Abstract: Sudden Stratospheric Warmings (SSW) are critical processes in stratosphere-troposphere coupling. Their signals can propagate downward to the troposphere, influencing weather and climate and serving as an important source of predictability for subseasonal to seasonal (S2S) prediction. Based on hindcasts (2008—2022) and real-time prediction data (2023 to present) from the China Meteorological Administration Climate Prediction System version 3 (CMA-CPSv3), this study systematically evaluates the prediction skill of CMA-CPSv3 for SSW events. A total of 11 SSW events occurred in the Northern Hemisphere during 2008—2024. Predictability varies among events, with an average predictable lead time of approximately 3—4 pentads. Composite analysis reveals that CMA-CPSv3 can capture the evolution characteristics of temperature and circulation fields associated with SSW about 3 pentads in advance. However, when the lead time exceeds 3 pentads, the intensity of the predicted anomalies weakens significantly. Further analysis shows that the prediction errors in the zonal wind circulation associated with SSW events are closely related to prediction errors in the intensity of planetary wave activity propagating upward from the troposphere to the stratosphere prior to SSW onset. When the lead time exceeds 3 pentads, the system tends to underestimate the intensity of planetary wave activity before SSW onset, thereby limiting its prediction skill for SSW events.