Skip to main content
Menu
Have The Democratic Debates Moved The Polls At All?

Wednesday brings the fifth Democratic primary debate, which means you’ll be hearing a lot about potential “game changing moments” and whatever else Beltway-types are obsessing over.

The only problem: For the most part, this cycle’s debates haven’t really been all that game changing.

That’s not always the case. There is evidence, after all, that primary debates can move poll numbers — particularly in the summer and fall of the year prior to the primary, which is where we are right now. Still, the 2020 Democratic campaign has been surprisingly stable so far. Since he entered the race in April, former Vice President Joe Biden has led most national polls, and although Sen. Elizabeth Warren has generally trended upward in the polls, her rise has been slow and steady and not necessarily closely tied to any one debate performance. This isn’t to say the primary debates haven’t moved the needle at all, though. Sen. Kamala Harris, for instance, shot up in the polls after her performance in the first debate, but the bump didn’t last.

But we wanted to better understand just how much of an effect these primary debates have had on the candidates’ standing in the race so far. Which debates have moved the national polls the most, and which candidates have gained or lost the most? And more importantly, have these changes really shaken up the candidate field or have most of the same contenders continued to hold the top spots in the primary race?



To answer this, I did two things. First, I averaged all of the national polls that were in the field two weeks before a debate and two weeks after,1 and from there, I calculated the difference between a candidate’s pre-debate and post-debate average. Next, I also averaged the absolute change of the top 10 candidates’ pre-debate and post-debate averages2 to see how much the polls had moved in one debate. (In other words, if one candidate’s average went up 2 percentage points and another’s went down 2 points, I treated both their polling averages as having moved 2 points.) This helped me understand which of the four debates had shaken up the national polling picture the most so far. Now, to be clear, some of the poll movement might not be entirely attributable to the debates. Still, given the amount of attention these events receive, it’s reasonable to think they have a lot to do with the shifts we observed.

And as you can see in the table below, the first debate not only saw the most movement in the polls — 2.4 points, on average — but it also saw the biggest drop or gain for any of the candidates. After that first debate, Harris gained 8.3 points and Biden lost 6.5 points.

The first Democratic debate shifted polls the most

Average change* in national polls and the candidates who moved the most after each 2019 Democratic primary debate so far

Debate Avg. Change Biggest gainer Biggest loser
June 26-27 +/- 2.4 Harris +8.3 Biden -6.5
July 30-31 0.9 Warren +3.0 Harris -3.0
Sept. 12 1.1 Warren +4.4 Sanders -2.0
Oct. 15 1.0 Buttigieg +1.6 Warren -3.5

*Average change in national polls evaluates the average absolute change in polling averages before and after each debate across the 10 candidates with the highest post-debate polling average. Polling averages were calculated using national polls conducted during the two weeks before and two weeks after each debate.

Source: Polls

Compared to the first debate, though, the last three debates just haven’t corresponded with as big of a move in the polls — about a one-point difference, on average. So there is evidence that the Democratic contest has, in fact, been pretty stable so far, or at the very least, the debates haven’t dramatically shaken up the race. But that got me thinking, is what we’re seeing in 2020 all that different from any other cycle?

For comparison, I crunched the numbers for the first four Republican presidential primary debates in 2015 (the field was crowded then, too)3 And in that case, I found that there was more movement in the primary after debates — but not that much more.

2015′s debates corresponded with shake ups a bit more

Average change* in national polls and the candidates who moved the most after the first four 2015 Republican primary debates

Debate Avg. Change Biggest gainer Biggest loser
Aug. 6 +/- 2.3 Carson +4.4 Walker -4.0
Sept. 16 2.4 Fiorina +7.0 Trump -6.8
Oct. 28 1.4 Cruz +3.8 Fiorina -2.6
Nov. 10 1.1 Trump +3.8 Carson -2.9

*Average change in national polls evaluates the average absolute change in polling averages before and after each debate across the 10 candidates with the highest post-debate polling average. Polling averages were calculated using national polls conducted during the two weeks before and two weeks after each debate, except for the period after the Oct. 28 debate and before the Nov. 10 debate, which was based on polls conducted in the same intervening 12 days.

Source: Polls

The biggest shifts for any GOP candidate happened in the second debate when former Hewlett-Packard CEO Carly Fiorina stood out for her sharp attacks and rebuttals against now-President Trump, who stumbled in the polls after his performance, although he continued to hold on to his lead.

The first debate didn’t see any individual candidate move up or down as much as Fiorina or Trump in the second debate, but it still had a number of candidates whose poll numbers moved at least 2 points up or down. The debate was critically damaging for Scott Walker, who was in third nationally at 11 percent heading into it, but things went so poorly that he dropped out of the race after the second debate. There isn’t a parallel to Walker among the leading Democrats running this time around, which is one reason why the GOP cycle had seen more movement in the polls at this point.

Yet don’t make too much of this difference, as both primaries had fairly stable polling before and after the first four debates, at least for candidates at the top of the polls, which is where you’d expect to find those most likely to win their party’s nomination. So far in the 2020 Democratic race, the same five candidates have been at the top of the polls before and after each debate — Biden, Pete Buttigieg, Harris, Sen. Bernie Sanders and Warren. And in the 2016 GOP contest, there was also a group that stayed pretty consistently at the top of the polls — Jeb Bush, Ben Carson and Trump — with Trump consistently in the lead like Biden has been this cycle.

Of course, there are still more opportunities in the 2020 cycle for the polls to move in response to the debates — with tomorrow night as a prime example. Currently, the Democratic National Committee has plans for eight more debates — two more in 2019, and six in 2020. But time might be running out for these debates to make much of a difference. Debates tend to have the most impact early on when voters aren’t as familiar with the candidates, Democrats are already four debates in, voters are paying plenty of attention to the race and many of the leading candidates are already quite well known. It’s possible that the slight shifts in the polls after the second, third and fourth debates might actually be the norm this cycle.

Footnotes

  1. In the case of the June and July debates, which both took place over two nights, the time periods I used for the polling averages were the two weeks before the first night and the two weeks after the second night. If the same pollster conducted more than one poll during the pre-debate or post-debate periods, I averaged the results of their surveys.

  2. I limited my analysis to the top 10 candidates in a debate to avoid including candidates polling near zero, as the change in their pre-debate and post-debate average is so small that it artificially lowers a debate’s average change.

  3. The Oct. 28, 2015, and Nov. 10, 2015, Republican primary debates were separated by 12 days, which complicated my methodology a bit. In that case, I used polls conducted in these 12 days, and not the two weeks before the November debate or the two weeks after the October debate.

Geoffrey Skelley is an elections analyst at FiveThirtyEight.

Comments