Elections are a well-documented example of collective human decision-making, with voting data available for elections globally over several decades. SFI Complexity Postdoctoral Fellow Aanjaneya Kumar, along with former colleagues at the Indian Institute of Science Education and Research (IISER) Pune, analyzed election data from 34 countries and devised a mathematical model to study if a universal behavior emerges in elections. In a study published in Physical Review Letters, they show that the margin of victory for any election can be predicted solely by the voter turnout.
Given a distribution of voter turnout across different constituencies, the model calculates how tight or lopsided an election will be. The relationship between voter turnout and electoral margins was independent of the number of voters, applying to elections ranging from municipal to national. The model’s predictions matched with empirical data from hundreds of diverse elections across seven decades and 32 countries. The researchers suggest that this universality could serve as a statistical tool to detect electoral malpractices.
Read the paper “Universal Statistics of Competition in Democratic Elections” in Physical Review Letters (January 9, 2025). DOI: 10.1103/PhysRevLett.134.017401
Read More
- January 10, 2025: "Voter turnout drives margins of victory ― if elections are fair," Nature
- January 11, 2025: "IISER Pune study shows how voter turnout can predict winning margins," The Indian Express
- January 12, 2025: "IISER’s global study unveils universal poll pattern, may aid in fraud detection," The Times of India