Not Available for Download

Predicting Litigation Outcomes

Posted: 11 Oct 2011  

Joanna Shepherd

Emory University School of Law

Date Written: August 2011


Economic theory asserts that, in general, the only cases going to trial should be cases with unpredictable outcomes. When the law applies to the facts to yield consistent and predictable outcomes, litigants have strong incentives to settle cases before trial. I test this theory using a dataset of decisions from the U.S. Courts of Appeals over a six-year period. My analysis includes over 70 variables such as the issues in the case, detailed information about the appellants and respondents, the legal history of the case, information about the litigants' attorneys, and extensive background information about the judges hearing the case. Employing a series of limited dependent variable models, I find that, despite the theoretical predictions, several case and judge variables are systematically related to case outcomes. Preliminary results from my study of decisions from U.S. District Courts find simlar relationships between case variables and case outcomes. The patterns revealed by the model have important implications for litigation strategy.

Keywords: Trial outcomes, litigation model, litigation strategy

JEL Classification: K0, K4, K41

Suggested Citation

Shepherd, Joanna, Predicting Litigation Outcomes (August 2011). Available at SSRN:

Joanna Shepherd (Contact Author)

Emory University School of Law ( email )

1301 Clifton Road
Atlanta, GA 30322
United States
404-727-8957 (Phone)

Paper statistics

Abstract Views