# Justice blocks and predictability of US Supreme Court votes

**Authors:** Roger Guimera, Marta Sales-Pardo  
**arXiv:** [1210.4768](https://arxiv.org/abs/1210.4768)  
**Date:** 2012-10-17  
**Summary page:** https://www.summarizepaper.com/en/paper/1210.4768v1/

## Key points

- The study "Justice Blocks and Predictability of US Supreme Court Votes" explores the predictability of US Supreme Court justices' votes based on their colleagues' votes in the same case.
- Models and methods developed for analyzing complex social networks are used to find that these methods are more accurate at predicting justice's votes than forecasts made by legal experts or algorithms that consider the content of cases.
- High predictability is a quantitative proxy for stable justice (and case) blocks, which likely reflect stable a priori attitudes toward the law.
- U.S. Supreme Court justice votes are more predictable than expected from an ideal court composed of perfectly independent justices, with deviations from ideal behavior most apparent in divided 5-4 decisions where justice blocks appear to be most stable.
- Justice predictability decreased during the 50-year period between the Warren Court and Rehnquist Court eras, and aggregate court predictability has been significantly lower during Democratic presidencies.
- Methods developed for analyzing complex social networks can be used to quantitatively investigate historical questions related to political decision-making, shedding light on how legal decisions are made and evolving behaviors within the judiciary.

## Summary

In their study "Justice Blocks and Predictability of US Supreme Court Votes," Roger Guimera and Marta Sales-Pardo explore the predictability of US Supreme Court justices' votes based on the votes of their colleagues in the same case. Using models and methods developed for analyzing complex social networks, the authors find that these methods are more accurate at predicting justice's votes than forecasts made by legal experts or algorithms that consider the content of cases. The authors argue that high predictability is a quantitative proxy for stable justice (and case) blocks, which likely reflect stable a priori attitudes toward the law. They discover that U.S. Supreme Court justice votes are more predictable than expected from an ideal court composed of perfectly independent justices, with deviations from ideal behavior most apparent in divided 5-4 decisions where justice blocks appear to be most stable. Additionally, they find evidence that justice predictability decreased during the 50-year period between the Warren Court and Rehnquist Court eras and aggregate court predictability has been significantly lower during Democratic presidencies. Overall, this study demonstrates how methods developed for analyzing complex social networks can be used to quantitatively investigate historical questions related to political decision-making, shedding light on how legal decisions are made and evolving behaviors within the judiciary.