Next Tuesday will mark the last day of voting in the 2024 general election. While we are feeling very certain that it will be many days after November 5th before we have an announced winner in the Presidential election (and potentially control of both Congress and the Senate), the likely outcomes are still murky. What feels most likely to us is that Vice President Kamala Harris will win the popular vote by a comfortable margin. What is less certain is who will win the Electoral College. Every outcome, from Harris or former President Donald Trump winning resoundingly to a too-close-to-call result in the Electoral College, is in the realm of likely.

To close out this election, the team at Trendency has updated the graphs we discussed on our October 17th webinar (if you missed it, there is a recording here). An interactive version of these graphs is posted on our website here (click here).

The National Horserace

Since late July, the Trendency numbers have shown Harris with an advantage on the average allocation question. As a reminder, we do not ask questions in a binary fashion (e.g. are you voting for Harris or Trump?) but instead ask about the chances of voting for each candidate, allowing our panelists to give more nuance to their answers and allowing us to understand how opinions are shifting, even if the shift is not from one choice to another. This is not an exact apples-to-apples comparison with what you will see reported from snapshot surveys, but it provides a more comprehensive look at how people are feeling about their choices, as well as their likely future behavior.

Across our national panels, Harris’ average allocation has been, for the most part, on the rise since she became the presumptive Democratic nominee back in late July. More recently, her support grew steadily through the beginning of October and has been increasing at a much slower pace over the past few weeks. During the past three months, Trump’s support has been more steady, with a slight increase in his average support over the past two weeks as we get closer to Election Day.

While traditional methods of research have pointed to a stagnant homestretch, our look at the data paints a very different picture. Both candidates have seen shifts in their support level among key demographic groups. Since October 15th, Harris has increased her level of support among most cohorts; her biggest increases have been among Black men, White women, and voters under the age of 40. She also has seen notable gains among White men. The only group where her support shrank during this period was among Latina voters.

During the same time period, Trump’s change in support has been a little bit more of a mixed bag. His biggest growth has been among Latino and Latina voters, while his biggest drop has been among Black men. Support among White women dropped slightly while White men increased by about a point. When it comes to age, Trump’s support grew by about 2 points on average among voters under 40, but dropped slightly among all other age groups.

Comparing the support for the two candidates, the biggest delta is among Black women followed by Black men. Older voters, as well as White women, currently have a near double-digit difference. Harris has a higher allocation among most of these groups, with the exceptions being among White men (where Trump has about a one-point advantage) and Latino voters (where his advantage is roughly seven points, on average).

When dealing with averages, the overall results can oftentimes mask what is really happening in the data. In the case of partisan politics, an average of 50 does not mean everyone is equally allocating between the two choices (aka completely undecided); indeed, most are strongly supporting one candidate while also being strongly opposed to the idea of supporting the other. As we can see in the chart below, almost all voters are at the bookends with very few in the middle reporting that they are unsure of their vote. For Harris, currently, 45% of voters are allocating 80% or higher on their likelihood to support her candidacy, while 47% state that there is very little to no chance that they will support her (under 20% chance). For Trump, his strong support is currently at 38%, with 54% of voters saying that they are unlikely to support his candidacy.

Looking at these distributions, there are a few clear takeaways. First, Harris has an advantage when it comes to both the strong supporters and having a lower concentration of strong detractors. Second, both candidates have a higher negative concentration than the positive, meaning there is more motivation to vote against a candidate than there is voting for a candidate. Third, even though both candidates are underwater in terms of the positive versus negative motivations, only one candidate has a majority of voters ruling out voting for them nationally: Trump. Fourth, and finally, there are very few voters in the middle who are leaning, but not sold on either candidate (roughly 8 percent).

Looking at the last two weeks, the Harris campaign has done a good job of solidifying her support. Her strong support increased from 39% on October 15th to 44.5% on the 30th. At the same time, the concentration of voters who are opposed to the idea of supporting her candidacy barely moved (down 0.3 percentage points).

During this same time period, Trump also increased his concentration of strong supporters. On October 15th, 35% of voters said there was an 80% chance or higher that they would be voting for the former President. Two weeks later, that number increased to 38.2%, an increase of 3.2 percentage points. At the same time, his negative numbers increased by 1.5 percentage points (going from 52% to 53.5%).

Looking at the strong supporters, what we refer to as the Commitment Index, both campaigns have done a good job of solidifying their support over the last two weeks. Since mid-August, the Harris campaign has done a remarkable job of increasing the concentration of voters in the committed range. In August, 37.3% of voters were at this level, and currently the number stands at 44.5%. This 7.2-point increase represents a 19% increase over the past two and a half months. For the Trump campaign, his committed voters increased from 35.5% in mid-August to 38.2% at the end of October, resulting in a 2.7 point increase or 7.6% increase over the last 75 days.

