Duran v. U.S. Bank National Assn.

(2014) 59 Cal.4th 1

Statistical sampling must satisfy due process. The sample must be representative, the methodology sound, and the defendant must retain the right to challenge individual claims. The constitutional floor for every extrapolation argument.

Holding

Trial by statistical sampling must satisfy due process. The sample must be representative, the methodology sound, and the defendant must have the opportunity to challenge individual claims. The trial court's use of a small, unrepresentative sample to extrapolate liability across the class was reversible error. Statistical evidence can be used to establish classwide liability patterns, but not to eliminate the defendant's right to raise individualized defenses.

Impact on Defense Practice

The constitutional floor for sampling methodology in both class actions and PAGA cases. Duran means that a plaintiff cannot extrapolate a violation rate from a handful of employees to the entire aggrieved population without a defensible statistical framework. The decision gave defense counsel the tools to challenge sample design, sample size, selection methodology, and extrapolation methodology — each of which can be outcome-determinative.

Defense Strategy

Challenge every element of plaintiff's sampling methodology: Is the sample truly random? Is the sampling frame complete? Does the sample size support the claimed confidence interval? Does the methodology allow for individualized defenses? In one case, the plaintiff's expert drew only from currently employed workers, excluding departed employees whose records showed different violation patterns. This selection bias was outcome-determinative. The 174-question expert deposition framework was designed specifically to expose these methodological vulnerabilities systematically.

This analysis is for informational purposes only. Case law is current as of Q1 2026.