A quick map showing early votes and mail-in ballots by voting precinct. Data from Rhode Island Secretary of State as of 10/31/2020 using registered voter data from 2018. 2 quick tables highlighting the top & bottom 10 precincts, as well as the ability to hover over the map for more detail.

A quick horizontal bar graph showing the method early voters choose to use between early voting at a local city/town hall, or voting by mail. The white line down the middle shows the 50% mark to easily identify which method a majority of early voters used.

A quick map showing early votes and mail-in ballots cast data from Rhode Island Secretary of State as of 10/25/2020 as a percent of turnout using registered voter data from the 2018 general election from the Board of Elections.

As a map won't work for this years primary, or rather, would look weird as some of the cities/towns didn't have an election this years primary. I chose to use a scatter plot to visualize the change in voter turnout between the 2016 primary election and this years 2020 primary. A teal dot, represents a precinct that has seen an increased turnout from 2016 and a purple dot a decrease in turnout.

Previously, my use of election data was limited to viewing turnout and showing votes for each candidate showing which recieved more votes. This year I used Python & Pandas to find some interesting facts hidden within the primary election data. Things like, which endorsed candidate lost their election, or if mail-in votes effect the outcome from the ballot box on election day.

With the COVID-19 pandemic mail-in voting has seen a huge bump as people are less likely to go to the polls. This post will take A look into the scale of mail-in voting for the 2020 primary compared to previous elections.

Last year I attempted to make a live election graph in Tableau. However, Tableau Public only allows data to automatically update once every 24 hours. Clearly... that didn't work out well, so this year I decided to try again and create my first Python web app using Flask, Bootstrap and AWS.

Using previous election results data from Rhode Island's Board of Elections since 2010, I normalized the data and removed multi-vote races using Python to find the 5 closest single-vote elections for every primary/general each year. I found it a bit surprising how many elections were decided by fewer then 10 votes between the top 2 candidates.