clock menu more-arrow no yes mobile

Filed under:

The Electoral College, explained

The bizarre, patchwork, Frankenstein’s monster of a system Americans use to choose their president.

Andrew Prokop is a senior politics correspondent at Vox, covering the White House, elections, and political scandals and investigations. He’s worked at Vox since the site’s launch in 2014, and before that, he worked as a research assistant at the New Yorker’s Washington, DC, bureau.

Donald Trump has a very small chance of actually winning more votes than Joe Biden in the presidential election — just 3 percent, according to FiveThirtyEight’s forecast model.

But he still has a somewhat better shot of winning anyway, the same way he did last time: through the magic of the Electoral College.

The presidential election isn’t decided by the number of votes you get overall. It’s decided by whether you get more votes in the right states. Whichever candidate ends up winning contests that add up to 270 electoral votes is the winner.

In other words, it doesn’t necessarily matter that polls show Biden ahead by nearly 9 percentage points nationally. What matters is the exact breakdown of the votes in key politically divided states like Florida, Pennsylvania, and North Carolina.

It gets weirder. In the election itself, the states aren’t technically choosing a presidential candidate. They’re choosing representatives in the Electoral College — the electors — who are actual people who will cast the electoral votes determining the president in December.

Overall, the Electoral College is a patchwork Frankenstein’s monster of a system, which in the best of times merely ensures millions of Americans’ votes are irrelevant to the outcome because they don’t live in competitive states, and in the worst of times could be vulnerable to a major crisis. And yet, for the foreseeable future, it’s the system we’re stuck with.

1) What is the Electoral College, and how does it work?

A generous definition of the Electoral College swing states and districts for 2020.
Andrew Prokop/Vox

The presidential election is generally portrayed as a battle to win states and their accompanying electoral votes. If Biden wins Vermont, he’ll get its three electoral votes. If Trump wins Alaska, he’ll get its three electoral votes. Whoever gets to 270 or more electoral votes first — a majority of the 538 total — wins the election.

So rather than simply trying to win the most actual votes in the country, a presidential campaign must try to put together a map of state victories that will amass more than 270 electoral votes. That’s the simplified version.

What’s happening under the hood, though, is more complicated. When people go to the polls to vote for a presidential candidate, what they are actually doing is voting for each party’s nominated slate of electors in their respective states (or, in the case of Maine and Nebraska, in congressional districts too).

So when Donald Trump won Alaska in 2016, the practical effect was that the Republican Party’s nominated elector slate there — former Gov. Sean Parnell, Jacqueline Tupou, and Carolyn Leman — officially became Alaska’s three electors.

This process repeated itself across the country, resulting in the selection of the Electoral College — the 538 electors who cast their votes for president in their respective states in December. (In the modern era, this ceremonial occasion has been a formality that reiterates an outcome known well in advance — though there was a bit of last-minute drama in 2016.)

2) But the outcome of the presidential election is really just settled in a few swing states, right?

The Democratic and Republican parties have each developed solid bases in a series of states that are all but certain to vote for them in a presidential year. But the Electoral College winner will be determined by those few swing states that are more divided politically and look like they could go either way.

The swing states’ dominance is a consequence of the fact that almost every state chooses to allot all its electoral votes to whoever comes in first place statewide, regardless of his or her margin of victory.

That is, it doesn’t matter whether Biden wins New York by a 30 percent margin or a 10 percent margin, since he’ll get the same amount of electoral votes either way. But the difference between winning Florida by 0.1 percent and losing it by 0.1 percent is crucial, since 29 electoral votes could flip.

Naturally, then, when the general election comes around, candidates ignore every non-competitive state — meaning the vast majority of the country — and pour their resources into the few that tend to swing back and forth between Republicans and Democrats. That’s the best strategy for reaching that magic number, 270.

3) That seems unfair.

Well, there’s a lot that’s unfair — or at the very least undemocratic — about the Electoral College.

For one, the winner of the nationwide popular vote can lose the presidency. In 2000, Al Gore won half a million more votes than George W. Bush nationwide, but Bush won the presidency after he was declared the winner in Florida by a mere 537 votes. In 2016, Hillary Clinton won an even bigger popular vote victory — by 2.1 percentage points — but she lost Wisconsin, Pennsylvania, and Michigan, each by a less than 1 percentage point margin, and so she lost the presidency. Electoral College/popular vote splits happened in 1876 and 1888 as well.

