To a casual observer of American politics, it might be confusing what constitutes a “battleground state” in a presidential election. For instance, in the most recent presidential election, a state might have voted one way by as much as double-digits, but pundits talk as if it might vote the other way in the next. Why would the state in question be treated as competitive?
To put it simply, there is no official definition of what makes a battleground state other than a state where the result of an election is in any amount of doubt. However, there is a mathematical way of measuring how competitive a state might be: the lean of the state relative to the nation. For example, in 2016, Minnesota voted for Hillary Clinton by 1.5% while she was winning the popular vote nationally by 2.1%. This means that Minnesota leaned 0.6% more Republican than the country.
The lean of a state relative to the nation can also give a hint as to how the state could vote in the future. Below are a series of graphs of how battleground states have voted relative to the country in the last five presidential elections. When the states are grouped together by region, a pattern emerges:
Besides Florida, every battleground state in the sun belt voted more Democratic relative to the nation in 2016 than in 2000. While past trends are not always a good indicator of the future, the consistent trends of some states give reason to believe they are more likely than not to continue in the direction they are going.
For example, North Carolina and Virginia have steadily trended towards Democrats in the last four elections. North Carolina is expected to be heavily contested in next year’s presidential race, while the Trump campaign has given signals that it will not spend much time in Virginia.
Colorado trended towards Democrats in every election except 2012, possibly due to Republican Mitt Romney’s personal appeal in the mountain west. Like Virginia, Colorado has a large population of college-educated white voters who are uncomfortable with Donald Trump, giving Democrats a clear edge in the state.
Arizona, Georgia and Texas all saw relatively strong trends towards Democrats in 2016, ranging from 4 to 8%. This was likely because of traditionally Republican voters being repelled by Donald Trump and switching their vote to Hillary Clinton or Gary Johnson. Barring an unlikely removal from office, Trump will be on the ballot again in 2020, meaning these states will most likely either remain relatively stable compared to the nation, or trend more towards Democrats.
The two remaining states, Florida and Nevada, both have high proportions of Hispanic voters which typically determine the outcomes of statewide and federal races. Although Florida has been elusive for Democrats in many recent races, it has stayed relatively stable in its lean relative to the nation. Democrats have had much more success in Nevada, where they now control both Senate seats, the governorship, both houses of the legislature, and four of the five row offices. While the trend towards Republicans in presidential elections since 2008 might look concerning for Democrats, their recent success leads one to believe that it could simply be a reversion to the mean after Barack Obama’s strong wins there in 2008 and 2012 and that Nevada will most likely vote close to the national average in 2020.
For the sake of similar patterns, I am putting the Midwest and Northeast into one category.
The states in this category might as well be the polar opposite of the battlegrounds in the sun belt. Every battleground state in these regions except New Hampshire voted more Republican relative to the nation in 2016 than they did in 2000.
Some of the trends in 2008 and 2012 might have been somewhat artificial as Barack Obama had unique strength in the upper Midwest, being from the region himself. This is most evident in Michigan and Wisconsin.
When comparing the trend history of these states to the states in the sun belt, there is one clear difference: the states in the Midwest and northeast were not clearly trending in one direction or the other prior to 2016, except perhaps Pennsylvania, which trended towards Republicans in three of the last four elections. This could imply that these states are still very much up-for-grabs.
As a counterpoint, one could argue that some of the trends between 2012 and 2016 were quite dramatic, and it would seem unlikely that there would be as strong a shift in the opposite direction in 2020.
One could also argue the reality is somewhere in between the two arguments. While Trump’s over-performance in some states, such as Iowa and Ohio, might have been a fluke, the result of Hillary Clinton being a poor fit for the region or a combination of both, Trump could possibly push a few of these states even more Republican-leaning compared to the nation. The state that is perhaps the most likely to do so is Wisconsin, which has the highest percentage of white voters without a college degree among the three decisive states (the others being Michigan and Pennsylvania) in the 2016 election.
As stated earlier, keep in mind that the recent trends of a state does not necessarily predict what it will do in the near future and should not be a huge factor when making a prediction. Every state has its own political culture, after all.