Accuracy, Optimality, and the Preponderance Standard

Law, Probability & Risk, Forthcoming

U of Alabama Legal Studies Research Paper No. 2547348

26 Pages Posted: 10 Jan 2015 Last revised: 12 Mar 2015

See all articles by Edward K. Cheng

Edward K. Cheng

Vanderbilt Law School

Michael S. Pardo

University of Alabama School of Law

Date Written: January 7, 2015


Law-and-economics scholars have recently argued that the legal system should set burdens of proof on the basis of ex ante welfare considerations. In this Article, we reject this welfarist approach, showing that it relies on contested normative principles, raises legitimacy concerns, and is nearly impossible to implement in practice. As an alternative, we propose a decision-theory model that we constrain to account for core legal values and the practical limitations of the trial process. Specifically, we require that the burden of proof prioritize accuracy (truth) over welfare, and that it be capable of operating without knowledge of the base rates or prior probabilities of activities. The resulting optimization problem can be solved using a minimax approach, which minimizes the maximum probability of error faced by each of the parties, and remarkably, the minimax solution turns out to be precisely the preponderance-of-the-evidence standard currently imposed by courts. We thereby not only refute recent welfare theories about the burden of proof, but also provide a new theoretical justification for the traditional preponderance standard.

Keywords: burden of proof, burden of persuasion, evidence, accuracy, welfare, minimax, error

JEL Classification: C44

Suggested Citation

Cheng, Edward K. and Pardo, Michael S., Accuracy, Optimality, and the Preponderance Standard (January 7, 2015). Law, Probability & Risk, Forthcoming; U of Alabama Legal Studies Research Paper No. 2547348. Available at SSRN:

Edward K. Cheng (Contact Author)

Vanderbilt Law School ( email )

131 21st Avenue South
Nashville, TN 37203-1181
United States
615-875-7630 (Phone)

Michael S. Pardo

University of Alabama School of Law ( email )

P.O. Box 870382
Tuscaloosa, AL 35487
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
PlumX Metrics