Crop yield and revenue insurance products with coverage based on actual production history (APH) yields dominate the U.S. Federal Crop Insurance Program. The APH yield, which plays a critical role in determining the coverage offered to producers, is based on a small sample of historical yields for the insured unit. The properties of this yield measure are critical in determining the value of the insurance to producers. Sampling error in APH yields has the potential to lead to over-insurance in some years and under-insurance in other years. Premiums, which are in part determined by the ratio of the APH yield to the county reference yield, are also affected by variations in APH yields. Congress has enacted two measures, yield substitution and yield floors, that are intended to limit the degree to which sampling error can reduce the insurance guarantee and producer welfare. We examine the impact of sampling error and related policy provisions for Texas cotton, Kansas wheat, and Illinois corn. The analysis is conducted using county level yield data from the National Agricultural Statistics Service and individual insured-unit-level yield data obtained from the Risk Management Agency’s insurance database. Our findings indicate that sampling error in APH yields has the potential to reduce producer welfare and that the magnitude of this effect differs substantially across crops. The yield substitution and yield floor provisions reduce the negative impact of sampling error but also bias guarantees upward, leading to increased government cost of the insurance programs.