Biden Holds Narrow Lead Over President Trump
National • Presidential Poll • Sept. 15, 2020
The race for President appears to be a repeat of 2016.
The Joe Biden-Kamala Harris ticket is ahead of the Donald Trump-Mike Pence ticket by a 48-45% margin among registered voters.
Among likely voters, the lead shrinks to 1.2%, which is similar to 2016 when Hilary Clinton won the popular vote by 2.1%.
| Registered Voters | Likely Voters | ||
|---|---|---|---|
| Biden | 47.9% | 47.5% | +1.2% |
| Trump | 44.8% | 46.3% | |
| Other | 7.3% | 6.1% | |
| Est. Turnout Reg. Voters | 86.4% |
Turnout among registered voters will be similar to 2016 when 86.8% of registered voters showed up at the polls. However, the composition of likely voters this year appears to be very different.
There’s been a lot of talk about shy Trump voters and the accuracy of the 2020 polls. We spent an excessive amount of time analyzing these results and delayed this release to spend more time identifying who will be voting this year.
We found a substantial block of older, white, male voters without a college degree who plan to vote for Trump in 2020. Historically, female voters turnout at a higher rate than men; however this year looks like a 50%/50% split.
Also, we tend to believe many public polls are too narrowly defining who will be voting this year. Polls with tight or narrow likely voter screens or that rely on prior voter history to identify “likely voters” could be missing a block of older male voters who do not always show up on election day, but plan to this year.
Methodology
The PollSmart MR online poll was conducted by Targoz Market Research, an independent research company with sample provided by Prodege, a leading market research panel.
Interviews were conducted from September 2-6, 2020, among a sample of 1,200 registered voters. A subsample of 1,039 has been defined as likely voters (LV), with a margin of error of ± 2.98 percentage points.
The estimated margin of sampling error for registered voters is ± 2.77 at the 95% confidence level. More information about the PollSmartMR polls can be obtained by visiting www.pollsmartmr.com or www.targoz.com.
Results are of likely voters unless otherwise noted. Some percentages may not add to 100% due to rounding. In the same way, percentages in “total” columns may be one point more or less than the sum of their parts due to rounding.
Our Analysis
Biden’s lead among likely voters narrowly trails Clinton’s 2016 popular vote margin. Biden leads by 10 points in urban/suburban areas of the country while Trump is up 40 points in exurban markets. Also, Trump is up nearly 7 points in the swing states, which account for almost a quarter of likely voters in this poll.
| Registered Voters | Likely Voters | Swing State Voters | Urban Voters | Exurban Voters | |
|---|---|---|---|---|---|
| Biden | 47.9% | 47.5% | 41.3% | 51.9% | 26.9% |
| Trump | 44.8% | 46.3% | 48.1% | 41.9% | 67.2% |
| Other | 7.3% | 6.1% | 10.6% | 6.3% | 5.9% |
| % of Likely Voters | 24.7% | 82.4% | 17.6% |
We tend to believe Trump voters are not silent when answering online surveys. Some are likely being screened out by narrow likely voter questions or are not flagged on known lists of active voters. The shy Trump voter could be an issue with live interviewing/phone polling.
We do believe online polls will have better success this year since they mostly eliminate social desirability bias. Typically, respondents to polls or surveys are more likely to state a preference regarding polarizing issues or say they don’t care about an issue or topic when staring at a blank screen.
With live interviewing, respondents are more likely to state an affirmative response when they really don’t care about an issue. During live interviewing, respondents are more likely to say they don’t know or that they are undecided when asked about polarizing topics such as which Presidential candidate they plan to vote for. While the shy Trump voter could be an issue with live interviewing, we believe it is less of an issue with online polls.
We do see a bloc of male voters without a college degree, who are not regular voters, that are planning to vote this year. They appear to be older (45+), white, and do not have a college degree. To borrow a P. J. O'Rourke book title, many in this group could be described as the “Don't Vote, It Just Encourages the Bastards” voting bloc. Trump leads this group by 10 points (53%/43%), which is a lower margin than in 2016. However, this block only accounted for 16% of voters in 2016. In 2020, it appears they could make up around 25% of voters.
Trump is also up 15 points among voters with household incomes of $100k or more, which account for over a third of likely voters at 34% and is up from 2016.
While dealing with the coronavirus is the number one issue among likely voters, economic issues such as bringing jobs and manufacturing back to the U.S., cutting unemployment and getting people back to work, and addressing crime, law, and order rank 2 through 4 in terms of importance to voters.
While Biden leads on dealing with the coronavirus, Trump holds a substantial advantage regarding economic issues and addressing law and order matters.
| Which of the following issues are most important to you in deciding on who to vote for President this year? Please choose up to three | Likely Voters | Biden Voters | Trump Voters |
|---|---|---|---|
| Dealing with coronavirus | 46% | 61% | 32% |
| Bringing jobs and manufacturing back to the U.S. | 36% | 21% | 53% |
| Cutting unemployment and getting people back to work | 31% | 20% | 40% |
| Addressing crime, law, and order | 30% | 14% | 46% |
| Expanding healthcare | 29% | 39% | 17% |
Concerned news stories regarding the Biden campaign’s outreach to Latino voters appear to have merit. While the subsample is somewhat small, it does appear the Trump campaign is doing better with Latino voters than in 2016.
