Executive Summary

Election integrity is fundamental to democratic governance. As electoral systems grow increasingly complex and data-dependent, the need for sophisticated analytical tools has never been greater. Swarmalytics specializes in developing and implementing advanced data analytics methodologies that enhance transparency, detect anomalies, and support evidence-based decision-making in election administration.

This whitepaper outlines six comprehensive analytical frameworks developed and deployed during the 2022 and 2024 North Carolina election cycles. These methodologies represent a non-partisan, objective approach to election data analysis, designed to identify irregularities, improve operational efficiency, and strengthen public confidence in electoral processes.

Our analytical suite encompasses voter registration integrity monitoring, absentee ballot tracking, fraud detection algorithms, and predictive modeling. Each methodology has been refined through real-world application and demonstrates the critical role that data science plays in modern election administration.

Introduction

In an era of heightened scrutiny of electoral systems, election officials and legislators face unprecedented demands for transparency and accountability. Traditional oversight methods, while valuable, often lack the granularity and real-time capabilities necessary to address modern challenges in election administration.

Swarmalytics bridges this gap through sophisticated data analytics that transform raw voter data into actionable intelligence. Our methodologies are grounded in three core principles:

• Non-partisan objectivity: Our analyses focus exclusively on data patterns and statistical anomalies, without regard to political outcomes

• Transparency: Methodologies are documented, replicable, and designed to enhance public understanding of electoral processes

• Actionable insights: Analyses are structured to support concrete decision-making by election officials and policymakers

The methodologies described in this whitepaper were developed and deployed during North Carolina's 2022 and 2024 election cycles, analyzing millions of voter records and ballot transactions to identify patterns, detect anomalies, and provide election administrators with enhanced situational awareness.

Voter Registration Integrity: Gain/Loss Analysis

Overview

The Gain/Loss Report represents a fundamental data integrity tool for monitoring voter registration databases. This methodology tracks changes to the voter roster on a week-over-week basis, identifying additions and removals at both the aggregate county level and individual voter record level.

Methodology

Using North Carolina's publicly available statewide voter file (ncvoter_Statewide.txt), updated weekly, our system performs differential analysis to identify all North Carolina Voter Identification Numbers (NCID) that appear or disappear between successive file releases. The analysis maintains two key components:

• Aggregate reporting: County-level summaries of net gains and losses, enabling jurisdictional comparisons and trend analysis

• Individual record tracking: A comprehensive database of all added or removed records with timestamps, preserving audit trails

Applications

This methodology serves two critical functions. First, it detects technical errors in voter file generation that could lead to eligible voters being erroneously excluded from the active roster. Such errors, while rare, can have significant consequences if undetected. Second, it identifies unusual patterns in voter registration activity that may warrant investigation, such as unexpected surges or declines in new registrations within specific jurisdictions.

Value Proposition

For election officials, the Gain/Loss Report provides continuous quality assurance for voter registration systems. For legislators, it offers transparency into the dynamics of voter registration and early warning of anomalies that may indicate systemic issues requiring policy attention.

Sequential Integrity Monitoring: Skipped NCID Analysis

Overview

Voter identification numbers are typically assigned in strict sequential order. Deviations from this pattern—instances where NCID are assigned out of sequence—can indicate various scenarios ranging from benign administrative practices to potential data integrity concerns. The Skipped NCID Report systematically identifies and tracks these deviations.

Methodology

Our system analyzes the creation sequence of NCID within the voter file, flagging any identifiers that are "skipped" in the sequential ordering. For example, if NCID AA123 is created immediately after AA121, the system flags AA122 as skipped. These skipped identifiers are catalogued and monitored across subsequent file releases to determine if they eventually appear in either the voter registration file or the voter history file (ncvhis_Statewide.txt).

Applications

Sequential integrity monitoring serves as a quality control mechanism for voter registration systems. While most skipped identifiers have benign explanations—such as cancelled registrations or administrative corrections—systematic patterns of skipped NCID can reveal inefficiencies in registration processes or, in rare cases, more serious irregularities. This methodology provides election administrators with granular visibility into the voter identification assignment process.

Value Proposition

This analysis enhances confidence in the integrity of voter identification systems by ensuring that all assigned identifiers are accounted for and that assignment processes follow expected patterns. It provides a low-level technical audit capability that complements higher-level registration monitoring.

Absentee Ballot Intelligence: Dashboard and Predictive Systems

Overview

Absentee voting has grown substantially in recent election cycles, creating new challenges for election planning and resource allocation. Our Absentee Dashboard provides real-time intelligence on absentee ballot requests and returns, enabling data-driven decision-making throughout the absentee voting period.

Dashboard Methodology

The dashboard aggregates absentee ballot data by day and county, presenting temporal and geographic patterns through interactive visualizations. These visualizations allow election officials to monitor absentee ballot activity in real-time, comparing current cycles against historical baselines and identifying jurisdictions that may require additional resources or attention.

