Many methods have been devised for setting baselines for greenhouse gas (GHG) mitigation projects. Of these, multiproject baseline methods have emerged that yield baseline emissions rates (or performance standards), which can be used for any number of projects in the corresponding sector and region. These methods frequently rely on deriving the performance standards from the outputs and associated emissions of recently completed facilities within an appropriate reference region. However, no standard or systematic method has been accepted for determining the appropriate spatial boundary and year of construction (or implementation) for facilities to be included in the reference set. This study examines the question of how GHG crediting programs may begin to methodically address these two parameters. We use examples from power sector and land-use change and forestry projects to illustrate Our approach, but the analysis is applicable to a wide variety of GHG mitigation projects. We find that performance standards can generally be grouped into five spatial categories: global, national (or other administrative region), infrastructural, biophysical, and generic. As for temporal periods, multiyear averages produce more consistent performance standards that are often more representative of the mix of resources or technologies that can be used for a given activity than a standard based on only 1 year. Published by Elsevier Ltd.