2016/12/31 by Daniel Katz, Daniel Martin Katz, Michael James Bommarito +2 · 452 citations
Mathematics · Physics and Astronomy · Social Sciences · #Artificial Intelligence in Law #Artificial intelligence #Classifier (UML) #Computer science #Context (archaeology) #Econometrics #Economic Justice #Geography #Judicial and Constitutional Studies #Law #Legal Education and Practice Innovations #Machine learning #Mathematics #Parametric statistics #Political science #Random forest #Sample (material) #Statistics #Supreme court #physics.soc-ph
paper · pdf · open access · doi:10.1371/journal.pone.0174698
published in PLoS ONE 12(4), e0174698 (Public Library of Science) · version 2.02; 18 pages, 5 figures. This paper is related to but distinct from arXiv:1407.6333, and the results herein supersede arXiv:1407.6333. Source code available at https://github.com/mjbommar/scotus-predict-v2
arxiv created 2017/01/17 · openalex publication_date 2017/04/12 · arxiv updated 2017/07/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/08
Building on developments in machine learning and prior work in the science of judicial prediction, we construct a model designed to predict the behavior of the Supreme Court of the United States in a generalized, out-of-sample context. To do so, we develop a time-evolving random forest classifier that leverages unique feature engineering to predict more than 240,000 justice votes and 28,000 cases outcomes over nearly two centuries (1816-2015). Using only data available prior to decision, our model outperforms null (baseline) models at both the justice and case level under both parametric and non-parametric tests. Over nearly two centuries, we achieve 70.2% accuracy at the case outcome level and 71.9% at the justice vote level. More recently, over the past century, we outperform an in-sample optimized null model by nearly 5%. Our performance is consistent with, and improves on the general level of prediction demonstrated by prior work; however, our model is distinctive because it can be applied out-of-sample to the entire past and future of the Court, not a single term. Our results represent an important advance for the science of quantitative legal prediction and portend a range of other potential applications.