In the United States, we elect our president and vice president using the Electoral College rather than the national popular vote. To win, a party’s ticket must win the most votes in enough states to attain an outright majority of electoral votes — at least 270 votes out of a total of 538 (sound familiar?).

But lurking in that overall figure is the potential for no candidate to win a majority, in the case of a 269-to-269 tie. While seemingly not in play in the 2024 election, it’s also possible that third parties could win sufficient electoral votes to prevent anyone from hitting 270. Regardless of how it happens, should no contender claim a majority, the Constitution calls for Congress to carry out the selection of our national executives via a “contingent election.” In this scenario, the U.S. House of Representatives would choose the president, although based not on the vote of the whole chamber but each state’s preferences — meaning a candidate needs majority support from 26 state delegations out of 50 to win — and the 100-seat U.S. Senate would pick the vice president based on the vote of individual senators, with 51 votes needed to win.

Could we run into this situation in the 2024 presidential race? It’s possible, though pretty unlikely: According to 538’s presidential forecast, there’s only about a 1-in-300 chance that neither Vice President Kamala Harris or former President Donald Trump attain 270 electoral votes, and the only plausible scenarios here involve a 269-269 tie in which key swing states shift in ways that appear less likely based on the current state of the race. With that in mind, let’s take a look at how we could end up with contingent elections for president and vice president, and why such a development would heavily favor Republicans.

How Harris and Trump could both end up shy of 270

Of the small number of scenarios in 538’s presidential forecast in which Harris and Trump tie, four scenarios are by far the most plausible, making up about 4 in 5 of those rare cases. In each, Harris would capture states and congressional districts worth 225 electoral votes, while Trump would capture 219. That leaves the 94 electoral votes in the seven principal battleground states — Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin — as well as Nebraska’s 2nd District in play for a possible tie, with the outcomes in those contests varying across all four cases, as shown in the table below.

Three of these four scenarios involve at least two of the three northern swing states shifting into the GOP column while the southern and western ones mostly end up in Democratic hands. That is somewhat at odds with our current forecast, which gives Trump a slightly better chance of flipping states in the Sun Belt than in the Frost Belt.

Of these, Scenario 1 is the most common — only with about a 1-in-1,000 shot of happening, however — and involves Trump holding onto North Carolina and flipping Pennsylvania and Michigan, while Harris retains Wisconsin, Arizona, Georgia and Nevada from President Biden’s 2020 showing. Next-most common is Scenario 2 (about a 1-in-1,200 shot), which includes mostly the same results, except with Harris and Trump trading Georgia and North Carolina’s identical electoral votes. Scenario 3, has about a 1-in-1,600 shot of happening — In it, Trump reclaims the Sun Belt states and Nebraska’s 2nd District while Harris holds onto the three northern battlegrounds. Scenario 4 is the least likely of this quartet (about a 1-in-2,000 chance), but it involves Trump carrying all the northern battlegrounds and Nevada while Harris wins the three southern swing states.

Beyond these scenarios, a handful of even more remote possibilities exist, but they have about a 1-in-10,000 (or worse) chance of occurring. Of those, the only ones that happen at least twice in our forecast’s roughly 34,000 simulations are four sets of state-by-state results in which Democratic-leaning New Hampshire or Virginia go Republican, while the outcomes in the seven key swing states vary (although Michigan remains in the Democratic column in all four). Out of the 12 other simulations that produce a tie, two involve Maine’s red-leaning 2nd District going Democratic.

Now, 538’s forecast assumes that electoral votes will go to the candidate who has won them based on the election results — but that’s not a 100-percent given. In 2020, for instance, Republicans in a handful of states that voted for Biden attempted to submit fake Republican electors to Congress in support of Trump’s unfounded claims that he’d actually won the election. And about two-thirds of House Republicans voted against certifying the results in one or both of Arizona and Pennsylvania. While these attempts to disrupt the 2020 election ultimately failed, they could definitely come into play in 2024, too.

Additionally, the electors of the Electoral College have some room for maneuver to cast votes for whomever they want, meaning a “faithless elector” could refuse to cast their vote for the ticket they were expected to support. For instance, in 2016, five Democratic electors and two Republican electors did not cast their votes for their party tickets but instead voted for other candidates. If an election outcome were to hang on every elector like in a 269-269 tie or another very close scenario, it’s unlikely that a faithless elector would cast what would amount to a protest vote — but it can’t absolutely be ruled out.

