Historically, whether for training or analysis, there has been a tension between the desirable detail of entity simulations and the reduced overhead of aggregate simulations. Often both are used – entity to provide detail in areas of particular interest and aggregate to provide an appropriately complex battlespace. When so used, care must be taken to control the interactions between entity and aggregate simulations due to inaccurate or even impossible interactions. For example: entities often cannot sense aggregates while aggregates often use completely different engagement models (fire/detonate vs Lanchester's equations). In this paper we present an innovative approach to aggregate simulation based on validated entity performance data. Our approach provides low overhead computation and operation while allowing entity level mobility, sensing and engagement compatible with more detailed entity simulations. We also present results of using this methodology during test and user events.