Just under six months out from Election Day, national polling averages are showing Donald Trump leading Joe Biden by just over 1 percentage point. They also show that, despite narratives of “momentum” for either candidate, the race has been fairly stable for the better part of a year.

But what’s happening under the surface? Which subgroups are driving this ~5-6-point national rightward shift from the 2020 election?

Over the past six months, I’ve been seeking to answer these questions by aggregating subgroup data from the crosstabs of reputable national 2024 general election polls (read about the full methodology here) and comparing them to justified baselines of 2020 election results.

Aggregating crosstabs across national polls gives us much larger sample sizes to examine, and thus a clearer picture of what reputable pollsters are collectively presenting as the state of the race.

Note that the numbers and analysis below reflect what polls have shown in April. Polls are merely snapshots in time and are not necessarily predictive of the outcome of the election this far out — overall or by subgroup. However, they are useful tools to show, at least directionally, demographic and subgroup trends in the electorate.

Subgroup Polling Averages from April 2024 Head-to-Head Polls

Overall, Trump leads Biden by 0.9 points across 12 selected polls conducted in April, which tracks closely with other averages.

Here are the aggregated results of head-to-head polls by subgroup for the month of April (see the full spreadsheet for more information, and scroll to the bottom of this article for the full methodology):

April national head-to-head polls showed Biden losing the most ground since 2020 among the following subgroups:

• Black (-23)
• Urban (-15)
• Independent, including partisan leaners (-14)
• Latino (-13)
• Moderate (13)*
• Age 18–29 (-12)

* Showing data for March due to insufficient data in April

These losses reflect withheld support for Biden, as Trump has gained less than what Biden has lost to voters declaring for undecided/other. In other words, they’re unhappy with Biden but have not realigned with Trump.

Greater losses among Black and Latino voters, who tend to be to the right of other groups in the Democratic base, indicate Biden may be more at risk of lasting defections from centrists rather than those on his left flank— despite all the progressive criticism regarding his actions around Israel/Palestine.

Biden has only gained relative to 2020 among the following subgroups:

• Republican, excluding leaners (+3)
• Rural (+3)
• Age 65+ (+2)
• Age 50–64 (+1)
• White, Non-College (+1)

Generally, the polls suggest that Trump has largely consolidated his coalition, despite continued support for Nikki Haley, even in closed primaries. Meanwhile, Biden still has room to improve, as his losses among the Democratic base have been partially offset by holding steady among high-propensity seniors, rural voters, and white voters (which may simply be an artifact of differential non-response and weighting effects).

Below I’ll be digging deeper into some of the subgroups where crosstab aggregates are showing particularly large shifts relative to 2020.

Race/Ethnicity

Black Voters

Among Black voters, in aggregate Trump is outperforming his 2020 margin by a whopping 23 points. Relative to 2020, Biden has lost more support (-16 points) than Trump has gained (+7 points), with the remaining 9% moving to undecided/other.

If Trump matches his April polling vote share (15%) among Black voters, it would be nearly double what he received in 2020, and would be the strongest performance by a Republican presidential nominee in nearly 50 years.

It should also be noted that April was one of Trump’s poorest polling months among Black voters — his average vote share since November is even higher (17%) and has been as high as 20%.

Based on his recent campaign schedule, it appears that the Biden campaign is well aware of his struggles among Black voters, despite publicly dismissing polling showing his slippage.

Some of Biden’s weakness among Black voters may be attributable to the fact that many are not yet paying close attention to the race. An Economist/YouGov poll conducted last week found that only 27% of Black voters are paying “a lot” of attention to the campaign — the lowest of any demographic group tested, apart from young voters.

It would not be a surprise if Trump gained a few points among Black voters relative to 2020. After all, Black voters are by-and-large fairly moderate. If that did occur, the most likely groups to abandon Biden for Trump appear to be younger Black voters (who are increasingly disconnected from the Civil Rights era), Black men, and Black conservatives that may be regressing to the historical mean after temporarily setting aside their ideology to support Barack Obama.

But it would be surprising and historically anomalous if Trump approached anywhere near 20% among Black voters, who have not shown willingness to broadly defect to Republicans in recent elections. Rather, the Biden campaign’s primary concern might be anemic turnout among Black voters due to lukewarm feelings toward Biden’s presidency to date.

Latino Voters

Among Latino voters, in aggregate Trump is outperforming his 2020 margin by 13 points. Once again, compared to 2020 Biden has lost more support (–9 points) than Trump has gained (+3 points).

If Trump ends up winning 40% of Latino voters, it would match the highest performance by a Republican presidential candidate in the last 50+ years (George W. Bush in 2004).