How the House would elect a president

If Congress finds the absence of a majority when it gathers on Jan. 6, 2025, to certify the Electoral College results, the 12th Amendment calls for the House to “immediately” choose the president from no more than the three candidates who won the most electoral votes. In a 269-269 tie, that would mean just the two major-party nominees (though a faithless elector scenario could technically introduce a third candidate). However, the choice is not based on a simple vote by the House’s 435 members as a whole, but instead is decided through a unique system whereby each state casts one vote according to the preferences of each member of that state’s House delegation — with states that have tied delegations potentially unable to cast a vote unless a member breaks with their party. To win, a candidate must win a majority of state delegations (26).

Under these rules, Republicans would likely hold an edge in this contingent election for president. 538’s House forecast shows a tight race for overall control of the House (which, of course, is based on which party controls more individual seats), but Republicans are far more likely than the Democrats to have an edge in the House delegations from at least 26 states in the next Congress. Currently, Republicans hold a majority of the House seats in 26 states compared with the Democrats’ 22, while two others (Minnesota and North Carolina) are tied. Yet if we take 538’s forecast and rate each state based on the race rating of its most competitive seat that would change party control, the GOP looks favored to control as many as 28 state delegations to Democrats’ 18 come January, with the four other states rated as toss-ups or likely to have tied delegations.

Based on this metric, Republicans are currently positioned to flip at least one state delegation in their favor, and possibly more. Thanks to a new Republican-drawn congressional map in North Carolina, the GOP is all but certain to shift that state from a 7-7 tied delegation to one with at least a 10-4 Republican edge. The GOP may also have a shot at capturing Michigan, where Democrats’ 7-6 lead looks tenuous due to two swingy Democratic-held seats that Republicans hope to flip. And, the GOP needs to flip only one highly competitive seat in Alaska, Pennsylvania or Virginia to control the House delegations in those states, while doing the same in Colorado and Maine would produce tied delegations in states currently held by Democrats.

Meanwhile, Democrats have no real path to 26 state delegations in a political environment where the presidential election produces a 269-269 tie in the Electoral College. At best, they could hope to hold onto their current states — save North Carolina — and flip two competitive Republican-held seats in Arizona to remain at 22 states. But if Democrats did better than that — say by flipping GOP-leaning seats in states such as Iowa, Montana and/or Wisconsin to potentially tie those delegations — that would probably signal that Democrats are doing well enough overall that Harris is winning at the top of the ticket, making the contingent election arithmetic purely academic.

We have only one precedent for a contingent election for president: the 1824 presidential race.* At that time, the Democratic-Republicans (known then as the Republicans but unrelated to today’s party) held a dominant political position amid the decline of the rival Federalist Party. As a result, four separate Democratic-Republicans contested the presidential race. To win, a candidate needed 131 electoral votes out of 261, but every contender fell short of that mark: General Andrew Jackson won 99 electoral votes, Secretary of State John Quincy Adams captured 84 electoral votes, Secretary of the Treasury William Crawford earned 41 electoral votes and House Speaker Henry Clay came in fourth with 37 electoral votes.

When the House gathered in February 1825 to choose a president, it could only consider Jackson, Adams and Crawford — the top-three recipients of electoral votes. While all were Democratic-Republicans, the party had internal divisions that would eventually split into a new party system by the 1828 election, and the votes in the House largely followed these cleavages. Clay, speaker of the chamber deciding the election, viewed Adams as most in line with his political views and publicly backed him. Accordingly, House members from the states that had supported Clay mostly shifted to Adams’ camp, which enabled Adams to win the 13 states that he needed to attain the presidency. (Jackson’s supporters soon argued that Adams and Clay had struck a “corrupt bargain,” whereby Clay became Adams’s secretary of state in exchange for Clay’s support. In 1828, Jackson went on to defeat Adams in a head-to-head rematch.)

Beyond its age and unusual political context, there are other reasons this 200-year-old case might not serve as much of a precedent for a modern contingent election, though. For one, due to the 20th Amendment, the newly-elected House would elect the president rather than the outgoing Congress that did the job in 1825, back at a time when the new Congress began in early March instead of on Jan. 3.

The choice that would confront the House in January 2025 would likely be more straightforward than the decision the chamber faced two centuries ago, too. While the House had to sort out an election involving three options ostensibly from the same party in 1825, it would likely be weighing two candidates from clearly opposing parties this time around — and the two-party division in the House today is readily apparent. As a result, it’s very difficult to imagine any Democrat or Republican deciding to vote for the other party’s candidate in a contingent election, a move that could effectively end the defector’s political career at a time of intense partisanship — especially if the vote is public. (In 1825, the House chose to conduct its contingent election in closed proceedings using a secret ballot, but the preferences of individual members soon became public knowledge.)

