We review an emerging technique for ionospheric forecast whereby empirical models and data assimilation techniques are combined and show that the blended technique produces a diminished result compared to using the empirical models alone. We then present a technique for the nowcast/forecast of ionospheric potential patterns using an analog forecast technique. This is a new type of model for describing ionospheric potential patterns based on matching solar wind conditions and available ground observations to previously archived assimilative mapping of ionospheric electrodynamics (AMIE) runs. The model combines the best features of a data assimilation--based model with the availability and forecast potential of an empirically based model. We define a set of comparison metrics, show quantitatively how this model compares with the Weimer (1996) model and detail the performance for a May 1998 event.