Second, there’s swing state privilege. Millions of votes in safe states end up being “wasted,” at least in terms of the presidential race, because it makes no difference whether Clinton wins California by 4 million votes, 400,000 votes, or 40 votes — in any scenario, she gets its 55 electors. Meanwhile, states like Florida and Ohio get the power to tip the outcome just because they happen to be closely divided politically.

Third, a small-state bias is also built in, since every state is guaranteed at least three electors (the combination of their representation in the House and Senate). The way this shakes out in the math, the 4 percent of the country’s population in the smallest states end up being allotted 8 percent of Electoral College votes.

And fourth, there’s the possibility for those electors themselves to hijack the outcome.

4) Wait, the electors can hijack the outcome of the presidential election? What?

For decades, it’s been assumed that the 538 electors will essentially rubber-stamp the outcome in their respective states, and they mostly have. But there’s scarily little assurance that they’ll actually do so.

According to the National Conference of State Legislatures, about 30 of the 50 states have passed laws “binding” their electors to vote in accordance with the presidential popular vote in their state. But in most, the penalty for not doing so is only a fine. The Supreme Court upheld the constitutionality of these penalties this year — but other states still don’t bind electors, and the justices didn’t require electors to abide by the vote in their state.

This issue hasn’t been a big deal in the past because, almost always, the parties do a good enough job of vetting their respective electoral slates to ensure that they will indeed loyally back their party’s presidential nominee.

But there have been a few rogue, faithless, or just plain incompetent electors over the years — and their votes have all been counted as cast.

  • In 1837, rogue electors from Virginia briefly blocked the seating of the vice president-elect because they were offended that he had a multiracial common-law wife. (The Senate overrode them.)
  • A Democratic elector from Tennessee cast his ballot for segregationist third-party candidate Strom Thurmond in 1948, and a Republican elector from North Carolina voted for segregationist third-party candidate George Wallace in 1968.
  • In 2000, an elector from Washington, DC, withheld an electoral vote from Al Gore, because she wanted to protest the fact that DC didn’t have representation in Congress.
  • Perhaps most bizarrely of all, in 2004, an elector from Minnesota who was supposed to vote for John Kerry for president instead voted for John Edwards. (It’s believed that this was an accident, but since the votes were cast anonymously, we don’t really know for sure.)
  • And 2016 brought a record seven faithless electors who had their votes cast and counted. Two Trump electors defected to vote for Ron Paul and John Kasich. And five Hillary Clinton electors defected — three voted for Colin Powell, one for Bernie Sanders, and one for Native American activist Faith Spotted Eagle.

Rogue electors have never been numerous enough to actually affect the outcome of a presidential race. But it’s unclear if they would be stopped, should they choose to do so.

Now, some defenders of the system have taken the comforting view that the power of electors to go rogue is a good thing, since they could conceivably save America from a popularly elected majoritarian candidate who could oppress the minority.

But it seems just as likely, if not more likely, that electors could install that candidate with dictatorial tendencies against that popular will. Perhaps some electors are wise sages with better judgment than the American people, but others are likely malign, corrupt, or driven by their own idiosyncratic beliefs. (You’ll notice above that several of those historical rogue electors in history had racist motivations.)

In any case, if we had a process in which the electors were notable citizens who were chosen because they’re supposed to exercise good judgment, maybe the defense would make sense. But in the system we have today, the electors are chosen specifically to be rubber stamps. As a result, there’s incredibly little attention paid to who those electors even are outside internal party machinations in each state. Any defection by an elector would, essentially, be a random act that could hold our system hostage, in an election that truly hinged on a handful of electoral votes.

5) Why do we use such a bizarre system, anyway?

The Constitutional Convention of 1787
The Constitutional Convention of 1787.
Hulton Archive/Getty Images

The Electoral College is, essentially, a vestigial structure — a leftover from a bygone era in which the founders specifically did not want a nationwide vote of the American people to choose their next president.

Instead, the framers gave a small, lucky group of people called the “electors” the power to make that choice. These would be some upstanding citizens chosen by the various states, who would make up their own minds on who should be the president (they’d have to vote on the same day in their respective home states, to make it tougher for them to coordinate with each other).

The Constitution remained silent on just how these elite electors would be chosen, saying only that each state legislature would decide how to appoint them. Initially, some state legislators picked the electors, while other states had some form of statewide vote in which the electors themselves would be candidates.

But over the new nation’s first few decades, two powerful trends in American politics brought attention to the Electoral College system’s shortcomings — the rise of national political parties that would contest presidential elections, and the growing consensus that all white men (not just the elite) should get the right to vote, including for president.

The parties and states responded to these trends by trying to jury-rig the existing system. Political parties began to nominate slates of electors in each state — electors they believed could be counted on to vote for the presidential nominee. Eventually, many states even passed laws requiring electors to vote for their party’s presidential nominee.

Meanwhile, by the 1830s, almost every state had changed its laws so that all electors were chosen winner-take-all through a statewide vote, according to Richard Berg-Andersson. The point of all this was to try to make the presidential election function like ordinary statewide elections for governor or senator, at least within each state.

6) Is the Electoral College system biased in Republicans’ favor?

The Electoral College delivered two of the five most recent elections to Republicans even though the Democrat won the popular vote (George W. Bush in 2000, and Donald Trump in 2016).

But more broadly, the way this question is generally measured is by comparing the margin for the winning candidate in the popular vote to that candidate’s margin in the “tipping point” Electoral College state. That’s the state that would get the candidate 270 Electoral Votes if they lost every other state where their margin is smaller.

Now, political coalitions shift over time, and in 2004, 2008, and 2012 it was Democrats who had a slight edge in the tipping point state compared to the popular vote.

  • In 2004, John Kerry lost the popular vote by 2.4 percentage points, but he lost the tipping point state, Ohio, by 2.1 percentage points (a 0.3 percentage point edge for Democrats).
  • In 2008, Barack Obama won the popular vote by 7.2 percentage points, and he won the tipping point state, Colorado, by 8.9 percentage points (a 1.7 percentage point edge for Democrats).
  • In 2012, Obama won the popular vote by 3.9 percentage points but won Colorado (again the tipping point state) by 5.4 percentage points (a 1.5 percentage point edge for Democrats).

But shifts in the parties’ support bases since the rise of Donald Trump may have made this irrelevant. In 2016, Trump lost the popular vote by 2.1 percentage points, and won the tipping point state (Wisconsin) by a 0.77 percentage point margin. So he had nearly a 2.9 percentage point Electoral College edge — significantly bigger than Kerry, Obama, or even Bush in 2000 managed.

And polls in 2020 show that Electoral College edge may have even grown. According to FiveThirtyEight’s averages, Joe Biden is winning the national popular vote by 8.6 percentage points, Biden is ahead in the tipping point state — Pennsylvania — by a smaller amount, 4.8 percentage points. That would mean the Electoral College is giving Trump a 3.8 percent edge. (Though of course, we’ll have to wait for the actual results to know for sure.)

As for the future, this look at recent history should be a reminder that political coalitions shift over time and trends can be difficult to predict. One potential looming problem for Republicans in the Electoral College is that one of their most important “safe states” — Texas, with 38 electoral votes — is becoming more competitive. Then again, perhaps the GOP will continue to make gains in Pennsylvania, Michigan, and Wisconsin (which combine for 46 electoral votes) over the longer term.

7) Are there arguments for the Electoral College?

It’s tough to argue with a straight face that this bizarre system is inherently better than just a simple vote. After all, why doesn’t any state elect its governor with an “Electoral College” of various counties? Why does pretty much every other country that elects a president use a simple popular vote, or a vote accompanied with a runoff?

Now, you can argue that the Electoral College’s seeming distortions of the popular will aren’t as bad as they seem — for instance, by pointing out that swing states tend to swing along with the nation rather than overriding its will, or that the popular vote winner almost always wins. But, of course, that’s not guaranteed to always be the case, and the recent major exceptions (the 2000 and 2016 elections) were incredibly consequential ones.

Others try to fearmonger about the prospect of a contested nationwide recount — which, sure, would be ugly, but if you’ll recall, the Florida recount was also extremely ugly. And since there are so many more votes cast nationally, it’s much less likely that the national vote would end up a near tie than that a tipping point state’s vote would end up as a near tie. A recent paper by Michael Geruso and Dean Spears looked at this question, and found that it was 40 times more likely that a small number of potentially disputed ballots would determine the outcome under the Electoral College system, compared to a national popular vote.

