Analyzing Swing States in Presidential Elections: The Case of Wisconsin
Michelle Zuo
TL;DR
The paper studies county-level voting patterns in Wisconsin across seven elections (2000–2024) to understand how urban mobilization and regional structure shape statewide outcomes. It combines absolute and percentage vote-change metrics with a Pearson-correlation–based County Similarity Matrix and choropleth visualizations to reveal spatial heterogeneity, including a pronounced southeastern urban cluster driving Democratic gains and a north–south divide in county trajectories. Key contributions include four derived datasets (Vote Change, County Similarity, Win Margin, Win Margin Heatmap) and multiple interactive maps that highlight geographic patterns and county-level similarities. The work provides a baseline for more predictive, district-level analyses and motivates extending the framework with methods such as Möbius inversion, optimal transport, temporal modeling, and demographic integration to better explain and forecast electoral dynamics in swing states like Wisconsin.
Abstract
This paper quantitatively analyzes county-level voting patterns in Wisconsin's presidential elections from 2000 to 2024. As a pivotal swing state, Wisconsin has alternated between Democratic and Republican candidates since 2012. Using data from the Wisconsin Elections Commission, we examine vote totals across 72 counties and seven election cycles. Pearson correlations measure similarity in county voting trajectories, while choropleth maps visualize spatial shifts. Results show strong clustering of vote changes: Democratic and Republican gains between 2016 and 2020 were concentrated in southeastern urban and suburban counties, with rural areas showing little change. Correlations reveal a north-south divide, as southern counties exhibit similar trends and northern ones diverge. These findings highlight spatial heterogeneity in electoral dynamics and the decisive role of urban mobilization in statewide outcomes.