Fundraising and vote distribution: a non-equilibrium statistical approach

TL;DR

This paper introduces a statistical framework using Shannon entropy and Superstatistics to relate campaign spending to vote distribution, aiding in outcome prediction and misconduct detection in elections.

Contribution

It develops a novel non-equilibrium statistical approach linking campaign finance data to voting outcomes, enabling efficient misconduct detection.

Findings

The model accurately predicts vote distributions from campaign spending data.

Application to real election data identifies only nine candidates for potential misconduct.

Framework enhances election integrity analysis with minimal data verification.

Abstract

The number of votes correlates strongly with the money spent in a campaign, but the relation between the two is not straightforward. Among other factors, the output of a ballot depends on the number of candidates, voters, and available resources. Here, we develop a conceptual framework based on Shannon entropy maximization and Superstatistics to establish a relation between the distributions of money spent by candidates and their votes. By establishing such a relation, we provide a tool to predict the outcome of a ballot and to alert for possible misconduct either in the report of fundraising and spending of campaigns or on vote counting. As an example, we consider real data from a proportional election with $6323$ candidates, where a detailed data verification is virtually impossible, and show that the number of potential misconducting candidates to audit can be reduced to only nine.