Story sifting (or story recognition) is the sifting of interesting stories that emerge from the interactions between characters in a simulated storyworld. Recent works in this area have demonstrated the ability to automatically sift stories within milliseconds by plotting generated events as visual story arcs and using object detection techniques to sift out the event sequences that match a criteria of interest. Those techniques are primarily applied after the interactions have occurred and stories unfolded. The focus of our work is to use automated story sifting techniques to detect partially complete story arcs, in order to prospectively guide and control virtual characters in the storyworld on their future interactions. We do this by developing a modified version of the story sifting tool ARC DETECTOR to sift out partial plot trajectories a virtual character is on. We show that our modified sifting tool is able to detect story arcs with 50% and 70% completions at 86% and 93% precision, respectively.