A data-driven blog that uses chart, maps, cartograms, and other visualizations to understand important societal issues.


2016 Election in Maps (and Cartograms): Impact of Third Party Votes

Description: Map (top) and cartogram (bottom) of the potential impact of third party votes in the 2016 US Presidential election. For the cartogram, the area of each county is proportional to its population.
Each county is colored according to the potential impact of third party votes. Specifically, all third party votes are added to the losing Trump-Clinton candidate, and the color of the county is proportional to the remaining percentage gap. For example, a light-Red county (0–1% in the legend), indicates that (a) Trump originally  won the county and (b) adding all the third party votes to Clinton means that the final winning vote gap (%) was between 0 to 1%. Green indicates that adding all third party votes to a Trump county is enough to turn the county Blue. Contrariwise, Black indicates that adding all third party votes to a Clinton county is enough to turn the county Red. This is a metric to indicate how much of an impact third party candidates had in the 2016 elections.
Analysis: Generally, third party candidates had only a small impact on the 2016 US Presidential election:

  • No third party candidate won a single county.
  • A third party candidate finished second in only 23 counties.
    • This happened in only one Blue county— Kalawao County, HI—where Jill Stein finished second (with five votes). Kalawao’s population is ~90 and 20 people voted!
    • McMullin came in second in 22 counties, dominated by Utah (his home state) and Idaho.
  • All third party candidates grouped together were enough to come in second in 32 (of 3,142) counties.
    • Four Blue counties (DC, Kalawao in HI, and two Alaskan counties).
    • 28 Red counties, dominated by the “McMullin effect”, but also including Wyoming and Nebraska.

Although third party counties showed little promise in terms of winning a county, they had the potential to swing some counties. These “swing counties” (Green=swing from Red to Blue, Black=swing from Blue to Red) are largely centered around Utah (Evan McMullin), New Mexico (Gary Johnson), and Vermont (Bernie Sanders). However, the swing states of Wisconsin and Michigan also show counties with the third-party effect. The two “swing counties” with the largest populations are Maricopa (Arizona) and Riverside (California); see the cartogram.
Methodology: 2016 election data were collected manually from the Secretary of State website for each state. 2012 election data were sourced from GitHub1. Spatial data was visualized using ESRI ArcMap2. Cartograms were made using ScapeToad3. Data was processed using custom VBA scripts in Microsoft Excel4.

  1. https://github.com/tonmcg/County_Level_Election_Results_12-16
  2. http://www.esri.com/data/esri_data
  3. https://scapetoad.choros.ch/
  4. https://products.office.com/en-us/excel


Leave a Reply