The 2024 election cycle has produced some startling and sometimes counterintuitive narratives about how demographic subgroups might end up voting. We may or may not be seeing a historical divide in the way men and women vote. Polls suggest we are facing the biggest racial realignment since the passage of the Civil Rights Act — or it could be a mirage. Young people could skip the election because they are disillusioned and vote for a third party — or they could turn out in record numbers for Kamala Harris. The more diverse Sun Belt states could pave the way for a Donald Trump victory — but the predominantly older and whiter “Blue Wall” states could elect the first black female president.

We’ll know soon enough. Although Election Day is just days away, at least 60 million people have already voted. Battleground states are meeting or exceeding their early voting records. And with polls of likely voters still showing an even race, any combination of factors, events or movements within the electorate could influence the outcome.

To that end, I’ve compiled a handful of questions that we at Vox have been tracking over the last year. Their ultimate answers could determine who wins the White House.

Will there be subsequent decision-makers? And what could change their minds?

The story of the final weeks of the 2024 election has been a scramble for undecided voters, whose numbers have dwindled in poll after poll. This share includes two groups: voters who cannot decide between the two candidates and voters who may have a preference but are not sure whether they want to vote at all.

However, we don’t know exactly who these late deciders are. Could these be the same type of working-class, non-college-educated voters (mostly white) who helped Trump win in the Rust Belt states in 2016 (and thus upset the polls)? Or will it be the throngs of new and young (mainly non-white) voters that could give Harris an edge in the Sun Belt states?

And for all of these subgroups, what kind of messaging or campaign development could persuade them to vote without already having been persuaded? Could Harris’s late revival of democracy and Trump’s authoritarian leanings resonate with them? Is something like the racist and extreme rhetoric at Trump’s Madison Square Garden rally a factor that could influence their decision? Or is something like President Joe Biden’s “garbage” gaffe this week something that could inspire more Trump support?

Regardless, these late deciders will be crucial. They defeated Trump by large margins in the last two elections in which he participated. But this third time everything could be different.

Will there be a Republican transition to Harris?

In this sense, Harris’ appeal to the fate of democracy and the juxtaposition of her “to-do” list with Trump’s “enemies” list are the clearest examples of how the Democratic campaign is targeting Trump-skeptical Republicans as a key element of prevention focused on a Trump victory. But will these registered Republicans cross party lines or simply be reluctant Trump voters?

Somewhere between 15 and 20 percent of Republican primary voters did not vote for Trump, and even after they dropped out, large portions of those voters chose to vote for former South Carolina Gov. Nikki Haley. Many of them are women, which explains some of the focus Harris has placed on touting Republican supporters such as former GOP Rep. Liz Cheney, her father and former Vice President Dick Cheney and numerous former Never-Trump Republican politicians.

However, partisanship is a damn powerful drug. Republicans, even if they personally dislike Trump, routinely stick with their party’s nominee. Harris keeps telling these Trump-wary Republicans to put “country over party.” But if they don’t, and Harris’ argument about Trump’s threat to democracy is correct, they may have to give up a “country.” above“Party.

Will Arab American voters lean Republican?

The Gaza war and Israel’s treatment of the Palestinians were one of the defining issues of last year, including in the electoral arena. Biden’s handling and response sparked significant discontent among more progressive and left-leaning members of the Democratic coalition, and that antipathy appears to have persisted to a lesser extent toward Harris. That includes a group of voters that is influential in a crucial swing state: Arab-American voters in Michigan.

Polls specifically among Arab Americans suggest that these voters will not go for Harris to the extent that they have supported Democratic candidates in the past: An Arab News-YouGov poll this week found that Trump favored Harris among Arabs Americans lead 45 to 43. That’s a strong statement. A reversal from 2020, when Biden led Trump by 24 points, and especially in 2016, when Hillary Clinton led Trump by 34 points.

However, this was not always the case. Before 9/11, Arab American voters leaned Republican. Only after the GOP’s anti-Muslim and anti-Arab turn during the George W. Bush years did this constituency shift toward Democrats, reaching a peak in 2004. And since that peak, these voters have leaned toward the GOP, with the share supporting John McCain, Mitt Romney and Trump growing from 2008 to 2012 and into 2020 (support fell slightly in 2016). The Gaza war may be accelerating a latent shift to the right that was already occurring as Republicans shifted their foreign policy priorities, championing conservative culture war themes and denouncing economic populism, while Democrats became more culturally progressive, including on issues of gender and sexuality.

Will Trump’s bet on younger black men pay off?

For much of the last year, the Trump campaign has increased its targeting of a specific segment of the electorate: Black men. With an avalanche of digital advertising aimed at younger black men and the use of surrogates and outside groups to reach young black voters, the campaign hopes to exploit two dynamics: Harris’ apparent weakness against black men and an overarching vulnerability of the Democrats versus younger Black Americans.

Traditional polls suggest that Harris faced the challenge of achieving the same margin of support that previous Democratic candidates have enjoyed among black voters and particularly black men. Both social and economic reasons explain this, including former President Barack Obama’s theory that a certain level of misogyny prevents some black men from supporting a black woman.

But there is also a larger weakness for Democrats, based on polls showing that younger black voters in particular may have weaker ties to the party than older cohorts of black voters and may be more conservative than their elders. And young black men appear more likely to support Trump this year, perhaps because of that weaker tie.

However, this is also the group of voters least likely to vote, and some polls suggest Harris is consolidating ahead of the election. And with Trump and his supporters increasingly blatantly racist and biased in the final weeks of the campaign, it’s not clear whether that investment will be big enough on Election Day to decide the race in battleground states.

Will Latino voters shift to the right in key states?

Whether Latino voters have turned to the Republican Party since the start of the Trump years isn’t really a matter of debate. Trump’s gains in 2020 remained with Republican candidates through the 2022 midterms, and polls suggest he will retain at least a majority of support a week from now. However, since the election is decided by the Electoral College and not the popular vote, the more interesting question is whether these gains will persist or increase in key states.

In 2020, much of the political media was fascinated by Trump’s massive advances in South Florida and South Texas, places that had given Democrats an advantage in Latino support for years. But Trump’s Hispanic gains also occurred across the country, particularly in immigrant communities and in both Democratic and Republican strongholds, which do not necessarily impact the Electoral College map results.

This year, it appears that states already likely to strongly support Trump or Harris may see their Latino populations shift further to the right (most clearly seen in Florida), even as Latino voters in Swing states like Arizona, Nevada and Pennsylvania polls suggest should buck this trend and go Democratic (or at least keep Democratic margins intact as of 2020).

That could result in Trump making greater gains among Latinos nationwide, but not enough in swing states to boost him in key presidential elections. This would be further evidence of an ongoing racial realignment between the parties, but one driven by Hispanic and Latino voters in California, New York and Texas. This has real implications for control of Congress, but unless the Latinos who switch party affiliations are in swing states, it will have no impact on who wins the White House.