This study proposes a methodology to determine road vulnerability to flood events, which combines link criticality in a road network with flood probability. Link criticality is evaluated based on the link role in maintaining network connectivity and enabling trips between different areas of a territory. While previous research on road network vulnerability typically requires simulating a disruption and evaluating the expected effects, the proposed methodology aims to identify the most critical links based on flow-related and connectivity characteristics of the network. The methodology integrates transport demand, network supply, and flood probability data at a macroscopic scale. Because of iterative disruption simulations are not required, this approach is more practical and feasible for real-world applications compared to existing methods. An application has been made to the Emilia-Romagna region (Italy), frequently affected by floods. The results highlight the importance of some road segments, particularly in peripheral areas with limited infrastructures, and show a positive relation between link criticality and the potential path travel time increase due to a disruption.