presidential-elections-from-a-data-science-perspective | historical data on US presidential elections

presidential-elections-from-a-data-science-perspective is a Jupyter Notebook library. presidential-elections-from-a-data-science-perspective has no bugs, it has no vulnerabilities, it has a Strong Copyleft License and it has low support. You can download it from GitHub.

There is an abundance of political commentary regarding the dynamics of US presidential elections. In some cases the commentary includes sound data-driven analysis. Publicly accessible analysis tends to focus on the coming election with only passing references to previous election results. Indeed, each presidential election is unique and often include decisive idiosyncratic events such as October surprises. Nevertheless, there are overall known trends such as the difficulty of winning an election without reaching a majority in the state of Ohio or the fact that an incumbent president usually wins a second term. These seemingly constant trends suggest that there might be interesting insights hidden in the historical data on presidential elections. Here we focus on an analysis of the margins of victory in all the states and nationally. Some of the results below agree well with common expectations while others suggest interesting novel political scenarios.