Address Verification Component

As a complementary analysis, our system cross-references address information between the voter registration file and absentee ballot request records. This verification process identifies discrepancies that could lead to ballot delivery issues or indicate data quality problems. In our deployments, this analysis found no significant discrepancies, validating the integrity of the address matching processes.

Predictive Modeling

Building on historical absentee voting patterns, demographic data, and population trends, we developed a predictive model that forecasts expected absentee ballot volumes by county prior to the opening of the absentee registration period. This model establishes baseline expectations, allowing election officials to identify counties with unusually high or low absentee ballot activity based on statistical deviations from predicted values.

Applications

For election administration, these tools enable proactive resource planning and early detection of potential issues. For policymakers, they provide evidence-based insights into absentee voting trends and help assess the adequacy of existing infrastructure to meet voter demand.

Fraud Detection Analytics

Overview

While voter fraud is statistically rare, robust detection mechanisms are essential for maintaining public confidence in electoral integrity. Swarmalytics developed two fraud detection methodologies that leverage data integration and pattern analysis to identify potential irregularities.

Deceased Voter Analysis

This analysis cross-references North Carolina voter history records from the 2022 elections against a proprietary database of deceased Americans. The objective is to identify instances where votes may have been cast fraudulently in the names of deceased individuals. Our analysis methodology:

• Matches voter records against death records using multiple identifiers to ensure accuracy

• Accounts for timing discrepancies between death dates and election dates

• Flags potential cases for further investigation while acknowledging alternative explanations

Findings from the 2022 analysis identified a small number of potential fraud cases. However, the scale of the phenomenon appeared to be relatively insignificant, consistent with academic research on voter fraud prevalence.

Multiple-State Voting Analysis

This analysis integrates voter participation data from multiple states—specifically North Carolina and Georgia—to detect instances where individuals may have cast ballots in both jurisdictions during the same election. The methodology employs sophisticated matching algorithms that account for name variations, address histories, and other identifying information to minimize false positives.

Our analysis identified a small number of cases where individuals appeared to have participated in both North Carolina and Georgia elections. These findings were flagged for investigation, recognizing that such patterns may reflect data quality issues, legitimate explanations, or actual fraud requiring enforcement action.

Value Proposition

These fraud detection methodologies provide election officials and law enforcement with targeted intelligence that focuses investigative resources on the most credible potential cases. For policymakers, they offer empirical data on the prevalence and nature of potential electoral fraud, supporting evidence-based policy decisions regarding election security measures.

Voter Behavior Analysis: Party Affiliation Dynamics

Overview

Understanding patterns in voter party affiliation provides valuable insights into electoral dynamics and can reveal strategic voting behaviors. Our Party Change Analysis tracks week-to-week shifts in party registration, distinguishing between newly registered voters and previously registered voters who change party affiliation.

Methodology

The analysis segments party affiliation changes into two categories:

• New registrants: Initial party selections by voters registering for the first time

• Party switchers: Previously registered voters who change their party affiliation

This segmentation allows for nuanced analysis of registration trends versus strategic party switching, particularly relevant in states with closed or semi-closed primary systems where party affiliation determines primary voting eligibility.

Research Application

This methodology was deployed to test a specific hypothesis: that voters from competing political parties were strategically switching party affiliations to participate in primary elections. Our analysis of 2022 and 2024 data did not reveal evidence of large-scale strategic party switching for primary participation. This finding provides empirical data relevant to ongoing policy debates about primary election systems.

Value Proposition

For legislators considering primary election reforms, this analysis offers objective data on voter behavior patterns. For political strategists and academics, it provides granular insights into party affiliation dynamics that inform understanding of electoral competition.

Conclusion

The methodologies detailed in this whitepaper represent a comprehensive approach to election data analysis that enhances transparency, improves operational efficiency, and strengthens public confidence in electoral systems. Swarmalytics' work demonstrates that sophisticated data analytics can provide election officials and policymakers with the tools necessary to address modern challenges in election administration.

Our analytical framework is distinguished by its commitment to non-partisan objectivity. Rather than advocating for specific policy positions or political outcomes, we focus on developing rigorous methodologies that surface relevant data patterns and enable evidence-based decision-making. This approach has proven valuable across multiple election cycles and diverse analytical challenges.

As electoral systems continue to evolve and face new challenges—from increasing absentee ballot usage to concerns about data security—the need for advanced analytical capabilities will only grow. Swarmalytics is committed to developing and refining methodologies that meet these emerging needs while maintaining the highest standards of analytical rigor and objectivity.

Looking Forward

The methodologies described in this whitepaper are continuously refined based on real-world deployment experience and evolving best practices in data science. Future development priorities include enhanced real-time monitoring capabilities, expanded cross-state data integration, and machine learning applications for anomaly detection.

We welcome the opportunity to discuss how these analytical capabilities can support your jurisdiction's election integrity objectives and policy development needs.