On the opposite side of the spectrum, we see a difference in the two candidates over the past two and a half months. There are two main motivators in a decision: you can either support your choice, or you can reject the idea of the choice you are not making. This double negative is not making many English teachers happy but is basically the idea of voting for a candidate versus voting against the other candidate. The motivation for voting against a candidate is what we refer to as the Rejection Index. A candidate would ideally have this number as low as possible. In mid-July, before Harris became the nominee, and while President Biden was still running for re-election, roughly 55% of voters rejected the idea of supporting Harris and Trump. After becoming the nominee, Harris’ Rejection Index dropped to 50%, while Trump’s dropped to 52.9%. Since that point, Harris’ Rejection Index has dropped slowly, but steadily, ending at 47.3%. Trump’s, on the other hand, remained steady through mid-October but has seen an increase over the last two weeks, ending at 53.7%.

If we combine these two measurements and subtract the Rejection from the Commitment, we can understand where the motivation is mainly coming from for voters. In an ideal world, a candidate would like to see the difference in positive territory.

With the election coming to a head next week, neither candidate has moved into positive territory in this measurement. Between the increase in strong support and the decrease in the rejection of her candidacy, Harris has seen a slow but steady rise in the difference between those strongly supporting her and those strongly opposed. She currently stands at -2.84 an increase of 5.52 points over the past two weeks and 7.81 since September 15th. During this same time period, Trump’s number increased by 1.54 points over the past two weeks and 1.57 over the past 45 days.

2024 vs. 2020

While each election is unique and we know the past is not necessarily prologue, we do find it interesting to look back and compare patterns leading up to elections. Since one of the candidates is the same as last time, it is very interesting to see the differences in voter sentiment. For Trump, he is ending the 2024 campaign with a slightly higher level of support than a similar point in time in the 2020 election, although the path to this point is certainly different.

In 2020, Trump was outperforming Biden until late July and even though his support dropped below Biden’s, Trump remained relatively steady until mid-September when he dropped a few points and then held steady until the election. This go-around, Trump’s support has been slightly more volatile through October and currently stands 1.3 points higher than in 2020.

Harris has followed a similar pattern to Biden in 2020, albeit her average support has always been lower than Biden’s was four years ago. Similar to Trump’s numbers, the last two weeks have shown some volatility compared to Biden’s in the last election. For both candidates, there is good news and bad news. For Trump, he is performing slightly higher than he did in 2020 (good news) but is running behind Harris similar to Biden in 2020 (bad news). For Harris, she is running ahead of Trump (good) but is underperforming Biden (bad).

Our research approach involves short check-ins with our audience members regularly to get a better understanding of how opinions and behavior are changing over time. This creates longitudinal data that allows us to not just look at what is currently happening, but also use past or current data to better understand the landscape we are dealing with. Many of our national panelists have been with us for over four years, which in turn gives us the ability to see how voting patterns have shifted among specific individuals.

After major elections, we will typically ask our audience members if they voted and which candidates they supported. While not everyone will share this information, most do, and since we ask this question within days of the election it gives us an accurate view of what happened versus asking someone what they did four years ago and then using that as a stand-in for behavior.

Looking at the 2020 Biden and Trump supporters in our national panels, we find that very little has changed for the vast majority of voters. Average support has remained basically steady since July among this subset of voters (the purple and blue lines) while the vast majority of 2020 Trump voters are planning on voting for him again - and the same is true for 2020 Biden voters and Harris. On October 30th, 3% of 2020 Trump voters indicated that they are definitely planning on voting for Harris, while 94% say there is little to no chance of that happening. Among 2020 Biden voters, 3% say that they are definitely voting for Trump and 93% say there is little to no chance of them supporting Trump.

Swing States

While it looks like Harris has a definitive advantage at the national level, that is not how we pick our Presidents in the United States. This election is going to come down to a handful of swing states (more accurately, a few swing counties in a handful of swing states) and at the state level, the numbers are less clear.

We looked at three states where we have statewide panels set up: Michigan, Arizona, and Nevada. Michigan was a typical blue state but went for Trump in 2016 before going back to Biden in 2020. Nevada has gone to the Democratic candidate for the past four elections but has been very close when Trump is on the ballot. Arizona went to Biden in 2020 and was the first time the Grand Canyon state went to the Democratic candidate since 1996. For all three states, Harris is ahead in the average allocation, but the other measurements are not as clear-cut.

