteed to swing the other way, and, well, Trump has a history of giving them an assist. The bad news cycle can also cause Republicans to break through likely-voter screens while Democrats become less likely to answer the polls; this “differential response” issue explains a lot of the ebb and flow of campaigns. But we have to wonder: If this didn’t catapult Trump to an electoral lead, what could?
If you’re truly looking for bad news for Clinton, it’s probably found more in the long-term trend lines. Since bottoming out on Aug. 9 at 39.9 percent of the polls, Trump had already risen to 42.9 percent by September 11, cutting a Clinton lead of 7.9 points to 3.1 points. It was a gradual slide for her, suggesting that this was “real” movement and not simply differential response. The Los Angeles Times tracking poll, which probably shows a Republican bias because it weights to 2012 polling results, would be largely immunized from differential response due to that weighting, and likewise shows movement toward Trump.
So in the end, I think we need to wait and see. If a couple weeks from now the polling is still close, without some major intervening event to fix it, then we have some pretty good information that the shift was “real,” and that the equilibrium of the race is (at the least) a very close election, where short-term variations in the news cycle really could put Trump over the top.
On the other hand, if this drifts back to a three- or four-point race, we’ll have evidence more consistent with a small but real Clinton lead being the equilibrium, with events having the potential to knock it off that equilibrium in a couple of directions. Of course, there are still a large number of undecideds who could ultimately break one way or the other down the stretch, something that won’t really come into focus until the very end.