What to Make of All These New and Often Conflicting Generic Ballot Polls?

Amy Walter
January 25, 2018

If you are confused about what to make of the latest slew of polling data about control of Congress in 2018, you are not alone. On the so-called generic ballot question which asks voters if they’d support a Democrat or Republican for Congress in 2018, Democrats have either lost ground or retained their double-digit lead. As my colleague Charlie Cook wrote, there’s a big difference between Democrats winning the national House vote by 5 points (as CNN showed) or 12 points (as the ABC/Washington Post found). The results of the ABC/Post poll predict a tsunami in November while the CNN poll projects a significant, but not overpowering wave next fall. Three polls — CNN, Marist/NewsHour/NPR and NBC/Wall Street Journal — showed Democrats going from double-digit leads on the generic ballot in December to single-digit leads in January. The Quinnipiac poll also found that Democrats lead shrunk, though it remains in double digits (from +15 in December to +13 in mid-January).

In looking through the data, the cause is independent and “soft” GOP voters moving toward the GOP. In the Marist poll, for example, Republicans picked up 10 points from independent voters between December and January (35% to 45%), while Democrats lost 9 points among these voters. In the CNN poll, a generic GOP candidate gained 6 points from independents between December and January (32 percent to 38 percent), while the generic Democrat lost 7 points among these voters. There wasn’t much movement among independent voters in the NBC/Wall Street Journal poll — which asks voters which party they’d like to see control Congress. Republicans picked up a point and Democrats lost 3 points among independents in that survey. Even the Quinnipiac survey showed Democrats had lost 14 points among independents between mid-December and mid-January.

Among so-called “soft” Republicans — those Republicans who are not as closely aligned or identified with the party, Republicans made significant gains in both the CNN and Marist polls (7 points in CNN and 6 in the Marist poll); a sign that the GOP base is “coming home.” That is likely due to the passage of the tax bill, a lack of GOP infighting and/or just the fact that Trump himself was pretty quiet over the holidays. The NBC and Marist poll found a drop off in support for the Democratic candidate among “soft Democrats” — a sign, perhaps, that with Trump off the front burner, they are not quite as reflexively anti-GOP as when Trump is front and center.

If this pattern feels familiar — Republicans flee when Trump is under fire, but return to the fold when the news is about anything other than Trump – that’s because it is. We saw this pattern play out time and again in 2016. Trump would say/do/tweet something controversial. His numbers would drop. But, as soon as Hillary or emails got back into the news and/or Trump stayed disciplined, the race would tighten up as Republicans and Republican-leaning independents came back into the GOP fold.

Which party’s candidate would you vote for in your congressional district?

If 2018 election were held today, which party’s candidate are you most likely to vote for in your congressional district?

What is your preference: Congress controlled by Republicans or by Democrats?

If the election were today, would you want to see the Republican Party or the Democratic Party win control of the United States House of Representatives?

However, both the ABC/Washington Post and Pew polls showed Democrats holding onto the double-digit lead on the generic ballot that they enjoyed in 2017. In November of 2017, the last time ABC/Washington Post poll asked the generic ballot question, Democrats had an 11 point lead. It was D+12 in their January survey. In July, the only time Pew asked a generic ballot question in 2017, Democrats held a 13 point lead. In January, Democrats held a 12 point lead in the Pew poll.

But, dig under the topline numbers and even some of the polls that show a tightening race for Congress also show Democrats with an enthusiasm advantage. For example, while the January CNN poll found an increase in the percentage of Republicans (especially soft Republicans) who said they were “enthusiastic” about voting, Democrats still held a 12 point advantage in enthusiasm (54 percent to 42 percent). Among those who said they were “very” enthusiastic to vote this November (think of these as the folks who’d walk over broken glass to get to the polls), Democrats hold a 13 point lead.

Change in percentage of Democrats and Republicans enthusiastic about voting

The CNN poll found Democrats with a 6 point lead on the generic among registered voters, but that lead grew to 15 points among those voters who said they were “very enthusiastic” about voting in November (56 percent to 41 percent). In other words, among people who are registered to vote, Democrats have a small lead. Among those who say — today — that they are enthusiastic about voting – Democrats have a big lead.

The January ABC poll also found a similar enthusiasm advantage for Democrats. “ Democrats hold a 15-point advantage over Republicans on the generic ballot among voters who say they are “absolutely certain” that they will vote, slightly larger than their 12-point edge among registered voters.”

The Quinnipiac poll found that 89 percent of Democrats were “very motivated” to vote, compared to 85 percent of Republicans. Not a huge advantage for Democrats.

The danger in looking at national polling, of course, is that this isn’t a national election. Democrats may have an enthusiasm advantage, but it also may not be distributed evenly. In other words, stronger turnout in college towns and urban areas doesn’t help Democrats pick up seats in more exurban or less dense suburban districts. The NBC/Wall Street Journal poll looked at the generic ballot by congressional district and found Republicans gained ground in GOP-held districts (going from +8 in December to +14 in January). Democrats also improved their performance in Democratic-held seats (from +32 in December to +35 in January).

However, the 14 point advantage held by Republicans in their own seats is basically in line with the advantage they held in 2010 (+15) and 2014 (+18) midterm elections. Meanwhile, Democrats 35 point advantage is 18 points higher than it was in those Democratic seats in 2010, 12 points higher than it was in 2012 and 6 points higher than their showing in 2016. In other words, it looks as if Democrats may “waste” their enthusiasm advantage in districts they already hold.

In recent days, however, there has been a great deal of hand-wringing by Democrats that the mishandling of the shutdown/DACA deal will dampen Democratic turnout this fall. There’s no doubt that liberal activists are frustrated with the lack of progress on DACA. It’s also true that there are going to be lots of twists and turns to come on DACA before the March deadline. As such, the jury is out as to whether this frustration in January turns into disillusion in November. However, it’s also true that the biggest driver of Democratic enthusiasm is Donald Trump. And, we can say with almost 100% certainty that between now and November he will say/do something that will enrage and engage Democrats.

Going forward, it’s going to be important to weave polling data with real-life voting results. In 2017, for example, the Democratic enthusiasm advantage wasn’t just theoretical — it was apparent in every single special election contest. My colleague David Wasserman crunched the numbers and found that in the six 2017 special elections, Democrats have won an average of 71 percent of the votes Hillary Clinton received in their states/districts, while Republicans have won an average of just 55 percent of the President Trump’s votes. This disparity in enthusiasm also showed up further down ballot. The Hill’s Reid Wilson found that Democrats saw their vote share increase in 27 of 37 contested special and regularly scheduled state legislative races in 2017. And, this year, Democrats picked up GOP-held district in Wisconsin that Trump had carried with 56 percent of the vote. The Democrat took 36 percent of the votes Hillary Clinton got in 2016 while the Republican got just 20 percent of Trump’s vote. In other words, you can follow the generic ballot question, but also follow the actual elections taking place. A special election on March 13th in Pennsylvania’s 18th CD will be the next “real” test of the generic ballot and enthusiasm gap.