Skip to main content
ABC News
Democrats Lead The Race For Congress — But The National Environment Will Probably Change

The 2020 electoral environment currently looks pretty good for Democrats.

As of Sunday evening, Democrats lead Republicans by roughly 8 percentage points in FiveThirtyEight’s generic ballot tracker, just a point less than their lead on Election Day 2018.

We know that early generic ballot polling has predictive value in both midterm and presidential cycles. These surveys, which ask voters which party they plan to support in the next congressional election, correctly foreshadowed big Democratic gains in the 2018 midterm elections as early as June 2017. They also have trended close to the final national popular vote for president.

But how confident should Democrats be that this lead will hold? We took a look at the movement in generic ballot polling in presidential cycles from 1996 to 2016. In the last six months of a cycle, the polls often shifted by meaningful amounts: There was, on average, a 4-point range between the largest and smallest generic ballot margins.1

The generic ballot can still shift before Election Day

The average generic-ballot polling margin at different points in the last six months of a presidential election cycle, since 1996

Time until election day
Cycle Six Months Four Months Two Months 10 days margin range
1996 D+8 D+6 D+4 D+4 4pt
2000 D+2 D+3 D+2 D+3 1
2004 D+7 D+6 D+1 EVEN 7
2008 D+12 D+11 D+7 D+9 4
2016 D+5 D+6 D+2 D+1 5

For the average margin at six, four, and two months from Election Day, we included all polls fielded around the relevant time point (180 days out, 120 days out, 60 days out). We also calculated an average for each pollster in a given period so a pollster wasn’t overrepresented, then averaged those averages. Figures are rounded.

Source: Polls

Four points may not sound like a lot, but that sort of shift in voter preferences could make or break the election for candidates in the closely fought seats that will determine control of the House. What’s more, there appears to be a tendency for the margin to narrow by Election Day, so Democrats should probably expect their lead to shrink in the coming months. In four of the six cycles we analyzed, the difference between the two parties shrank from 6 months out to 10 days before the election. Only in 2000 and 2012 was there little change in the overall electoral environment.

Of course, there’s no guarantee the generic ballot will move much this time around, but it’s still worth thinking about how a change in the national environment could make the 2020 House race more of a toss-up. Take 2016: Republicans went from trailing by 6 points in the generic ballot four months before the election to trailing by just 1 point in the final 10 days. Notably, they ended up narrowly winning the House popular vote, too — 48 percent to 47 percent. This speaks to the consequential shifts generic ballot polls can experience between now and November.

Currently, FiveThirtyEight’s tracker puts Democrats up by about 8 points. If that were to dwindle to a little less than 4 points by November, that might still be a sufficient lead for Democrats to retain control of the House. But it could also give Republicans a better shot at winning some Democratic-held House seats that don’t look especially vulnerable right now. There are some sizable GOP pickup opportunities, as Democrats are defending 30 districts that President Trump carried in 2016 (Republicans just need to win 18 seats to retake the House).2

Moreover, because the generic ballot also says something about the overall electoral environment, a narrow Democratic lead could signal that Trump has a better chance of winning reelection. In 2016, Trump lost the national popular vote by 2.1 points to Hillary Clinton, but he won in the Electoral College because he performed better in key swing states like Michigan, Pennsylvania and Wisconsin. In fact, Wisconsin was the tipping-point state in 2016 and Trump won it by 0.8 points, meaning that the decisive contest in the Electoral College was about 3 points more Republican than the country as a whole. So if Trump once again has an advantage in the Electoral College, a smaller Democratic lead in the generic ballot could point to a more competitive presidential election.

The generic ballot average has been relatively stable in 2020, but recent history suggests it could still move and make for a competitive battle for control of the House as well as the White House. In other words, Democrats still have reasons to be optimistic about November, but a small swing in the electoral environment could be the difference between Democrats maintaining their House majority and Republicans capturing one.

How Justin Amash could affect the general elections | FiveThirtyEight


  1. We calculated the average margin between Democrats and Republicans at four different points in time — six months out from Election Day (about where we are now), four months out (in the summer before the conventions), two months out (around early September) and over the final 10 days of the campaign. And at each point (except for the final 10 days), we included all polls fielded around the relevant time point. For instance, at the six month mark, that meant all the polls 15 days before and the 15 days after. We did this to account for noise in the polls. We also took an average of the margins in each pollster’s surveys, then averaged those averages. That way, one pollster couldn’t skew the numbers if it released a disproportionate number of polls.

  2. Assuming they hold onto four vacant seats in heavily Republican districts that the party previously controlled.

Geoffrey Skelley is a senior elections analyst at FiveThirtyEight.