However, many national pollsters cut corners/costs by not offering Latino respondents the option to take the survey in Spanish. This means that Latinos that primarily speak Spanish at home — who have disproportionately supported Democrats in recent elections — may be underrepresented in national polls, so Trump’s strength among Latinos may be directionally accurate, but slightly overstated.

A few other notes on Latino voters:

• One pollster estimates that for nearly one-quarter of Latino voters (who are disproportionately young), this year will be their first time voting in a presidential election.
• Latino voters are not a monolith, and their voting behavior reflects that, meaning Trump’s gains among Latino voters will not be spread evenly across the country.
  • Protestant Latinos (and especially evangelicals) tend to be more conservative that Catholic Latinos.
  • Cuban Americans tend to be more conservative/Republican than non-Cubans (especially Mexican Americans).
  • Those whose families have moved to the U.S. more recently and those who primarily speak Spanish at home are more likely to support Democrats.
• An Economist/YouGov poll conducted last week found that only 32% of Latino voters are paying “a lot” of attention to the campaign — 10 percentage points lower than white voters.
• On the whole, while Latino voters are very pro-choice, they are fairly moderate or at times conservative on many issues.

White Voters

White voters are holding fairly steady compared to 2020. Biden has dropped by 4 points and Trump has dropped by 3 points, with the balance moving to undecided/other.

Biden has dipped slightly among white college grads, while he has gained slightly among white non-college grads:

• Among white college grads, Biden’s vote share has dropped by 5 points since 2020, while Trump’s has dropped by 1.5 points.
• Among white non-college grads, Biden’s vote share has dropped by 3 points since 2020, while Trump’s has dropped by 4 points.

Looking at trend lines in the two-party vote over the last six months, Biden has made some slight gains among all Black and Latino voters (which roughly tracks with the slight gains he’s made nationally in this period), but definitely far short of the base consolidation that his campaign is counting on.

If the shifts we are currently seeing among Black and Latino voters came to fruition in November, it would amount to the largest single-election racial realignment that we have seen since the Civil Rights Act was passed in 1964.

Note: “2020 Results” are a flat average of post-election estimates from Catalist, Pew’s validated voter study, and AP VoteCast

Age

Young Voters

Among voters aged 18–29, in aggregate Trump is outperforming his 2020 margin by 12 points. Relative to 2020, Biden has lost more support (–9 points) than Trump has gained (+4 points). The last time a Republican presidential candidate won at least 40% among this age group was George W. Bush in 2004.

While we don’t have reliable 2020 (or historical) estimates for voters aged 18–34 to compare against, we can be reasonably confident that Biden winning this group by just 2 points would be a seismic rightward shift given that voters aged 18-29 backed Biden by 24 points four years ago.

Similar to Black voters, some of Biden’s weakness among young voters may be attributable to the fact that many are not yet paying close attention to the race. An Economist/YouGov poll conducted last week found that only 23% of voters aged 18-29 are paying “a lot” of attention to the campaign — the lowest of any demographic group tested.

Another risk to the Biden campaign among this cohort are abstentions or third party defections, which could cost him a significant amount of support among those who backed him four years ago. The progressives among them are comparatively unlikely to vote for Trump.

Seniors

Among senior citizens, Biden is actually outperforming his 2020 margin by 2 points. This time, Trump has lost more support (–5 points) than Biden (–2 points). Seniors are also paying much closer attention to the campaign than most other demographic groups, so they may be less likely to shift in either direction over the next six months. Notably, Biden also polled well among seniors in many 2020 polls, but ended up losing them by 4 points.

The trended data for age is fairly noisy month-to-month, but the two-party vote trend lines since November show that Biden has lost some ground among young voters (using either definition), Trump has lost ground among voters aged 35–49, and things have held relatively steady among voters aged 45–64 and 65+.

Note: There are no reliable 2020 post-election estimates for Age 18–34 and Age 35–49.

Note: Pollsters either define young voters using Age 18–29 or Age 18–34. That, in part, explains why there can be such a large gap between these two groups, not because voters aged 30–34 are highly Republican.

Independents and Moderates

In aggregate, Trump is outperforming his 2020 margin among independents (including partisan leaners) by 14 points and among moderates by 13 points, which dwarfs the national swing of R+5.5. Both of these groups were critical to Biden’s victory four years ago.

Relative to 2020, Trump’s vote share is only 2 points higher among both independents and moderates, but Biden’s vote share has dropped by double digits among both groups (–12 points among independents, –10 points among moderates). It’s worth noting that under 30% of both groups are paying “a lot” of attention to the campaign to this point, far lower than most other subgroups.

