Skip to main content
ABC News
With Trump In The Race, The Battleground Is Everywhere

As the presidential campaigns shift their focus to the general election, there’s a big question about where the battleground will be. Which states will be competitive? What state will be the tipping point? The unusual candidacy of Donald Trump has some people wondering whether the electoral map will look different this year than it has in the past.

There are good reasons to think the map won’t change much; John Sides and Andrew Gelman, for example, recently suggested that, because states are moving increasingly in unison from election to election, it is unlikely that Trump can “scramble” the map.

But new data that I have analyzed with Bernard Fraga of Indiana University tells a different story.

The premise of this story is that although recent presidential elections have featured a predictable set of swing states and safe states, the extraordinary unpopularity of both major candidates (especially Trump) might mean that voter enthusiasm — and therefore turnout — might be different from that of past years. Additionally, Trump is so unlike recent Republican nominees that the past few presidential election results might not be very predictive of his performance.

Suppose, therefore, that we broaden our view and investigate where Democrats and Republicans have generated close contests beyond presidential races. Surprisingly, nearly the whole country has seen very close contests during the past few years.

Consider this map:


Fraga and I first looked at every general election for presidential, senatorial, gubernatorial, congressional and state legislative offices that voters saw between 2006 and 2012. In this period of just four election cycles, about 90 percent of Americans saw at least one close election between a Democrat and a Republican.

It is commonly known that most voters do not see competition in any one type of election in any given year. For example, in this period, less than 20 percent of voters saw a competitive U.S. House of Representatives race each year. But more than half of voters saw competition in some important office every year.

Of course, this analysis doesn’t even include primaries, municipal elections, or down-ballot races such as attorney general. In just the larger races, most of the country saw stiff Democratic-versus-Republican competition at least once. As the chart below shows, the typical voter in this period saw a competitive race for 1 in 4 contests for which he/she could vote.

0% 26m
1-10% 33
10-20% 54
20-30% 58
30-40% 28
40-50% 18
≥50% 9
Number of voters by share of elections in their area that were competitive, 2006-12

Competitive elections have a winning vote margin of less than 10 percentage points

Source: Fraga & HERSH

Historical data shows a similar pattern of robust competition. During the past 100 years, about half of the states have never gone more than four election cycles without a close statewide race. The only states that have gone a long time without competition are in the South, prior to the enfranchisement of black voters.


When looking across offices and years, the potential battleground seems expansive. Almost everywhere, Democrats and Republicans can generate a close race. In some ways, this evidence is surprising and in other ways it is obvious. It is surprising because party brands are pretty strong: voters are loyal to their parties. If all voters were partisans, always showed up to vote and never defected from their parties, we wouldn’t see this spread of competition.

But the evidence is also obvious because it is an enduring result of our electoral system, as laid out in the Constitution. Unlike many other democracies, we have frequent elections held at regular intervals. Sometimes elections are held when the Democrats happen to be popular and sometimes when the Republicans happen to be popular. So, in 2010 (a good year for Republicans), a different part of the country was competitive than in 2008 or 2006 when the Democrats were popular. This happens because of a combination of swing voting and enthusiasm/turnout differentials between partisan supporters.

The U.S. also has lots of elective offices and a federal system. Typically, both parties want to be competitive for all these offices and in every state, so they field viable candidates. Particularly in the governors’ races, we see candidates appealing to voters on partisan terms different from those that characterize national party competition. Some very Democratic states have Republican governors (New Jersey, Illinois) and some very Republican states have Democratic governors (Louisiana, Montana).

Some of these governors, such as Republican Charlie Baker in Massachusetts, are certainly more moderate than the typical Republican. Others (such as New Jersey Gov. Chris Christie) are broadly ideological but adeptly emphasize nonpartisan aspects of their candidacies.

Which brings us back to Trump. There really isn’t a great precedent for Trump, and so it isn’t obvious that we should look to past presidential elections to gauge where Trump will be competitive. Of course, presidential elections are quite different from gubernatorial elections, but if you want to know whether a Trump-like character could win executive office in a blue state such as New Jersey… voila!

The bottom line is that voters show a willingness to split their tickets or abstain from races depending on the context. In a year when their preferred party is unpopular, some voters sit out the election. A race that features an incumbent or not, a scandal or not, a big issue or not can affect how and whether voters cast ballots. The result is an extraordinarily large potential battleground for partisan races.

Listen to the latest episode of the FiveThirtyEight politics podcast.


The spread of competition that Fraga and I describe is also important, beyond what it can tell us about the 2016 election. In any one year, most voters do not see a competitive race for any one level of office (e.g., U.S. House, president). Some have concluded that this means we live in a “dubious democracy.” But the picture is much rosier than that. In a brief time, most voters do see competitive races. Nearly all voters are also represented by at least one Democrat and one Republican at some level of government. For voters who feel attached to a party, they almost always get a chance to win, no matter where they live.

For readers out there who feel bothered by the fact that they have not seen a competitive race in the past few years, take comfort: You’ll get your chance soon, maybe even this November.

And for readers who take comfort in the stability in competition that has characterized recent presidential elections, gird yourself.

Eitan Hersh is an assistant professor of political science at Yale University.