Rust Belt Correlation Cluster (Blue Wall)

Overview

This pillar analyzes the correlated voting behavior of Pennsylvania, Michigan, and Wisconsin. It provides insights into how shifts in one Rust Belt state can predict outcomes in the others, offering a powerful regional view.

What It Does

The model calculates a real-time correlation score based on historical election results, demographic similarities, and aggregated polling data across the three key states. It identifies regional trends that individual state polls might miss. This allows for the creation of conditional probabilities, modeling how a swing in one state impacts the entire cluster.

Why It Matters

Presidential elections are often won or lost in this specific cluster of states. Understanding their interconnectedness provides a significant predictive edge over analyzing them in isolation, capturing the shared political and economic sentiment that drives regional voting blocs.

How It Works

First, the pillar aggregates historical presidential election results since 1992 for PA, MI, and WI. It then layers in current polling averages and key demographic data, such as union membership and manufacturing employment. Using this data, it calculates a weighted correlation coefficient that models how the states move together, updating as new polls are released.

Methodology

The core of the analysis is a Pearson correlation coefficient calculated on state-level presidential election results and polling averages over a rolling 90-day window. The model incorporates a custom demographic similarity index, which is a weighted score based on census data for non-college educated populations, manufacturing employment rates, and union household percentages. New polls are weighted more heavily using an exponential decay formula.

Edge & Advantage

This pillar provides a macro signal on the election's tipping point region, offering an edge by identifying momentum that is not yet fully priced into individual state markets.

Key Indicators

• Inter-State Polling Correlation

  high

  Measures the statistical relationship between polling shifts in PA, MI, and WI.

• Union Household Trends

  medium

  Tracks polling and economic sentiment within union households, a key shared demographic.

• Demographic Similarity Index

  medium

  A custom score measuring the overlap in key voter demographics across the three states.

Data Sources

• FiveThirtyEight / RealClearPolitics

  Provides aggregated state-level polling data for correlation analysis.

• U.S. Census Bureau

  Supplies demographic data on education, employment, and population for the similarity index.

• Federal Election Commission (FEC)

  Offers official historical election results for baseline correlation.

Example Questions This Pillar Answers

• → Will the Democratic candidate win Pennsylvania in the 2028 Presidential Election?
• → Which party will win at least two of Michigan, Pennsylvania, and Wisconsin?
• → What will be the Democratic margin of victory in Wisconsin if they win Michigan by 3% or more?

Tags