In two recent posts I discussed forecasts for the November midterms, including predictions about the potential for a Blue Wave and the odds of Republicans maintaining control of the Senate. But those polling numbers, interesting as they are, may not capture an aspect of the forthcoming blue wave. That aspect was touched on recently by both Sen. Ben Sasse on the latest episode of The Remnant and Harry Enten writing at CNN: Turnout sometimes just overwhelms the polling.*
It’s important to understand how pollsters operate. They don’t just call up 1,038 random people and the chips fall where they may. Pollsters balance the sample they look at based on what they think turnout will look like on Election Day. Using dissimilar turnout models, two different pollsters, working from the same data set, could come up with different numbers. So pollsters make educated guesses as to what the turnout will be in a various elections, based on voter enthusiasm.
In wave elections, pollsters underestimate voter turnout in favor of the party benefiting from the wave.
Enten went back to 2006 and looked at polling data at this point in the cycle as compared to performance on Election Day. The party that won the national House popular vote overperformed polling at this point by about three points. So if Democrats are currently ahead in the generic congressional ballot by an average of D+8.3 then, well, you can do that math, but it looks like they’re going to have a big night.
Enten also looked at individual district polls and found that the effect is larger in district polls. In 2006, the average district poll had Dems down by 1.5 points, and they ended up winning by, on average, four points. So if you look at individual polls, add another 5.5 points.
And that’s just the average! I think it’s fair to say we are in extraordinary political times. Despite a hot economy, presidential disapproval is over 50 percent. The one issue on the ballot this November, the issue that’s driving Democratic enthusiasm, is the President. If take a look at FiveThirtyEight’s map forecasting every congressional district, there are lots of districts marked as leaning Republican that could get swept up in a big wave of Democratic turnout.
At the same time that Democrats are eager to get to the polls, Republicans may not be as motivated. A recent poll showed that a majority of Republican voters don’t think the Democrats will take control of the House, including 57 percent of those who strongly support the President. (This perception is no doubt bolstered by talk about a Red Wave.)
And maybe there will be a Red Wave. Maybe, despite historic presidential unpopularity, the historical trend for the party controlling the White House to lose seats in midterm elections, Americans’ preference for divided government, and polling data, the Republicans might keep control of the House.
But I wouldn’t bet on it. In fact, I’d bet the other way: that there will be an enormous Democratic turnout, and the predictions of modest Democratic gains will turn out to be incorrect, and that it will be a big night for the Democrats.
*That may be a direct quote or close paraphrase of Sen. Sasse. I wasn’t able to find the precise words skimming back through the podcast.