This paper presents the evaluation of three different methods for determining zone temperature setpoint variations that limit peak electrical demand in buildings. The methods were developed in a companion paper [Lee K-H, Braun JE. Development of methods for determining demand-limiting setpoint trajectories in commercial buildings using short-term measurements. Building and Environment 2007, in press.] and are evaluated in the current paper through simulation for a small, medium, and large commercial building. Inverse models were employed for the simulation where the parameters were estimated with nonlinear regression techniques using hourly data. Two of the demand-limiting methods are based on the use of simple building models that capture dynamics of the building cooling loads in response to setpoint variations over a short time scale. The third method is data driven and only relies on load data to directly determine setpoint variations that minimize peak cooling demand. All three demand-limiting methods work well in terms of peak demand reduction for individual buildings. However, the data-driven method has slightly better performance than the other methods, is easier to implement, and is directly applicable for peak load reduction of aggregated buildings.