TL;DR: Maps represent land area, not population density. Therefore the amount of red vs. blue on a Presidential election result map is misleading unless the map accounts for population density in some way.

Over the past year, I have engaged in debate with several Trump supporters who, when confronted with the fact that Donald Trump did not win the popular vote, and in fact received 3 million fewer votes than Hilary Clinton, immediately point to a map like this one, claiming that the immense amount of red on the map proves how popular Trump is.

There’s only one problem with this argument:
Land doesn’t vote, people vote.

Maps represent land area, not population density, so it wouldn’t matter if there was only one tiny blue dot on the map. What matters is how many people live within that area and how they voted.

A fairer representation of the voting distribution would be something like this map, where the state’s size in the map is shown relative to the number of votes cast.

notice the big circle between Virginia and Maryland. That’s Washington D.C. which, despite being a tiny speck on the first map, has a population larger than Alaska.

However, even this map is deceptive since:

a) The red is mostly concentrated in the center, making it look larger than it actually is.

b) As a color, red tends to be more visually prominent than blue, also making it look larger than the blue.

c) This map doesn’t show how close the races are in each state, only showing the “winner take all” electoral college tally. So, Trump might have only carried a state by a few votes, while Clinton carried her’s by thousands, and this map wouldn’t indicate that.

One way to account for these issues is to use shades of red and blue to indicate the strength of the support in a given area, as with this map.

However, the best way to present the popular vote, is an old-fashioned bar or pie graph, where you can clearly see the actual amount of votes each candidate received.

This chart also shows the number of people ineligible to vote and who simply didn’t vote, which gives us an even more honest—and depressing—view of U.S. Voting in 2016.

Unfortunately, with a straight graph like a bar or pie, we lose all geographic perspective to help us understand were people where voting for a particular candidate. To overcome that, the map shows the county-by-county vote, but with population indicated using height. Although there is still a lot of red, it is at least in some part set off by the fact that the red counties are rarely very tall.

The important thing to remember here isn’t that maps are best suited to show location, not population, so arguing popularity based on the amount of land area the people who voted for Trump happen to occupy makes about as much sense as electing a realty TV star to the most important office in a democracy: none at all.