Rain-induced landslides are one of the most recurrent geohazards around the world, posing great threats to the safety and property of people living in mountainous areas. One of the most effective strategies for rain-induced landslide risk management and reduction is to analyze landslide susceptibility during rain events. To characterize the impact of varying rainfall conditions on landslide occurrence, a maximum rolling rainfall index (MRRI) is proposed in this study for the spatiotemporal landslide susceptibility modeling. During rainfall, MRRI can be updated based on the real-time maximum rolling 2-, 4-, 6-, 8-, 12-, 18-, and 24-h rainfall data. Furthermore, a spatiotemporal landslide susceptibility modeling approach, in which the MRRI index is taken as a conditioning factor, is developed. To illustrate the effectiveness and versatility of the proposed approach, landslide susceptibility models based on the random forest technique are trained in the central area of Hong Kong. The resulting landslide susceptibility models are then applied to two historical rainstorms, and the application results show that derived time-series landslide susceptibility maps are in good agreement with the spatial distribution of real landslides.