1. "The NBI Study compared a set of new buildings to a set of old buildings...CBECS includes buildings that were built as early as 1920. The LEED sample consists exclusively of buildings built or renovated after 2000.
2. "Certification does not require actual energy use data"

Anyone who's poked around in CBECS knows that EUI for pre-1959 buildings is less than for newer buildings (http://www.eia.gov/emeu/cbecs/cbecs2003/detailed_tables_2003/2003set14/2...), so comparing LEED buildings to very old buildings would more likely diminish the amount of savings rather than over-report, as is being implied here. (Note this is for the average of buildings of all sizes, but there is variance when you start looking at building size).

And really, the plaintiffs should find some way to distinguish their BD&C gripes from all of LEED, as EBOM is chock full of empirical data.