Michigan

In the Great Lake State (aka the Mitten), Trump had a surprising win in 2016 winning by less than half a percentage point. This was the first time a Republican presidential candidate won since 1988. In 2020, the state went back to the Democrats, with Biden winning by a little less than three percentage points.

Currently, Harris has an incredibly slim lead on the average allocation (0.4%) after having a stronger lead for August and September.

When it comes to the Commitment Index, both candidates are ending up at about the same spot that they were in mid-August, but the path has certainly not been a straight line. Trump had a strong increase in those strongly supporting his candidacy in late September and early October. However, over the past two weeks, strong support dropped just as quickly. For Harris, the opposite was the case, with mid-October marking a low mark for her base support (although the overall averages did not move much).

While the strong support showed fluctuations, the Rejection Indexes of the candidates did not follow a reflexive pattern for either candidate. For Trump, while his strong support slowly increased from mid-August to mid-September, the rejection of his candidacy dropped more rapidly with the pattern continuing through to mid-October. The last two weeks, however, have not been great for the former President on this measure, with a majority of voters in Michigan now rejecting the idea of voting for him.

Harris, on the other hand, saw her strong support hold steady, dip, and then recover. On the negative side of the equation, the rejection of her candidacy rose during the late summer but then has been slowly declining since mid-September.

Arizona

Moving to the Southwest, the candidates have been moving in similar patterns for the past month. Harris has put a little bit of daylight between herself and the former President in September. However, the Trump campaign closed the gap in early October and even jumped ahead for a few days. Over the past week, the Harris campaign regained the lead and is heading into Election Day with a 1.3-point advantage on the overall allocation average.

For the Commitment Index, there has been a similar pattern to Michigan in Arizona except the movement for the candidates was about a month earlier. Late summer was a very good time period for Trump with a three-point increase (3.6 points) in the concentration of voters strongly supporting his candidacy. Since the middle of September however, the level of strong support has been slowly declining ending the campaign, ending just slightly higher than the middle of August (0.44 points).

After the drop to just over 40% of voters strongly supporting her campaign in the middle of September (40.46%), Harris’ Commitment has grown to 45.72% ending 2.28 points higher than Trump on this measurement.

On the negative side, unlike in Michigan, the movement was similar in nature to the positive. Heading into Election Day, Trump is about where he was in mid-August with 51.61% of voters rejecting the idea of supporting his candidacy (it was 52.05 in August). Harris is ending at 49.22% rejecting her candidacy. On the plus side for Harris, this is a lower number than Trump and is below 50%. On the negative side, it is a higher number than where she was in mid-August.

Nevada

In the Silver State, we see yet another pattern over the past few months. On the average allocation, Harris had been increasing her support through mid-October although not in a steady pattern. Trump had not moved the numbers at all during most of September but then stated increasing in early October. Over the past two weeks, we have seen a downward trend for Harris and an upward trend for Trump. However, both candidates ended up roughly in the same place they were in the middle of October.

Interestingly, we see Harris with a slight advantage on the average, but Trump with a slight advantage on the Commitment and Rejection Indexes. Nevada is also the only state of the three swing states where the commitment to the candidates is hovering around 40% (a good 5 to 6 points lower than the other states).

On the negative side, it is also the only state where both candidates are below 50%.

State Conclusions

In the States, it is tough to get a clean read. As of today, every measurement is leaning toward Harris in Michigan, albeit by a close margin. For Arizona and Nevada, there is a mix of who has the advantage going in. Harris has a bigger lead in the average but in Arizona support for Trump is higher among Republicans and there is a higher level of strong support for Trump among Democrats than strong support for Harris among Republicans.

In Nevada, the overall average goes to Harris, but Trump has an advantage on the underlying measurements. However, Harris has an advantage over the most recent responses from panelists as well as base party support and crossover support. The only thing that seems unlikely in either state is a comfortable win for either candidate.

In Conclusion

The last three elections have had Donald Trump as the nominee and this brings a level of unpredictability to the data. Trump is not a traditional candidate and voters have not been following familiar patterns. There also have been large differences in the landscape adding further noise to the data. It is not surprising that analysts and prognosticators have been hedging bets like crazy over the past month or two. A close election or a blowout on the Electoral College is equally probable, which is a first in my recollection.

Based on our data there is one thing that seems to be certain, and that is that Harris will win the popular vote. This is not exactly a groundbreaking reading of the tea leaves, but it does seem like it should be by a comfortable margin. Only time will how this ends up with the Electoral College, but we’re in for an interesting ride.