Weekend Roundup: The Post-Debate Debate, Part 2

The Post-Debate Debate, Part 2

After the second and likely final debate between the two major parties’ presidential nominees, American voters who watched the matchup overwhelmingly believe that Vice President Kamala Harris outperformed former President Donald Trump. However, this has not yet translated into a meaningful change in the overall race.

When asked to rate each candidate’s performance, voters allocate an average of 59 points out of 100 to Harris and 35 points to Trump (Figure 1).

Overall, nearly a majority of voters nationwide allocated 80 points or higher for Harris while just 23% scored the debate at the same level for Trump (Figure 2). On the opposite end of the spectrum, 56% feel strongly that Trump had a bad night, while 28% feel the same about Harris’ performance. Roughly 1 in 5 voters think it was not a straight-up “win” for either candidate. 

Debate performance is one thing, but votes are another. We know that this country is incredibly divided by political tribalism, and there are very few people living between the two camps. This fact coupled with the understanding that it has only been a few days since the debate, massive movement is certainly not the expectation.  Harris has been on the upswing since becoming the presumptive nominee while Trump has also been increasing his support since the middle of August, albeit at a slower rate. So far, the dynamics of the race have stayed relatively stable, although Harris has bumped up slightly over the past 48 hours while Trump’s support ticked down ever so slightly (Figure 3).

Looking at how the support breaks down, not surprisingly, there has certainly been a shift for Harris 48 hours after her debate compared to where President Biden was 48 hours after he debated with Trump (Figure 4). Two days after the debate in June, 36% of voters were highly likely to vote for Biden, indicating there was an 80 to 100 percent chance that they would vote for him in November. Currently, 39% of voters say the same for Harris. When comparing these two points in time, the biggest difference is at the negative end of the spectrum. At the end of June, 54% of voters said there was little to no chance that they would vote for Biden. Currently, 48% say the same of Harris. 

The concentration of those definitively supporting Trump has not changed when comparing the 48 hours after each debate (Figure 5). At the end of June 35% of voters indicated that they were very likely to support the former President and the exact same percentage say the same today. The percentage of those who indicate there is very little to no chance that they will support Trump dropped by about a percentage point when comparing these two points in time.

What is interesting to see is how the supporters of the two candidates engaged with the debate. Overall, those who watched the whole debate are more likely to be strongly committed to voting for Harris than Trump (Figure 6). Nearly half of the viewers who tuned in for the entirety of the debate (48%) are strongly committed to supporting Harris (the Commitment Index), while 42% register as a Trump supporter on the same Index. An additional 2% say they are definitely going to be supporting a third-party candidate.

Those who watched part of the debate or didn’t tune in at all are more likely to be strongly behind Trump. Among those who watched some of the debate, 48% are strongly committed to Trump, while 40% are in the same range for Harris. Those who didn’t tune in to any of the debate skewed even more towards Trump with 48% committed to his candidacy compared to the 35% strongly committed to Harris. Overall there was a close to even level of non-committed voters in each category. About 8% of the viewers who watched the whole debate are non-committed to any candidate. Nine (9) percent of those who watched some of the debate are non-committed, as were 7% of those who did not watch any of the debate.

Since we are engaging with the same individuals over time we are able to use present-day information to look backward and analyze past data. While correlation does not necessarily equal causation, we can see that those who watched the entire debate have been significantly increasing their support for Harris since the middle of August (Figure 7). At the same time, Trump's numbers among this cohort have been fluctuating up and down over the same period. In the past 48 hours, Harris’ numbers have increased slightly, while Trump’s remained flat.

For those who watched some of the debate (Figure 8), Trump’s support has been much stronger going back to early August with his level of support generally growing. His level of support has dipped after the debate by a little more than a percentage point. Harris’ support has been flat to negative over the past five weeks but jumped about 2.5 points after the debate.

For those who didn’t watch Tuesday’s debate at all (Figure 9), there is an interesting dynamic in the data. Trump’s support among this cohort had been relatively high through most of August (right around 50%). His support levels then increased steadily over the end of August into early September. But his support dropped quickly starting the day before the debate and now stands four points lower than on September 9th. Harris’ support had been slowly increasing through August but then dipped in early September. She then moved in a positive direction a day after Trump’s support started increasing and continued on this path the 48 hours after the debate.

Turning to everyone’s favorite topic – the swing states – Harris and Trump are tied in Nevada, and Harris is still up in both Michigan and Arizona after the debate Tuesday night (Figure 10). Nevertheless, the race is tight with Harris enjoying a 2-point lead on average. This is an increase since late August when we showed both candidates with an average support of 45.

Arizona has tightened over the same time period and now stands tied. In late August, Harris was at 47 and Trump at 43. Nevada was a 1-point race at the end of August according to our data and now is tied. Both candidates have similar support distribution in Arizona, with 44% of Arizonans highly likely to vote for Harris and the same number highly likely to vote for Trump (Figure 11). The Vice President has a leg up among voters who aren’t entirely sure about their vote yet: 3% of voters are somewhat likely to vote for Harris (60-79 points), while 0% of voters are somewhat likely to vote for Trump in Arizona.

On the negative end of the scale, 50% of Arizona have ruled out voting for Harris, while 51% feel the same way about Trump. For Harris, her committed voters increased by 1 point over the past two weeks while those rejecting her candidacy grew by 2 points. For Trump, the committed growth was 3 points, while his rejection dropped by 2 points. Overall a better two-week period for the former President than the current VP.

Michigan has been better for Harris over the past fortnight. Since late August, her commitment grew by 1 point and her rejection dropped by 3 points. Over the same time period, Trump's commitment and rejection held at the same levels. However for those in the middle 6% used to rate their chances of voting for Trump at 40% or higher in late August and that now stands at 3%. 

Diving into some of the demographics, Harris is up by 6 points against Trump among White women, she’s up by 55 points among Black women, and she’s up by 7 points among Latinx/Hispanic women (Figure 12). A month ago Harris was up by 8 points among White women and 60 points among Black women. Among Latinx/Hispanic women, she was down by 15 points, a massive swing. 

On the other hand, white men allocate 3 more points on average for Trump than for Harris, Black men allocate 7 more points for Harris than for Trump, and Latinx/Hispanic men allocate 15 more points for Trump than for Harris. In mid-August, Trump was up by 2 points among white men and trailed among Black men by 1 point and 17 points among Latinx/Hispanic men.

The commitment indexes for men and women tell a similar story (Figure 13). While Harris sees a higher commitment index among women who are white, Black, and Latinx/Hispanic, Trump’s commitment levels among men vary by race and ethnicity. White men lend a 2% higher commitment index for Trump than they do for Harris but Harris receives an 8% higher commitment index from Black men compared to Trump. Latinx/Hispanic men, on the other hand, are 11% more committed to voting for Trump than to voting for Harris, in contrast to Latinx/Hispanic women, who are 4% more committed to Harris.

Currently, the groups with the largest concentration of non-committed voters are Black men and women and Hispanic men. Groups we will keep a close eye on. We are less than two months out from this year’s election, and if we’ve learned anything from the past few months, anything could happen. How this election ends will be determined by the turnout of key voters in key states, and Trendency is here to track them all the way up to November 5th.