In American politics, the term swing state refers to any state that could reasonably be won by either the Democratic or Republican presidential candidate. These states are usually targeted by both major-party campaigns, especially in competitive elections.  Meanwhile, the states that regularly lean to a single party are known as safe states, as it is generally assumed that one candidate has a base of support from which they can draw a sufficient share of the electorate.
Due to the winner-take-all style of the Electoral College, candidates often campaign only in competitive states, which is why a select group of states frequently receives a majority of the advertisements and partisan media.  The battlegrounds may change in certain election cycles, and may be reflected in overall polling, demographics, and the ideological appeal of the nominees. Election analytics website FiveThirtyEight identifies the states of Colorado, Florida, Iowa, Michigan, Minnesota, Ohio, Nevada, New Hampshire, North Carolina, Pennsylvania, Virginia, and Wisconsin as "perennial" swing states that have regularly seen close contests over the last few presidential campaigns. 
In American presidential elections, each state is free to decide the method by which its electors to the Electoral College will be chosen. To increase its voting power in the Electoral College system, every state, with the exceptions of Maine and Nebraska, has adopted a winner-take-all system, where the candidate who wins the most popular votes in a state wins all of that state's electoral votes. The expectation was that the candidates would look after the interests of the states with the most electoral votes. However, in practice, most voters tend not to change party allegiance from one election to the next, leading presidential candidates to concentrate their limited time and resources campaigning in those states that they believe they can swing towards them or stop states from swinging away from them, and not to spend time or resources in states they expect to win or lose. Because of the electoral system, the campaigns are less concerned with increasing a candidate's national popular vote, tending instead to concentrate on the popular vote only in those states which will provide the electoral votes it needs to win the election, and it is far from unheard of for a candidate to secure sufficient electoral votes while not having won the national popular vote, such as in the case of the 2016 presidential election.
From recent past electoral results, a Republican candidate can expect to easily win most of the mountain states and Great Plains, such as Idaho, Wyoming, the Dakotas, Montana, Utah, Kansas, Oklahoma, and Nebraska, as well as most of the South, including Alabama, Mississippi, Louisiana, Texas, Arkansas, Tennessee and South Carolina. A Democrat usually takes the Mid-Atlantic states, including New York, New Jersey, Maryland and Delaware, along with New England, particularly Vermont, Massachusetts, Rhode Island, and Connecticut, and the West Coast states of California, Oregon and Washington.
However, states that consistently vote for one party at the presidential level occasionally elect a governor of the opposite party; this is currently the case in Massachusetts, Maryland, Illinois, New Mexico and Vermont, which all have Republican governors, as well as in Louisiana and Montana, which currently have Democratic governors. Even in presidential election years, voters may split presidential and gubernatorial tickets. In 2016, this occurred in Vermont and New Hampshire, which elected Republican governors even as Hillary Clinton won both states, while Montana and North Carolina elected Democratic governors despite also voting for Donald Trump.
In Maine and Nebraska, the apportionment of electoral votes parallels that for Senators and Congressional Representatives. Two electoral votes go to the person who wins a plurality in the state, and a candidate gets one additional electoral vote for each Congressional District in which they receive a plurality. Both of these states have relatively few electoral votes – a total of 4 and 5, respectively. Neither Maine, which is generally considered a Democratic-leaning state, nor Nebraska, typically thought to be safely Republican, would become battlegrounds in the event of a close national race. Despite their rules, only once has each state 'split' its electoral votes – in 2008, when Nebraska gave 4 votes to Republican John McCain, and one to Democrat Barack Obama; and in 2016, when one of Maine's congressional districts was won by Donald Trump, and the other district and the state itself were won by Hillary Clinton.
States where the election has a close result become less meaningful in landslide elections. Instead, states which vote similarly to the national vote proportions are more likely to appear as the closest states. For example, the states in the 1984 election with the tightest results were Minnesota and Massachusetts. A campaign strategy centered on them, however, would not have been meaningful in the Electoral College, as Democratic nominee Walter Mondale required victories in many more states than Massachusetts, Republican Ronald Reagan still would have won by a large margin.  Instead, the tipping point state that year was Michigan, as it gave Reagan the decisive electoral vote. The difference in Michigan was nineteen percentage points, quite similar to Reagan's national margin of eighteen percent.  Michigan would have been more relevant to the election results had the election been closer.
Similarly, Barack Obama's narrow victory in Indiana in the 2008 election inaccurately portrays its status as a battleground. Obama lost Indiana by more than ten percentage points in the closer 2012 election, but triumphed despite losing fewer Republican states like North Carolina, Arizona, Georgia, Missouri, and Montana.  In 2012, the states of North Carolina, Florida, Ohio, and Virginia were decided by a margin of less than five percent. However, none of them were considered the tipping-point state, as Romney would not have been able to defeat Obama even if he had emerged victorious in all of them. Rather, Colorado was most in-step with the rest of the country. Coloradans voted for Obama by just over 5 points.  Had the election come out closer, Romney's path to victory would probably have involved also winning Wisconsin, Nevada, New Hampshire, or Iowa, as these states had comparable margins to Colorado, and had been battlegrounds during the election.
