Many supply chain and finished goods distribution networks involve intercity freight transportation. Shipping customers secure transportation services by matching their requirements to available service in an effort to minimize their total logistics costs subject to service level constraints. Frequently, shippers’ modal decisions are constrained by short-term capacity constraints restricting one of the available options, or gaps in shipper knowledge or carrier marketing programs. As a result, the observed traffic flows may not reflect the potential demand for the mode. Because the potential demand for a mode is not directly measurable, when planning road and rail capacity, governments and railroads cannot make accurate capacity planning decisions based on current traffic flows. The model developed here identifies the potential demand for intercity full truckload and intermodal shipments over the most heavily utilized 75,000 shipment lanes in the western United States by estimating minimum total logistics costs by mode. These flows are compared with actual U.S. freight flows in order to determine the differences between observed flows and the model estimated potential demand. The results indicate potential demand for intermodal transportation is high; considerable freight volumes could be delivered with lower logistics cost by switching from truck to intermodal transportation. This evidence suggests that observed traffic flows and trends may not be a sound basis for planning freight transportation infrastructure in the United States.