Get more FiveThirtyEight
From one presidential election to the next, the battleground states that make — or break — the election remain largely the same. But that doesn’t mean there aren’t gradual (and sometimes, not so gradual) shifts underway. We zoomed in on how 16 battleground states have voted relative to the country as a whole since 2000 — or how much more Republican or Democratic they are relative to the nation1For instance, Hillary Clinton won the national popular vote by about 2 points in 2016, but President Trump won Florida by about 1 point in the same election, so Florida was about 3 points more Republican than the country. — and we found an electoral map undergoing a series of changes, some steady and others abrupt.
Take Iowa and Ohio, which went from uber-competitive states to near blowouts for President Trump in 2016. Or Maine and Michigan, which hadn’t been all that competitive in 2008 or 2012, but lurched to the right in 2016. In other words, 2016 marked a significant departure from how these states had voted in recent years; each state swung 7 points or more to the right, the biggest swings in that election.
One explanation for why these four states moved so suddenly to the right is that they each have a large share of voters (at least 55 percent2Of the population age 25 years or older.) who are non-Hispanic white with less than a bachelor’s degree, a bloc that moved sharply toward the GOP in 2016. Additionally, Iowa and Maine rank among the most rural states in the country, which is another predictor of GOP-leaning politics. FiveThirtyEight’s forecast currently expects these states to step slightly to the left in 2020, but there’s a lot of uncertainty here, especially in a state like Maine where the error bars are particularly large. The next five states also shifted to the right in 2016, but they didn’t veer quite as far as the previous four did.
One reason why these states didn’t lurch as far to the right is that four of them3Minnesota, Nevada, Pennsylvania and Wisconsin. have at least one large metropolitan area that votes heavily Democratic. This offsets the rest of those states, which usually vote far more Republican. However, as was true in the first four states we looked at, there has been a slow yet noticeable move to the right in these four states over the last several elections. FiveThirtyEight’s forecast anticipates that some of these states, like New Hampshire and Wisconsin, might bounce back slightly to the left in 2020, but also that others, like Minnesota, may continue to shift to the right.
On the other end of the spectrum, some 2020 battleground states moved to the left — considerably so — in the last election. These states predominantly lie in the South and West, and the following trio of traditionally red states could all be up for grabs this November.
Arizona, Georgia and Texas all moved at least 4 points to the left in 2016, and it’s possible they’ll move even farther in 2020. After all, the 2018 midterm elections showed these states could elect Democrats statewide, or at least, come very close. Democrats won a U.S. Senate seat in Arizona for the first time since 1988, while Republicans only narrowly won Texas’s Senate race and Georgia’s gubernatorial contest.
What explains the leftward shift in these traditionally Republican states? For one thing, these states are more racially and ethnically diverse than most of the other states we’ve looked at — Arizona and Texas have large Hispanic populations, for instance, while Georgia has a sizable Black electorate — and people of color tend to vote more Democratic. But these fairly urban states have also seen their major metropolitan areas such as Atlanta, Dallas, Houston and Phoenix become increasingly Democratic because of the surge in college-educated voters. At present, the FiveThirtyEight forecast anticipates these states will lean similar to how they did 2016, although further shifts to the left are plausible.
For Democrats, the hope would be that those three states trend in ways similar to Colorado and Virginia, two formerly red states whose diverse and highly educated electorates have moved them to the left over the past two decades.
Colorado’s population is about one-fifth Hispanic and Virginia’s is about one-fifth Black, and both are only about two-thirds white. And white voters in these states are more likely to hold at least a four-year college degree than in the other states we’ve examined. Driven by increasingly Democratic vote shares in suburban and urban areas — especially around Denver and Washington, D.C. — Colorado and Virginia have moved far enough to the left that, in an environment in which Joe Biden leads by about 9 points nationally, they lie at the periphery of the competitive states.
That said, Florida and North Carolina are also racially diverse and home to a decent number of highly educated voters, but they haven’t become liberal bastions. Instead, they have tended to vote a bit to the right of the country with little variation in recent years.
Florida is a hard state to categorize politically. It has an elderly population that usually leans toward the GOP, but it also has a large Hispanic and Black population that leans Democratic -- with the caveat that a large share of its Latino vote is Cuban American, a group that has shifted toward Democrats over the last decade but remains far more Republican-leaning than other groups of Latinos. North Carolina is also a swing state, even though it has a fairly consistent Republican lean. North Carolina’s white college-educated population share isn’t that much smaller than Virginia’s, but it has a larger share of white voters who don’t have a four-year degree. Additionally, North Carolina’s white voters are somewhat more Republican-leaning, and the state tends to be more rural than Virginia. As things stand, the forecast model expects these states to vote more or less like they did in 2016 — so they’ll be as competitive as ever.
This is all to say that 2020 will be a pivotal year in understanding underlying trends within the Electoral College. Will states like Iowa or Ohio move further to the right? Georgia and Texas further to the left? Or should we expect more of a reversion to the mean? Suffice it to say that, with Trump on the ballot this November, many of the forces we saw in 2016 will be prevalent again in 2020. But those forces may also define the parties moving forward, which means this election could also tell us a great deal about how the electoral map will look beyond 2020.