President-elect Joe Biden and the real power of data visualization

We have all anxiously followed the dramatic head-to-head between Trump and Biden, played on the thread of votes and poisons. Of all the infographics I went through in the newspapers and online, I was struck by the States’ choropleth maps that show counties colored in blue or red depending on the party that collected the most votes. See, for example, the maps on the New York Times, or on Repubblica.it.

Let’s take Michigan, for example. Almost all counties are colored red, except for a few blue spots. But look at the data table; it’s Biden who won there! At a glance, you would say there is a mistake. But no: the blue spots are on the cities with the highest population density, such as Detroit and Ann Arbor. In the few highly populated cities, people vote Dem, while people in the countryside and small towns vote Republican. It came out as a recurring pattern: in fact, the same thing happened in other States such as Nevada and Pennsylvania — and, by the way, the are similar electoral dynamics between left- and right-wing parties in Italy too, nowadays.

Misleading infographics? At first glance, one would say yes. But no: in their binary either-red-or-blue logic, these maps communicate this particular distribution of votes in a much more powerful way. Indeed, their apparent imprecision is a warning to us all. We must be careful to interpret the response from the polls, regardless of who was elected President. These maps tell us that the USA is a society split in two, and numbers go beyond geographical borders: there are no longer “red” or “blue” States, rather only “centers” and “peripheries.”

Donald Trump knows the power of color in choropleth maps very well. His controversial “Impeach this” 2019 campaign featured an animated map showing that the USA is, well, almost 100% red. Fact-checking articles such as this one from CNN demonstrated that Trump’s map was inaccurate, but the power and clarity of his message were undeniable.

Back to the 2020 elections, Bloomberg CityLab’s aptly-titled article How to Spot Misleading Election Maps by Laura Bliss and Marie Patino digs deeper into the issue. I also suggest reading the review of the best electoral maps published by AnyChart.

I hate the phrase “a picture is worth a thousand words.” But it is undeniable how, in this case, these almost entirely red maps can tell what is happening in the USA in a much more objective, effective, and easy to memorize way than any written commentary could.