Trump enters Election Day with the tiniest of edges over Kamala Harris

Kamala Harris and Donald Trump
If a pollster says they know who is going to win, they’re lying - RYAN COLLERDRYAN M. KELLY/AFP

On Wednesday evening, I sat with a group of black, white and Asian undecided voters in the suburbs of Detroit. It is my last focus group of 2024. I am ending as I started: I was last here in January, talking to independents about the race between Joe Biden and Donald Trump.

The city was packed with snow then, but the words voters use about Trump are the same now as then and in the scores of other focus groups in between. “Crude”. “Nasty”. “Aggressive”. “Nonsense”. “Racist”. “Ignorant”. They see the former president as a brute; a heaving, unbound colossus who crushes anyone in his way and exhibits the worst of America’s excesses.

All eight respondents felt like this. But when it came to how they would vote, half of them picked Donald Trump regardless.

He might be a “wolf”, said one black woman, but he is not a “wolf in sheep’s clothing like Harris”. For an Asian American in the group, Trump’s aggression also means strength, which he would bring to bear on fixing the border. For a black man, Tim, he was a “disruptor” who would shake things up. For Randi, a 50-year-old white housewife, he is like a “film director”: “the most hated person on the set but with the vision to lead”.

The other white women in the group picked Kamala Harris, referencing her commitment to women’s rights and “not being Trump”.

There are two dynamics at play here. The first undermines the adage that the least favourably viewed candidate loses; people dislike Trump, but they still pick him to be their man.

The second is that non-white voters are more open to voting Republican than they have ever been before. Our final JL Partners poll of the cycle has Trump winning 45 per cent of Hispanic voters, a ten-point jump on what he achieved in 2020. It has him on 15 per cent of black voters, double his performance from four years ago. Eleven months ago, I might have raised an eyebrow at these numbers, but after meeting scores of people from both groups, I think this is a real shift. The economy, frustrations with the border, an increasingly liberal Democrat party on family values, and a dislike of Harris have all underscored the move.

Meanwhile, Harris’ hold over her most loyal non-white voters is slipping. Not only is her support among black voters lower than the Democrats enjoyed under Biden or Clinton, but black turnout in early voting statistics is mixed. While Harris will take heart from strong turnout in Detroit and North Carolina, in Georgia the black percentage of the early vote is lower than any recent election there, on 26.3 per cent compared to 28.3 per cent in 2020 and 27.8 per cent in 2016.

On the other hand, Trump’s base seems to be buzzing with energy. Our poll finds 74 per cent of Trump voters are “very enthusiastic” about voting, compared to just 67 per cent of Harris supporters. Rural counties are turning out in Georgia, with the top counties for early votes being ones that heavily backed Trump in 2020.

Many of these voters are people who have not voted before. That again bodes well for Trump, who leads Harris among previous non-voters by double digits. Trump won in 2016 by turning out these non-voters, and he may be about to do the same again.

So far, so good for Trump. But what of the white, college-educated women – Wendy and Nora – in my focus group?

They may end up being Harris’ own trump card. Angered by Republican positions on abortion, positive about Harris’ character (“she just makes me feel good”) and horrified by Trump, women like Wendy and Nora could storm the suburbs for Harris and deliver her the states she needs to win. Harris still leads Trump by two points in America’s suburban areas, which have often plotted the path to the White House.

This is the group that Ann Selzer, an esteemed pollster in Iowa, thinks will march Harris to victory. She found Harris up 3 points in Iowa, a deep red state in 2016 and 2020, largely resting on female independents and women over the age of 65. This group votes, and women have swung to Harris since 2020 – she now leads among women by 14 points to Biden’s 12.

Though Trump has also increased his lead with men (from 5 points over Biden in 2020 to 22), men are less likely to get off the couch and get to the voting booth.

We can see that in the early voting. More women have voted so far than men, with 54 per cent of the early vote in states that register this being female, compared to 44 per cent for men. That is not historically unusual, but with the swing to women that should alarm the Trump campaign.

All these factors leave pollsters like me in a sorry position. Anyone telling you they know who will win is lying to you. The polls have been wrong in the past and could be again. It is exceptionally difficult to model who will turn out in a given swing state, especially with such tight margins. A better ground game on the day in Pennsylvania – the state most likely to hand a candidate victory – could change everything.

But looking at the polling and what has happened already, I believe Trump has the very slightest of advantages. He is galvanising his base in the countryside. He is facing a tidal wave of angry white women, but he is also eroding a traditional blue mountain of non-white voters.

Back in Detroit, I ask Sheree, a 31-year-old undecided black woman, who she would vote for if she had to decide today. After countless polls and interviews, it is the last question I ask a voter this year.

She mumbles, twists her face, and looks tortured. Then finally she says it. “Trump.” Nora and Wendy, the white women, look on in shock.

“Why do you say that, Sheree?”, I ask.

“Because of the economy”.

I don’t know who is going to win. But as I gather up my notes from my final focus group, Sheree’s voice ringing in my ears, I think Trump goes into it with the tiniest of edges in an historically close race.


James Johnson is the co-founder of JL Partners