Also, Trump is doing better with 18 to 29-year-old voters. Some in the media have speculated that voter outreach on college campuses, typically a strong area for democrat candidates, are subdued due to limitations imposed by the coronavirus. This polling corroborates those concerns.
Regardless of how they plan to vote, nearly a third (32%) of likely voters say they say they don’t particularly like Donald Trump or Joe Biden. This includes 33% of Biden voters and 28% of Trump voters.
While many voters bemoan their choices for President, they can agree on one issue: almost half of likely voters (48%) believe some voters in their community support Donald Trump but won't tell anyone about it.
When asked which candidate their neighbors are voting for, nearly two in five (38%) thought most of their community were voting for Trump, and over a third (34%) thought they would vote for Biden.
In a testament to confirmation bias or the homogeneity of residential communities, 68% of Trump voters believed their neighbors would vote for Trump, and 66% of Biden supporters believed their neighbors supported Biden.
Fears that we will not know the winner on election night probably have merit. Mail or absentee voting could account for 31% of total votes, including 30% of voters in the swing states. Nearly half (45%) of Biden voters indicate they plan to vote by mail. It could be a long November if things remain the same until election day.
| Likely Voters | Biden Voters | Trump Voters | |
|---|---|---|---|
| In-person on election day at your voting location | 44% | 34% | 56% |
| In-person at an early voting location | 18% | 12% | 25% |
| By mail or absentee | 31% | 45% | 16% |
| Not Sure | 7% | 9% | 3% |
On election night, without early counting of mail or absentee ballots, it could appear Trump would be coasting to an easy victory. As mail or absentee ballots start to roll in, we could see Trump’s lead decline quickly which would be the most 2020 event ever, easily eclipsing the Florida recount fiasco from the 2000 election for President.
| Registered Voters | Likely Voters | In-Person or Early Voters | Mail or Absentee Voters | |
|---|---|---|---|---|
| Biden | 47.9% | 47.5% | 33.7% | 68.9% |
| Trump | 44.8% | 46.3% | 61.2% | 24.3% |
| Other | 7.3% | 6.1% | 5.2% | 6.8% |
Overall, 39% of likely voters said they watched or listened to the Democrat convention, and 35% indicated they watched the Republican convention.
More than half of self-identified Democrats (54%) and 28% of Independents said they watched the Democrat Convention. More than six in ten (65%) Democrats and 30% of Independents who watched the Democrat convention said they now had a more favorable view of Biden as a result.
More than half (55%) of self-identified Republicans and 27% of Independents said they watched the Republican convention. Eight in ten (81%) Republicans and 31% of Independents who watched said they now had a more favorable opinion of Donald Trump.
Just over one in five (21%) likely voters said they watched both conventions. Of this group, over half (51%) identified as a Republican, 38% Democrat, and 12% Independent. When we asked this group how they compared the two conventions, most scored the Republican Convention higher on each of the measures we tested.
| Which convention: | Republican Convention | Democrat Convention | Neither Convention | No Opinion |
|---|---|---|---|---|
| Was more patriotic | 56% | 23% | 15% | 5% |
| Was more hopeful | 51% | 35% | 12% | 2% |
| Looked more professional | 46% | 32% | 6% | 17% |
| Maintained the right balance between criticizing their opponents and saying positive things about themselves | 41% | 33% | 13% | 13% |
| Offered a positive vision for the country | 53% | 34% | 12% | 2% |
Coming Up
It appears we could be headed toward another 2016 type election. Unlike 2016, we might have to wait a few days to find out, with certainty, who won. This will all depend on what happens in the swing states, particularly in the Midwest.
We plan to poll Arizona next, and we will update these national results with a deeper dive into other items we covered in this poll, which will be posted later on www.pollsmartmr.com
Download The Report
Download Crosstabs
| Ballot | ||||||||
|---|---|---|---|---|---|---|---|---|
| 2016 Results | 2020 Poll | |||||||
| Clinton | 49% | 2.10% | Biden | 48% | 1.20% | |||
| Trump | 46% | Trump | 46% | |||||
| Other | 5% | Other | 6% | |||||
| Turnout reg. voters | 86.80% | Est. turnout reg. voters | 86.40% | |||||
| Comparison to National Averages | ||||||||
| Real Clear Politics Average 9/6/2016 | Five Thirty Eight Average 9/6/2016 | Real Clear Politics Average 9/6/2020 | Five Thirty Eight Average 9/6/2020 | |||||
| Clinton | 41% | 42% | Biden | 50% | 51% | |||
| Trump | 39% | 39% | Trump | 43% | 43% | |||
| Other | 11% | 19% | Other | 8% | 7% | |||
| Gender | ||||||||
| Clinton | Trump | Other/No Answer | Biden | Trump | Other/No Answer | |||
| Male | 41% | 52% | 7% | Male | 43% | 52% | 5% | |
| 47% | 50.00% | |||||||
| Female | 54% | 41% | 5% | Female | 52% | 40% | 8% | |
| 53% | 50% | |||||||
| 24558 respondents | 1039 respondents |