Typically, mobile vehicles follow the same paths repeatedly, resulting in a common path along with some variance. These paths are often punctuated by branches into other paths based on decision-making in the area around the branch. The objective of this work is to apply a statistical methodology to determine decision-making regions for branching paths. An average path in a spatial s-coordinate frame is defined in the proposed algorithm, as well as boundaries representing percentiles in variances along the s-coordinate. The percentile boundaries along each branching path intersect near the decision point; these intersections in path variances are used to determine path branching locations. The resulting analysis provides decision points that are very robust to typical path conditions, such as two paths that may not clearly diverge at a specific location. Additionally, the methodology defines decision-making radii that encompass statistical percentile memberships of a location relative to the branching paths. To validate the proposed technique, an off-line implementation of the decision-making region algorithm is applied to previously classified wheelchair path subsets. Results show robust detection of decision regions that intuitively agree with user decision-making in real-world path following. A potential 73% reduction in user inputs is demonstrated to be feasible.