Historically, generic ballot polls have undercounted Republican support in congressional elections. And so at the moment, with what seems like widespread unhappiness with the Trump administration and the status quo, it was a bit surprising the other day to see that the Politico/Morning Consult poll had the two parties tied — 40 to 40 percent — for partisan preference ahead of the 2018 midterm election.
Today, a slightly less surprising result, although still one that’s better for the GOP than I would have expected: A four-point percent Democratic lead among registered voters in the new NBC/Wall Street Journal poll:
A new poll from NBC News and the Wall Street Journal shows that 47 percent of registered voters say they would prefer a Democratic-controlled Congress after the next midterm elections, while 43 percent say they would like to see Republicans in charge. While Democrats have a four-point edge at this very early stage, their advantage still falls short of the double-digit lead they held on the same question shortly before their big electoral victories in 2006 and 2008.
I would expect the Democratic lead to grow by quite a bit, just because we’re going into Trump’s first midterm and no one expects House Republicans to do particularly well, even if they hold the House. (The Senate is kind of a different story just because of which seats are in play.) The fact is, Democrats are going to have to post some better numbers if they expect to win anything in 2018, let alone take over either House of Congress. You can see why if you look back at some generic ballot polling history from this poll in particular, especially among registered voters:
- Days before the 2016 election, the last NBC/WSJ generic ballot poll had Republicans losing the generic ballot by two points among registered voters and winning it by three points among likely voters. Republicans won the popular vote for House by six points. So the margin was eight points off in Democrats’ favor.
- The last NBC/WSJ generic ballot poll before the 2014 election had Republicans four points behind with registered voters and one point ahead with likely voters. The national result? A six-point Republican victory in the popular vote for House. Ten points off.
- The last NBC/WSJ poll before the 2012 election showed Democrats leading by four points among registered voters. The result? A one-point Democratic victory in the national House popular vote. Just three points off.
- The last poll before the 2010 election had Democrats leading by two points among registered voters but trailing by six among likely voters. Republicans won by seven points. That’s 13 points off with the registered result, and four with the likely voter result.
- This particular poll wasn’t conducted in 2006. But the polls that were conducted showed Democrats winning by as many as 20 points (CNN), 18 points (NYT) and 13 points (Fox News) immediately before the election. Democrats won the House popular vote by eight points that year. (To be fair, the last Washington Post poll nailed it, at least with its likely voter results.)
The point is that for one reason or another (and this isn’t the only poll for which this is the case), generic ballot polls tend to favor Democrats, sometimes by quite a bit, such that Republican victories turn out to be a surprise.
There are still 19 months for Democrats to make their case, and no shortage of material in the Trump administration to help them do it. But at this moment in time, Republicans cannot be unhappy with the generic polls that have been released recently after a few high-profile failures in Congress. So far, they’re hanging on better than you’d expect in this environment.
