Most studies on dynamic ridesharing, which allows any driver to ferry separate passengers with overlapping routes in a single trip, have focused on developing efficient algorithms. While dynamic ridesharing demands extensive computing infrastructure, this paper suggests a simple taxi ridesharing approach that allows only passengers from certain taxi hotspots to share a ride with another who has a similar destination and measures the effects, spatiotemporal patterns and its benefits. Compared to dynamic ridesharing, the proposed method does not require dedicated driver fleets, networked communication system, or monopoly of all passenger information. We identify taxi pickup hotspots and analyze the spatiotemporal patterns of the simple ridesharing approach. The results show that 48 % of rides from hotspots could be shared, reducing the overall vehicle-km traveled by 1.2 km for each shared ride. We also find that spatiotemporal patterns of the ridesharing could represent urban char-acteristics. For example, places with high ridesharing potential and low saved-trip distances could imply low public transportation accessibility while areas with high shareability during working hours on both weekdays and weekends could represent public transportation hubs. The proposed method is expected to be useful to identify taxi stands that have high ridesharing opportunities. Policy makers can use our approach to support simple ridesharing scheme. Moreover, with the characterized taxi stands, such as longer saved-trip distance and nightlife peaks, the proposed method could be used as decision support tools for temporary allowed ridesharing. In addition, spatiotemporal patterns of the taxi stands would be used in designing public transportation systems.