The 538 members of the U.S. electoral college are meeting Monday. Despite some public pressure on electors to reject Trump, they're expected to overrule the results of the popular vote — something that has only happened five times in our nation's history. Hillary Clinton won the popular vote 65.8 million to 63 million, but she earned only 232 electoral votes to Donald Trump’s 306.
The electoral college is a source of serious debate — whether it should exist at all, and how we should draw the boundaries that decide how individual votes are pooled together, and therefore who wins. It might seem strange to leave so much up to the vagaries of cartographers and gerrymanders: By moving the boundary of a state or a district by a few miles, you can massively change election results.
Following the election, an artist and urban planner named Neil Freeman created a fascinating tool he dubbed “Random States of America.” The map randomly generated state boundaries and showed which candidate would win based on the population of those new areas.
At the request of Josh Wallaert, senior editor of Places Journal, Freeman then built on the idea to create a new series of five U.S. maps, organized around different systems than our currents states and districts. Part land-use planning and part science fiction, these fascinating maps show how reworkings of U.S. cartography would have resulted in different election outcomes.
- If our cellphone conversations decided the election
The first is based on a 2011 map that researchers at the MIT Senseable City Lab created based on cellphone data. They examined the parts of the country in which people tend to call each other more, and used the data to divide the map into “telecom communities,” discrete units joined by social connections.
With borders redrawn this way, into 28 states and one federal district, Hillary Clinton wins the electoral college 270 to 224, the researchers say.
- If the economy decided the election
The second map is based on one we recently featured on Wonkblog, in which Garrett Nelson and Alasdair Rae used census data on commute paths to separate the U.S. into “mega-regions” — economic units organized around cities and satellite communities, and determined by where Americans live, work and spend their free time.
In this scenario, Trump wins the electoral college 294 to 258, and even takes over part of California. But Hillary turns part of Texas blue, and carves off Miami and Cleveland.
- If ease of governance decided the election
The third map is based on what Wallaert calls a “classic” — a map designed in 1973 by geography professor George Etzel Pearcy for easier government administration. Pearcy looked at U.S. population density, transportation routes and geography to create a nation of 28 states.
Here, Clinton wins a narrower victory, taking the electoral college 271 to 241.
- If equally populated states decided the election
The fourth map is based on a design from Freeman himself. In 2012, he redrew the U.S. map so that each of the 50 states were equally populated and arrayed around the country’s 50 largest cities.
As Freeman and Wallaert note, even in this scenario, the electoral college outcome is different than the popular vote. Like Mitt Romney before him, Trump wins this map, 280 to 220.
- If states centered on cities decided the election
The final map is a new creation by Freeman. Here, the U.S. is split into 48 states centered around the country’s 48 largest urban areas. Freeman uses a technique known as a Voronoi partition, dividing the country so that each point in the state is closer to its “state capital” than any other capital.
The outcome is a near tie. But because the vote of the District of Columbia is allowed to count in this scenario, Clinton squeaks by 270 to 268.
Note: A previous version of this post cited older figures for the results of the popular vote. The figure has been updated to reflect more recent results.
More from Wonkblog: