See how demographic swings could impact the 2024 election

538’s new Swing-O-Matic shows which states could flip under different scenarios.

By Geoffrey Skelley, 538

Early polling and special election results have raised many questions about how the electorate might vote come November. Will younger voters swing sharply to the right? Might President Joe Biden gain more ground among white college-educated voters? Has former President Donald Trump made significant inroads with voters of color? Could turnout plummet because of unhappiness with a Biden-Trump rematch? Any of these developments could potentially change who wins the election.

With this in mind, 538 is proud to present its 2024 Swing-O-Matic interactive! This tool engages with these questions by allowing you to play around with how different parts of the electorate might vote, altering how much a group prefers one party’s candidate or the other, and just how high or low that group’s turnout is. As you increase support here or decrease turnout there, the Electoral College map will adjust as some states flip to red or blue. Remember, state demographics vary quite a bit, so certain changes could help one party more in some places than others.

Let’s take a look at how the Swing-O-Matic works. Using polling and demographic data, we’ve broken down the national electorate into subgroups based on five components: age, education, sex, income and race. Within these categories, you can adjust both the turnout and the share of each subgroup supporting either the Democratic or Republican candidate (i.e., Biden or Trump), as well as the share supporting third-party candidates. And if you want to input a uniform swing affecting party preference or turnout levels of the electorate as a whole, you can use the “Shift all voters” tool.

Each time you change a group’s preferences or turnout at the national level, the Swing-O-Matic applies the adjustment to each state based on its individual demographic traits. If a state flips to a different party than it did in 2020, based on your scenario, it will have a dotted outline around it on the map. You can also dig deeper into specific state-by-state results in the tables at the bottom of the interactive, which will detail just what your changes wrought across each state and demographic subgroup. Note that the interactive uses an iterative and additive approach, so if you were to swing college-educated voters by 5 points and 18-to-29 year olds by 5 points, that would collectively swing college-educated voters who are age 18 to 29 by 10 points.

But fear not, we aren’t going to just throw you into the deep end with the Swing-O-Matic. The interactive includes four hypothetical scenarios that help show how changes in party preference and turnout among different parts of the electorate could affect state-level results and the Electoral College outcome. These scenarios aren’t just for demonstration purposes either — you can use them as a starting place for your own electoral tailoring. Otherwise, you’ll begin with the default scenario, which applies each demographic subgroup’s preferences and turnout levels from 2020 to the makeup of the current voting-eligible population.

So what hypothetical swings are included? One of the Swing-O-Matic’s scenarios lays out what would happen if voters without four-year college degrees and voters of color shifted to the right, a possibility given recent trends and some early polling. In 2020, Biden won among nonwhite voters nationwide by 48 percentage points, according to our estimates built on data from the Cooperative Election Study, while Trump carried voters without a college degree by 7 percentage points. But in this scenario, Biden loses a little ground among Black, Latino, AAPI and other voters of color, such that his margin among the nonwhite electorate falls to 45 points, while Trump’s advantage among voters without a college degree expands to 9 points. The result? Trump wins.

In this scenario, Trump would flip the Sun Belt trio of Arizona, Georgia and Nevada as well as the Frost Belt duo of Pennsylvania and Wisconsin. Adding these five states to his 2020 map would give Trump 297 electoral votes to Biden’s 241, surpassing the 270 electoral votes needed to win the presidency. Notably, Biden would actually still narrowly win the popular vote, illustrating the GOP’s recent advantage in the Electoral College compared with its popular vote support. After all, Trump won in 2016 under somewhat similar circumstances while losing the popular vote. This result wouldn’t be a big surprise — all of these are swing states that Biden won in 2020 by less than his national margin of 4.5 points.

Another scenario involves Biden making gains among older voters and white voters, which early surveys have also pointed to as a possibility. In 2020, Trump won voters who were 65 or older by 11 points and non-Hispanic white voters by about 13 points. But if Trump’s margin among older voters fell by 3 points (to R+8) and slid by 4 points among white voters (to R+9) while turnout remained the same, that would buttress Biden’s showing and allow him to win by slightly more than he did in 2020 — both in the popular and electoral vote.

The other two scenarios in the interactive illustrate how changes in turnout among specific groups could affect the election outcome. In one hypothetical, voters ages 18 to 29 turn out at lower rates and give a slightly larger share of their vote to third-party contenders, which hurts Biden and helps Trump because this group has been strongly Democratic-leaning in recent times. As a result, Trump would claim a narrow victory in the Electoral College — 272 to 266 — even as Biden still wins the popular vote by almost 4 points. In the other scenario, reduced turnout among white voters and voters without four-year college degrees hurts Trump’s standing and aids Biden’s, as both groups are Republican-leaning. Biden would add one additional state (North Carolina) to his 2020 showing in this hypothetical.

Each of these scenarios provides a useful overview of how tweaking support and turnout could affect the 2024 election outcome. Of course, the thing about demographic trends is that every group will probably move one way or the other, even if only by small margins, compared to the last election. And while we know that presidential turnout is likely to be high, it’s impossible to know if it’ll be nearly as high as the record-setting turnout we saw in 2020, and just how it might vary from group to group.

At this point, everything about 2024 is a moving target. So, dear reader, have fun adjusting the electorate and seeing just how the election could play out under different circumstances. Now, you’re welcome to adjust the numbers any way you like, but do remember that dramatic swings aren’t likely. Namely, a group that has been heavily Democratic or Republican is unlikely to suddenly swing all the way in the other direction to overwhelmingly support the other party. For instance, Latino voters became less Democratic-leaning in 2020. Yet while surveys found that the Democrats’ margin among Latino voters shrank compared with 2016 — perhaps by as little as 5 points or as much as nearly 20 points — around 60 percent or more still voted Democratic. And while turnout could shift up or down from 2020, it’s nigh impossible that a given group’s turnout will surge to almost 100 percent — or fall close to zero. Still, you’re in control — go tinker to your heart’s content!


Geoffrey Skelley is a senior elections analyst at 538. @geoffreyvs

Be the first to comment

Leave a Reply

Your email address will not be published.