Incorporating an instability predictor into a portable sensor has a number of clinically relevant applications. This study investigated the feasibility of developing a real-time assessment tool to predict stepping during standing by monitoring Center of Pressure (COP) measurements. Forward and backward perturbations were performed on 16 able-bodied subjects using a pulley system attached to the subjects' waist. A linear relationship was found between the peak COP velocity (COPv) and the peak COP position caused by the perturbations. As the peak COPv occurs considerably before the peak COP, the peak COP estimated using a regression equation from the peak COPv may serve as an instability predictor. By comparing stepping thresholds with the estimated peak COP, we found that the stepping predictor successfully predicted instability (stepping) earlier than those predictors using actual COP. Results show that the proposed model is a viable solution to predict stepping, and the feasibility of incorporating the model into a neuroprosthesis system for standing.