Nevertheless, moderates and independents are broadly very sour on his presidency and are disproportionately more likely to swing against him, and will likely be much harder to win over than holdouts on the left.

Trump’s lead among independents (+1 to +7) and Biden’s lead among moderates (+12 to +16) has been fairly consistent over the last six months.

Note: Independents include partisan leaners

Note: There was an insufficient number of head-to-head polls in April that released ideology (Moderates) data to include in the dataset

Population Density

When examining the composite results by population density, the polls suggest urban areas are swinging the most towards Trump, with a 15-point shift. This is followed by a suburban Republican shift of 4 points%, followed by a Democratic gain of 3 points% in rural areas.

This can be explained by the more racially diverse, younger urban demographics being the most likely to cross over or defect. Trump’s support rising in suburban areas is interesting, considering Haley’s aforementioned staying power even past the conclusion of the Republicanpresidential primary. Biden gaining among rural voters, meanwhile, can be explained by his slight gains among groups associated with rural communities: a more Republican, white, and non–college educated slice of the country.

Two-party vote margins among urban and suburban voters have remained fairly steady since November, while Trump has gained some ground among rural voters.

Subgroup Polling Averages from April 2024 3+ Candidate Polls

In April national polls testing a ballot with 3+ candidates, Trump leads Biden by an average of 1.3 points, with RFK Jr. pulling 9% and Cornel West/Jill Stein winning 4%.It should be noted that we do not yet know which, or how many, state ballots these candidates will qualify for.

Here are the aggregated results of 3+ candidate polls by subgroup for the month of April:

RFK Jr.’s vote share was the highest among voters aged 18-34, Independents (including partisan learners), and moderates — consistently his strongest three groups since November. These groups of voters also appear to be among the least likely to be paying close attention to the presidential election (at least to this point).

Over the last six months, polls have shown Jill Stein and Cornel West unsurprisingly pulling a disproportionate amount of support among disaffected traditionally Democratic constituencies — voters aged 18-34 (11%), Black voters (8%, mostly going to West), and liberals (7%). However, they’re also appealing to some voters in the center — pulling 7% of independents (including partisan leaners) and 6% of moderates. They are barely registering at all among senior citizens, Republicans, or most importantly, 2020 Trump voters.

These defections could explain the asymmetric losses Biden is seeing with the Democratic base. Republican-leaning demographics are seeing less defection; perhaps they are truly more satisfied with their presidential nominee, or perhaps those who lean Democratic are more willing to express dissatisfaction.

RFK Jr.’s vote share was highest in November (15%). Since January, it has dropped by 1% each month. Stein and West’s combined vote share has consistently hovered around 4%, and their support by subgroup has remained fairly constant over the last six months. Historically, minor party candidates have almost always underperformed their pre-election polling.

Since November, RFK Jr.’s vote share dropped the most among Latino voters, while it held relatively steady among white non-college voters. His drops among Black and white college grad voters closely matched his national decline during this period.

Over the last six months, RFK Jr.’s vote share dropped disproportionately among independents (including partisan learners) and moderates.

Note: Independents include partisan leaners

Note: There was an insufficient number of 3+ candidate polls in January that released ideology (Moderates) data to include in the dataset

From November to April, RFK Jr.’s vote share has dropped by 3-6 points among men, women, age 18-34, age 65+, Democrats, Republicans, 2020 Biden voters, and 2020 Trump voters – similar to the 6-point drop he experienced among all voters during that period. 

Conclusions

Whether you made it all the way through this weighty tome or if you skipped to the bottom (no judgments), here are the five key things to remember:

• If recent national polls are gauging real movement (even directionally) away from Biden among Black, Latino, young, urban, independent, and moderate voters, and that persists through Election Day, his campaign is in serious trouble. But it’s important to remember that polls are imperfect tools prone to error (especially at the subgroup level) and are merely snapshots in time.

• Biden has slipped among most subgroups relative to 2020, but he’s losing more support than Trump is gaining, meaning many are not (yet) eagerly embracing Trump and might still be persuadable.

• There is still plenty of time for Biden to bring home Black, Latino, and young voters that backed him in 2020, and the base usually comes home, but turnout, portended by current polled defection, may still be an issue for him.

• Despite the media’s obsession with Biden’s issues on his left flank, his more serious and lasting issue may be among voters in the center.

• RFK Jr.’s vote share has already begun to drop, even before the forthcoming barrage of attack ads, and minor party candidates almost always underperform their polling. But he could still sway the race in a close election.