Modern residential land development has trended toward densification, resulting in limited greenspace on individual parcels. As a result, land developers and homeowners are often constrained for space to plant shade trees. This can result in suboptimal placement of shade trees for the provision of energy conservation benefits. Using a simulation program called EnergyPlus, we examined the effects of existing trees on the energy consumption of recently constructed homes in three U.S. cities with distinctly different climates: Metro Minneapolis, MN, Charlotte, NC, and Metro Orlando, FL. Remote sensing was used to identify placement of existing trees around the homes, revealing that there were 1.5 to 2.9 trees within 15 m of the homes on average. These existing trees, when modeled as large-stature deciduous trees in the simulator, provided average annual energy conservation benefits per parcel of 14 kWh (MN), 25 kWh (NC), and 44 kWh (FL). We then developed an alternative tree placement strategy that spatially reconfigured the existing trees to both minimize space conflicts and maximize energy savings. This alternative strategy significantly improved annual energy savings per parcel to 57 kWh (MN), 47 kWh (NC), and 103 kWh (FL). Our alternative tree placement strategy was equally effective as the conventional strategy (always plant a shade tree on the west aspect) in MN and NC, but not in FL. However, the alternative strategy was more responsive to space constraints and therefore could be used to maximize energy savings while concurrently minimizing space conflicts between trees and site uses.