Who's winning the 7 key swing states, Harris or Trump? Inside the latest polls.

In the United States, winning the most votes nationwide isn’t enough to win you the presidency. Just ask Hillary Clinton, who bested Donald Trump by nearly 2.9 million votes in 2016 — and still lost the election.

The real trick is to rack up the most votes in the right places.

So which states will be this year’s battlegrounds? And which forces — local or otherwise — will determine whether each battleground state ultimately sides with Trump or Vice President Kamala Harris, the presumptive Democratic nominee?

Here’s a handy Yahoo News guide to the 2024 map.

Why do battleground states matter?

To win a U.S. presidential election, you don’t necessarily need to win the popular vote (i.e., the combined tally across all 50 states). You need to stockpile electoral votes by finishing first in individual states.

With minor exceptions, the candidate who gets the most votes in a particular state on Election Day is awarded all of that state’s electoral votes, which are proportional to its population. Whoever finishes first in enough states to secure at least 270 electoral votes — a majority — wins the Oval Office.

Because of these winner-take-all rules, candidates tend not to campaign in states that consistently vote Democratic (blue states) or Republican (red states). Instead, they focus on the small number of states (known as swing states, battleground states, toss-up states or purple states) that could go one way or the other.

How do we know which states will be battlegrounds in November?

History helps. Certain demographic groups tend to vote Democratic or Republican, and certain states tend to have a 50/50 mix of Democratic- and Republican-leaning groups. The vote in those states tends to be close.

But state demographics can change over time. The political preferences within each demographic group can change too. As a result, a former battleground state might start to lean red or blue.

Ohio is a good example. From 1900 to 2012, the Buckeye State accurately chose the winning presidential candidate 93% of the time (including in every election from 1964 to 2016). But as white, working-class voters increasingly gravitated toward Republicans, so did Ohio. In 2020, Trump won there by a comfortable 8-point margin — even though Joe Biden won the election.

Missouri followed the same trajectory; Florida has been trending rightward as well. New Mexico, Colorado and Virginia, on the other hand, have veered to the left.

So to figure out this year’s battlegrounds, you can’t just cut and paste the previous list. You need to look at which states are too close to call right now. And to do that, you need to consult the latest polls.

Can we trust state polls?

To a degree. National polls tend to be more reliable; on the state level, it’s hard for pollsters to drill down on the most representative mix of respondents (i.e., one that will have the same demographic characteristics as the eventual electorate).

But you can still get a fairly accurate sense of where things stand by averaging together the most trustworthy state polls — and you can further improve that average by incorporating it into a statistical model that includes campaign “fundamentals” as well (such as how the economy is performing and how the state has voted in the past).

Data journalist Nate Silver, formerly of the New York Times and FiveThirtyEight, does just that with the latest version of his Silver Bulletin election forecast. It’s the most comprehensive ranking of 2024’s battleground states to date.

So what are the numbers saying right now?

The Silver Bulletin model currently projects that Harris and Trump could finish within 3 points of each other in seven states. Here’s how those states rank in terms of closeness, along with how they voted in 2020:

1. Arizona (Trump +0.6% in 2024; Biden +0.3% in 2020)

2. Georgia (Trump +0.9% in 2024; Biden +0.2% in 2020)

3. Pennsylvania (Harris +1.0% in 2024; Biden +1.2% in 2020)

4. Nevada (Harris +1.1% in 2024; Biden +2.4% in 2020)

5. Wisconsin (Harris +1.9% in 2024; Biden +0.6% in 2020)

6. North Carolina (Trump +2.0% in 2024; Trump +1.4% in 2020)

7. Michigan (Harris +2.6% in 2024; Biden +2.8% in 2020)

These were the seven closest states in 2020 as well. But under the surface, key dynamics seem to be shifting.

Before Biden dropped out in July, polls showed Trump cutting into the president’s 2020 margins among several traditionally Democratic groups: Latino voters, Black voters (especially Black men) and younger voters.

Previously, Biden had been outpacing his Democratic predecessors with some groups that usually favor Republicans, including college-educated white voters and senior citizens (especially white seniors).

That kept him in the hunt.

But the president’s swing-state numbers started to crater after his worrisome debate performance on June 27, and Trump pulled ahead nearly everywhere. Lacking a clear path to victory, Biden was finally forced to step aside.

Since entering the race, Harris has largely reversed the damage, taking narrow (forecasted) leads in the Rust Belt states of Michigan, Wisconsin and Pennsylvania.

Yet Trump remains relatively strong — compared to 2020 — in the younger, more diverse Sun Belt states.

The big question going forward is whether Harris — a younger, more diverse candidate than Biden — can surpass Trump in the Sun Belt as well. Unlike Biden, Harris now leads Trump in the Nevada forecast, and she’s partially closed the gap in Georgia, Arizona and North Carolina.

Who would win if the election were held today?

The national polls now show Harris edging Trump by an average of about 2 to 3 percentage points. She was trailing when she entered the race.

In 2020, Biden beat Trump with electoral votes to spare: 306 to 232. If 2024’s eventual results match today’s Silver Bulletin forecast — with Arizona and Georgia flipping to Trump; with North Carolina remaining red; and with Wisconsin, Michigan, Nevada and Pennsylvania staying blue — Harris would win with 279 electoral votes.

But that’s a big if. Another way to think about the election is in terms of probabilities. The Silver Bulletin model currently thinks Harris would win the Electoral College 55 times out of 100; Trump would win the other 45 times. It’s hard to get much closer than that.

What might change between now and Election Day?

In general, voter demographics and distribution determine the vast majority of each state’s eventual outcome. College-educated white Pennsylvanians, for instance, don’t vote in a wildly different way than college-educated white Arizonans. The question is how many of them live and vote in each state.

But when the winning margins are narrow — like they were in 2020 — local dynamics can make a difference. A few to watch between now and November:

• Battleground states with key Senate and gubernatorial elections, including Arizona, Michigan, Nevada, Wisconsin, Pennsylvania and North Carolina. Why? Because high-profile statewide races can affect who turns out to vote for president.

• Battleground states with hot-button ballot measures, including (potentially) abortion in Arizona and Nevada and immigration in Arizona. Why? Because such measures can get otherwise unmotivated voters to show up on Election Day to vote for or against the new policy — then cast a ballot for president while they’re at it.

• Battleground states where Robert F. Kennedy Jr. is on the ballot, including Nevada, Michigan and North Carolina (and possibly Georgia, Wisconsin and Pennsylvania as well). Why? Because RFK Jr. could play the spoiler, siphoning just enough votes to put Harris or Trump over the top.

• Battleground states with their own idiosyncrasies, with Michigan being the prime example. Home to the nation’s biggest bloc of Arab Americans, the Great Lakes State could prove unpredictable if opposition to the Biden-Harris administration’s Gaza war policy disrupts that group’s usual Democratic voting patterns.

Could Harris or Trump expand the battleground map by putting other states in play?

That’s unlikely. If some big event or force — such as an economic collapse or a federal Trump conviction — were to alter the fundamental dynamics of the race in favor of one candidate or the other, then yes, the next tier of states could become battlegrounds.

These include Florida (Trump +4.7%, according to Silver Bulletin), Virginia (Harris +5.5%), New Hampshire (Harris +5.8%), Texas (Trump +6.8%), Maine (Harris +7.6%), New Mexico (Harris +8.0%) and Ohio (Trump +8.1%).

But even then, the current, core group of seven battleground states would still prove decisive, because the margins there are closer — so they would break for Trump or Harris first. Piling up extra electoral votes elsewhere would just be icing on the proverbial cake.