As many mathematical analysts have noted, however, the state voting in a fashion most similar to that of the nation as a whole is not necessarily the tipping-point.  For example, if a candidate wins only a few states but does so by a wide margin, while the other candidate's victories are much closer, the popular vote would likely favor the former.   However, although the vast majority of the states leaned to the latter candidate in comparison to the entire country, many of them would end up having voted for the loser in greater numbers than did the tipping-point state.  The presidential election in 2016 was a notable example, as it featured one of the largest historical disparities between the Electoral College and popular vote.   As the election was quite close, the winner of the Electoral College did not capture the popular vote. Additionally, this "split" was a lot larger in both directions than in previous, tighter elections, such as the one that took place in the year of 2000.  In that election, Vice President Al Gore won the popular vote by less than 1 percent, while incoming president George W. Bush won the Electoral College by only 4 votes.  In contrast, 2016 Democratic nominee Hillary Clinton won the popular vote by over 2 percentage points.   This meant that Donald Trump would have picked up New Hampshire, Nevada, and Minnesota if the popular vote had been tied, assuming a uniform shift among the battleground states.   On the other hand, Clinton would have had to win the popular vote by at least 3 points in order to win the Electoral College, as Trump, the Republican nominee, won the tipping-point state of Wisconsin by less than 1 percent. 
Swing states have generally changed over time. For instance, the swing states of Ohio, Connecticut, Indiana, New Jersey and New York were key to the outcome of the 1888 election.  Likewise, Illinois  and Texas were key to the outcome of the 1960 election, Florida and New Hampshire were key in deciding the 2000 election, and Ohio was important during the 2004 election. Ohio has gained its reputation as a regular swing state after 1980,   and last voted against the winner in 1960. If current trends from the 2012 and 2016 elections continue, the closest results in 2020 will occur in Arizona, Florida, Maine, Michigan, Minnesota, Nebraska's second congressional district, New Hampshire, North Carolina, Pennsylvania, and Wisconsin.  Other potential swing states include Colorado, Iowa, New Mexico, Nevada, Ohio, and Virginia, all of which have voted for both a Republican candidate and a Democratic candidate within a five-point margin multiple times in the previous six presidential elections. 
Professor Joel Bloom has mentioned opinion polls, previous election results, media attention, candidate campaign stops, and major advertising buys as crucial factors in identifying swing states. A 2004 article in the Oregon Daily Emerald also cites movie director Leighton Woodhouse opining that there is a general consensus among most groups regarding a majority of the states typically thought of as swing states.  Additionally, the swing-state "map" may transform dramatically between election cycles, especially depending on the candidates and their policies. In addition, gradual shifts can occur within states due to changes in demography, geography, or population patterns. For example, many West-Coast or currently Republican states, like Arkansas, Missouri, Tennessee, and West Virginia, had been battlegrounds as recently as 2004. 
A broad pundit consensus regarding the status of future battleground states developed in the years following the 2012 presidential election.  Contributors included Larry Sabato's Crystal Ball, Nate Silver of FiveThirtyEight, and other electoral analysts.  From the results of recent presidential elections, a general conclusion was reached that the Democratic and Republican parties start with a default electoral vote count of 188 each.  In this scenario, the twelve competitive states are Wisconsin, Pennsylvania, New Hampshire, Minnesota, Ohio, Iowa, Virginia, Florida, Michigan, Nevada, Colorado, and North Carolina.  However, this projection was not specific to any particular election cycle, and assumed similar levels of support for both parties. 
The electoral college encourages political campaigners to focus on these swing states while ignoring the rest of the country. States in which polling shows no clear favorite are usually inundated with campaign visits, television advertising, get-out-the-vote efforts by party organizers and debates, while "four out of five" voters in the national election are "absolutely ignored," according to one assessment. 
Since most states use a winner-takes-all arrangement, in which the candidate with the most votes in that state receives all of the state's electoral votes, there is a clear incentive to focus almost exclusively on only a few undecided states. In contrast, many states with large populations such as California, Texas, and New York, have in recent elections been considered "safe" for a particular party, and therefore not a priority for campaign visits and money. Meanwhile, twelve of the thirteen smallest states are thought of as safe for either party – only New Hampshire is regularly a swing state, according to critic George Edwards.  Additionally, campaigns stopped mounting nationwide electoral efforts in the last few months near/at the ends of the blowout 2008 election, but rather targeted only a handful of battlegrounds. 
Proponents of the Electoral College claim that adoption of a national popular vote would shift the disproportionate focus to large cities while ignoring rural areas.  Candidates might also be inclined to campaign hardest in their base areas to maximize turnout among core supporters. In some cases, they may also see greater opportunities in gaining votes from districts heavily supporting their opponent, the effects of which many feel would come at the expense of those in more closely divided parts of the country. Proponents of direct popular vote argue that the disproportionate influence the electoral college system affords swing states in determining the outcome of elections is unfair and essentially undemocratic.
- "Larry J. Sabato's Crystal Ball » The Electoral College: The Only Thing That Matters". www.centerforpolitics.org. Retrieved 2017-01-27.
