Neighboring houses display signs for opposing presidential candidates in Northumberland, Pa., a state being watched with great anticipation in the 2020 Election.
Paul Weaver/Pacific Press/LightRocket—Getty Images
Updated: November 3, 2020 9:51 AM EST | Originally published: November 3, 2020 9:32 AM EST

While Americans across 50 states and Washington, D.C. are voting on Election Day—or already voted early—some states’ votes will be more closely watched than others. Over the last half-century, the share of Americans who vote for the same party in every election has increased; unsurprisingly, some states are consistent in just the same way. And so all eyes will be on the handful states that do vary their votes: the swing states.

So how did those states get to be so influential?

Before the 1970s, one could find rare, sporadic references to the term swing state, but back then it meant something more like a bellwether state—a state that political journalists looked to because they believed its residents would vote for the eventual winner of the Presidential election. “The phenomenon has been around for along time but had a different label,” says Alex Keyssar, author of Why Do We Still Have the Electoral College? and professor of History and Social Policy at Harvard University. Political pundits used other terms to describe swing states, such as “doubtful” states.

The 1976 election, when outsider Georgia Governor Jimmy Carter defeated incumbent Gerald Ford, was rife with swing states, Keyssar says, in part because it was still in the period when many Southern Democrats were shifting to the Republican Party. That situation contributed to a situation in which a high number of states could have gone either way.

But the difference between now and then was about more than just vocabulary and timing.

“The two political parties had been more ideologically diverse, so you had less polarization,” says David Schultz, an editor of the anthology Presidential Swing States: Why Only Ten Matter. “As polarization becomes more rigidified since the 1970s—with people less likely to split-ticket vote; people less likely to vote Republican in one election, Democrat the next—that’s kicked in the notion that states have also more firmly become partisan one way or another.”

Schultz and his co-editor Stacey Hunter Hecht define a swing state as one that has frequently boasted a 5% or smaller difference in votes for the two major candidates, a vote result that often matches the national popular vote, and a history of flipping between parties. With the election of George H.W. Bush in 1988, the swing-state situation had started to evolve to the way things stand today, with just a handful of states as perennial battlegrounds.

“Between 1888 and 1988 it was common for close national elections to have 30 to 40% of the states to be competitive, within five percentage points. The 1960 election is a great typical example, a very close election and it had 20 competitive states. Also in that period, the roster of competitive and swinging states would change frequently like a revolving cast of characters,” says Scott McLean, a contributor to Presidential Swing States and a professor of Political Science at Quinnipiac University. “But since 1988, the swing club has been a smaller group, even when the national election is close, with most of the states reappearing for multiple election cycles. “

Schultz and Hecht’s book concentrates on the period beginning with the 1988 election but, even since it was released in 2015, the number of swing states has continued to decrease.

“Since I did the first edition of the book in 2015 and second edition in 2018, states like Colorado, Nevada and New Mexico—which were swing—have moved more firmly into the Democratic side, and states like Missouri have fallen more firmly to the Republican side,” says Schultz. The 2018 edition added Pennsylvania, which went from being reliably Democratic to voting Republican in the 2016 presidential election.

And Schultz now believes the 2020 election will come down to the same states it came down to in 2016: Pennsylvania, Wisconsin and Michigan.

These are states where working-class white voters without college degrees, who in the past tended to vote for Democrats, started voting Republican when their well-paying manufacturing jobs disappeared. They also have big college-educated populations and renowned state and private universities, as well as urban areas that are racially diverse and Democratic-leaning. “A combination of all those factors put them in a really nice position to look kind of swingy,” as Schultz puts it. “What you really have here is the juxtaposition of college-educated suburban women moving in one direction, and white Caucasian males without college degrees moving in a different direction. The intersection of those two populations within those three swing states, that’s the whole election right there.”

Get your history fix in one place: sign up for the weekly TIME History newsletter

And Americans know it: the images of electoral maps showing red Republican states and blue Democratic states has become an emblematic illustration of that polarization of the electorate.

In recent years, the media has played a big role in popularizing the term swing states, as the increase in partisan TV channels and news websites has allowed consumers to only listen to like-minded pundits. In contrast, back in the 1960s, there were only a few TV channels. “Those networks converged towards the center to maximize audience share, but now you’ve got lots of different news sources catering to different constituencies and interests,” says Schultz. The proliferation of partisan media sources helped exacerbate the polarization that has contributed to more voters who consistently support either party, he argues.

And that media focus on swing states only increased following the heated 2000 election—which came down to a single state, Florida, and ended in a Supreme Court decision halting a recount of votes in the state—and the 2004 election, when some alleged that purges of the voter rolls in Ohio tipped the result in Bush’s favor. According to the 2015 edition of Presidential Swing States, a keyword search of “swing states” in the New York Times turned up 139 references to the term in the 2000 election, 321 in 2004, 231 in 2008, and 581 in 2012.

Media fascination with swing state coverage is made further possible by the narrowing of the number of swing states and the rise of 24-hour cable news. Swing states also become less expensive to poll, so those polls have proliferated and continued to feed interest in the horse race in key swing states.

So what’s the impact of presidential elections coming down to a handful of swing states on American society?

McLean, for one, argues that the swing state dynamic has exacerbated distrust in American government and institutions that stems back to the 1970s Watergate scandal. “The sad effect is that the majority of voters are largely relegated to the role of spectators,” he says. “It is no wonder, then, that when the same swing states always seem to decide national elections, it can erode the faith that elections are the voice of the people.”

Write to Olivia B. Waxman at olivia.waxman@time.com.

Read More From TIME

Related Stories

EDIT POST