Some argue that the Electoral College ensures regional balance, since it’s mathematically impossible for a candidate with overwhelming support from just one region to be elected. But realistically, the country is big and broad enough that this couldn’t happen under a popular vote system either — any regional candidate would need to get some support outside his or her region.

When we get down to brass tacks, the most serious objections to reforming the Electoral College come from rural and small-state elites who fear that under a national popular vote system, they’d be ignored and elections would be decided by people who live in cities.

Gary Gregg of the University of Louisville wrote in 2012 that eliminating the Electoral College would lead to “dire consequences.” Specifically, he feared that elections would “strongly tilt” in favor of “candidates who can win huge electoral margins in the country’s major metropolitan areas.” He continued:

If the United States does away with the Electoral College, future presidential elections will go to candidates and parties willing to cater to urban voters and skew the nation’s policies toward big-city interests. Small-town issues and rural values will no longer be their concern.

And Pete du Pont, a former governor of Delaware (three electoral votes), has made a similar case, calling proposals for a national popular vote an “urban power grab.”

But a national popular vote system wouldn’t devalue the votes of people who live in rural states and small towns. It would accurately value them by treating them equal to people who live in cities, rather than giving them an extra weighting. Furthermore, small-state interests are built into the Senate’s math (where Delaware absurdly gets as many senators as California), and many House districts are rural. So rural and small-state areas are hardly hurting for national political representation.

Sure, candidates might end up spending less time stumping in the rural areas that currently happen to be lucky enough to fall within the borders of swing states, and more time in urban centers. But plenty of rural areas in non-swing states get totally ignored by the presidential candidates. And even if they didn’t, is that really a convincing rebuttal to the pretty basic and obvious argument that in the most important electoral choice Americans make, their votes should be treated equally?

8) Is there any hope that the US will ditch the Electoral College someday?

For decades, polls have shown that large majorities of Americans would prefer a popular vote system instead of the Electoral College. For instance, a September 2020 Gallup poll found that 61 percent of US adult respondents wanted to do away with it, and 38 percent wanted to keep it. (Naturally, most Democrats want to get rid of it and most Republicans want to keep it, but most independents side with the abolishers.)

But to ditch the Electoral College entirely, the US would have to pass a constitutional amendment (passed by two-thirds of the House and Senate and approved by 38 states) — or convene a constitutional convention (which has never been done but would have to be called for by 34 states). Either method is vanishingly unlikely, because each would require many small states to approve a change that would reduce their influence on the presidential outcome.

There is one potential workaround, however: the National Popular Vote Interstate Compact, a clever proposal that uses the Constitution’s ambiguity on electors to its own ends.

A state signing on to the compact agrees that it will pledge all its electors not to its state winner but to the victor in the national popular vote — but only if states controlling 270 or more electoral votes have agreed to do the same. If they do, and everything works as planned, then whoever wins the popular vote will necessarily win the electoral vote too.

It’s a fun proposal that’s already been enacted into law by 15 states (including massive California and New York) and the District of Columbia, which together control 165 electoral votes. But there’s one big obstacle: All the states that have adopted it are solidly Democratic, with zero being Republican or swing states.

States that have adopted the National Popular Vote Interstate Compact, according to their vote in the past three presidential cycles. They add up to 196 electoral votes for Democrats, 0 for Republicans
States that have adopted the National Popular Vote Interstate Compact, according to their vote in the past three presidential cycles.
Andrew Prokop/Vox

So unless a bunch of swing states decide to reduce their own power, or Republican politicians conclude that a system bringing the power of small and rural states in line with that of big urban centers is a good idea, the compact isn’t going to get the support it needs. (It also wouldn’t solve the rogue elector problem.)

So we’re likely stuck with the Electoral College for some time. Your safe state vote might be wasted, or it might even be subverted by rogue electors.

But at least you’ll get to draw fun maps.

A version of this article was originally published in 2016. It has been updated for the 2020 election.

Sign up for the newsletter Today, Explained

Understand the world with a daily explainer plus the most compelling stories of the day.