We see it every election: a map of the United States, each state color coded red or blue according to whether voters swung Republican or Democrat.
But as we all know, infographics can lie: a 51/49 split between Republicans and Democrats is so neck-and-neck that it seems ridiculous to color the whole state red. It’s just not an accurate representation of how people are voting.
So designers try to split the difference, mixing the red voters and blue voters by state until they have a purple blob that they feel better represents the true political temperature of a region.
But there’s a problem with purple in election maps. Our brains can’t actually tell how blue or red a shade of purple is by itself. And our perception is skewed by surrounding colors. A purple blob surrounded by red looks like a “redder” purple to us than that same color purple surrounded by blue. So our brains see a district that is going Republican, when it actually might be going Democrat, or vice versa.
According to Lawrence Weru, who studied election maps during the 2012 presidential elections while at Florida State University, this is why purple election maps are misleading. But there’s a solution. By adding an appropriate amount of green to each blob, we can neutralize the color purple from our election maps. Districts will appear redder or bluer according to how many Republicans or Democrats are voting in them, but neutral states will be white or gray.
There’s a big difference, and as Weru notes, this technique allows our eyes to easily identify a 51/49 voting split in a way that a purple blob never could. Read more about how Weru color-balances election maps on here.