- Beachler, Donald W.; Bergbower, Matthew L.; Cooper, Chris; Damore, David F.; Dooren, Bas Van; Foreman, Sean D.; Gill, Rebecca; Hendriks, Henriët; Hoffmann, Donna (2015-10-29). Schultz, David; Hecht, Stacey Hunter, eds. Presidential Swing States: Why Only Ten Matter. Lexington Books. ISBN 9780739195246.
- "The Odds Of An Electoral College-Popular Vote Split Are Increasing". FiveThirtyEight. 2016-11-01. Retrieved 2016-11-06.
- Silver, Nate (2012-04-27). "Arizona Is (Probably) Not a Swing State". The New York Times. Retrieved 2013-06-06.
- Silver, Nate (2012-11-08). "As Nation and Parties Change, Republicans Are at an Electoral College Disadvantage". Retrieved 2013-06-06.
- Silver, Nate (2016-09-20). "2016 Senate Forecast | FiveThirtyEight". FiveThirtyEight. Retrieved 2016-11-06.
- "Larry J. Sabato's Crystal Ball » SENATE 2016: FLIP FLOP". www.centerforpolitics.org. Retrieved 2017-01-27.
- "The Electoral College Blind Spot". FiveThirtyEight. 2017-01-23. Retrieved 2017-01-27.
- "Election Update: North Carolina Is Becoming A Problem For Trump". FiveThirtyEight. 2016-10-05. Retrieved 2017-01-27.
- "Larry J. Sabato's Crystal Ball". www.centerforpolitics.org. Retrieved 2017-01-27.
- "The Real Story Of 2016". FiveThirtyEight. 2017-01-19. Retrieved 2017-01-27.
- "The Odds Of An Electoral College-Popular Vote Split Are Increasing". FiveThirtyEight. 2016-11-01. Retrieved 2017-01-27.
- Chang, Alvin. "Trump will be the 4th president to win the Electoral College after getting fewer votes than his opponent". Vox. Retrieved 2017-01-27.
- "Clinton's popular vote lead surpasses 2 million". USA TODAY. Retrieved 2017-01-27.
- "Why FiveThirtyEight Gave Trump A Better Chance Than Almost Anyone Else". FiveThirtyEight. 2016-11-11. Retrieved 2017-01-27.
- "Clinton's Leading In Exactly The States She Needs To Win". FiveThirtyEight. 2016-09-22. Retrieved 2017-01-27.
- Malone, Clare (2016-07-18). "The End Of A Republican Party". FiveThirtyEight. Retrieved 2017-01-27.
- " 1888 Overview" p.4, HarpWeek.
- " Daley Remembered as Last of the Big-City Bosses", David Rosenbaum, New York Times, April 21, 2005.
- Trolling the Campuses for Swing-State Votes, Julie Salamon, " The New York Times", October 2, 2004
- Game Theory for Swingers, Jordan Ellenberg, " Slate.com", October 25, 2004
- Weaver, Dustin (2017-11-24). "How Dem insiders rank the 2020 contenders". TheHill. Retrieved 2018-01-13.
- Timeline - Margin of Victory View 270 To Win.
- " Portrait of a swing State", Cuniff, Meghann. Oregon Daily Emerald. October 4, 2004.
- "Battleground States Poll - June 21, 2004". Wall Street Journal. 2004-06-21. Retrieved 2017-07-05.
- "The 2016 Results We Can Already Predict". POLITICO Magazine. Retrieved 2016-06-11.
- Silver, Nate (2016-06-29). "2016 Election Forecast | FiveThirtyEight". Retrieved 2016-08-23.
- "2016 Presidential Election Interactive Map". 270toWin.com. Retrieved 2018-10-14.
- "Larry J. Sabato's Crystal Ball » The Electoral College: The Only Thing That Matters". www.centerforpolitics.org. Retrieved 2016-05-21.
- "Larry J. Sabato's Crystal Ball » The Electoral College: Pennsylvania Moves Toward Clinton". www.centerforpolitics.org. Retrieved 2015-09-30.
- Katrina vanden Heuvel (November 7, 2012).
"It's Time to End the Electoral College". The Nation. Retrieved November 8, 2012.
Electoral college defenders offer a range of arguments, from the openly anti-democratic (direct election equals mob rule), to the nostalgic (we’ve always done it this way), to the opportunistic (your little state will get ignored! More vote-counting means more controversies! The Electoral College protects hurricane victims!). But none of those arguments overcome this one: One person, one vote.
- Edwards III, George C. (2011). Why the Electoral College is Bad for America (Second ed.). New Haven and London: Yale University Press. pp. 1, 37, 61, 176–7, 193–4. ISBN 978-0-300-16649-1.
- Hands Off the Electoral College by Rep. Ron Paul, MD, December 28, 2004
- The Critical 2012 Swing States
- Battleground States 2008 via the Washington Post
- Swing State Ohio Documentary
- Swing State feature documentary project
- Guide to the 2004 swing states from Slate
- Battleground states from Democracy in Action site hosted by George Washington University
- How close were Presidential Elections? Influential States - Michael Sheppard
- The Bush campaign memo detailing its look at the swing states (PDF file)