How the Senate would elect a vice president

Harris’s rise to become her party’s standard bearer is only the latest demonstration of how important the vice presidency is — and it’s the Senate that would decide who fills this key position if no vice presidential candidate earned a majority in the Electoral College. Unlike the House, a Senate contingent election vote would be based on a vote of the chamber as a whole, in which a candidate needs 51 of 100 votes to win. Additionally, the Senate can only consider the top two recipients of electoral votes, so even if a faithless elector casts a vote for another vice presidential candidate, the Senate would find itself picking between the major-party running mates: Ohio Sen. J.D. Vance and Minnesota Gov. Tim Walz.

The Senate math favors Republicans to win a contingent election for vice president, too, due to their seeming edge in the race to control the upper chamber. Currently, Democrats hold a 51-to-49 majority, including the four independents who caucus with them. But Republicans are essentially guaranteed to pick up a Democratic-held seat in dark red West Virginia to get to 50-50. Next, Democratic Sen. Jon Tester is at least a slight underdog to win reelection in Republican-leaning Montana, while Democratic Sen. Sherrod Brown is in a toss-up race to keep hold of his Ohio seat. Furthermore, Democrats are fighting to defend five seats in key presidential swing states (Arizona, Michigan, Nevada, Pennsylvania and Wisconsin), which could also bear some fruit for the GOP. On top of this, Democrats have only a small chance of capturing GOP-held seats in states like Texas and Florida.

Now, it’s possible that the Senate math would not be as straightforward as the House because the GOP caucus will include some senators who have been more openly critical of Trump. For instance, three Republican senators who voted to impeach Trump after the events of Jan. 6 — Bill Cassidy of Louisiana, Susan Collins of Maine and Lisa Murkowski of Alaska — will still be in office this January. One of them (particularly Murkowski, reelected in 2022 by Alaska’s ranked-choice voting system) might vote against Vance in a contingent election. Still, it’s more likely than not that the GOP would have the upper hand.

That being said, any unexpected delays in the House’s process to elect a president could also cloud the Senate picture. Based on the 12th Amendment, if the House fails to elect a president by Inauguration Day (Jan. 20), the Senate’s choice for vice president would serve as acting president until the House elects a president. And if the Senate also fails to choose a vice president, the Presidential Succession Act calls for the Speaker of the House to be next in line to serve as acting president. (Assuming that individual meets the presidential eligibility requirements and that the House is able to select one — the latter hasn’t happened easily in recent times.)

Just once has the Senate had to elect a vice president, and no, it actually wasn’t in 1824-25 when the House had to elect a president. In that election, Sen. John Calhoun won a majority of votes for the position in the Electoral College from a mix of electors who mainly backed Jackson or Adams for president. Rather, the one time the Senate elected a vice president was in February 1837, following the 1836 election when no vice presidential candidate won a majority of electoral votes because of intraparty dislike for Richard Mentor Johnson, the running mate of the victorious presidential candidate, Democrat Martin Van Buren.

In the 1836 election, the Van Buren-Johnson ticket carried states worth 170 electoral votes, surpassing the necessary 148 of 294 to win. However, while Virginia’s 23 Democratic electors voted for Van Buren, they voted for a different vice presidential candidate, leaving Johnson just shy of a majority with 147 electoral votes. A Kentucky representative and former senator, Johnson had alienated some in his party, especially in the South, due to his common-law marriage to an enslaved Black woman. However, a mostly party-line vote in the Senate easily elected Johnson as vice president, 33 to 16. Johnson’s political difficulties continued, though, as the Democrats refused to renominate him at their 1840 convention, although he served as the party’s de facto nominee in most states as Van Buren lost reelection.

***

It’s quite unlikely, but not impossible, that we could find ourselves in a situation where Congress has to decide the next president and vice president. But as the 1824-25 example shows, a contingent election to decide the next occupant of the White House could bring with it ample controversy and dissatisfaction that could carry on for years to come.

Footnote

*The Electoral College result in the 1800 election also required the House to choose the president, but that took place before the passage of the 12th Amendment (that election precipitated its implementation), so the 1824 election is the only contingent election for president to occur under the rules that would govern one today.

CORRECTION (Oct. 22, 2024, 10:15 a.m. Eastern): This article originally said that two Republican senators who voted to impeach Trump in 2021 will still be in the Senate in January 2025. That missed Louisiana Sen. Bill Cassidy, so the figure is actually three.