Abstract. Stratospheric extreme events during Northern winter and spring have been shown to sometimes enhance the sub-seasonal predictability of large-scale tropospheric circulation patterns such as the North Atlantic oscillation (NAO) and Greenland/European blocking. However, it remains unclear whether event-to-event differences in the observed tropospheric evolution after individual sudden stratospheric warmings (SSWs) represent a robust difference in the tropospheric response to the events, or whether such differences in tropospheric evolutions are simply caused by tropospheric variability. To make progress on this question, we robustly quantify the tropospheric response with an ensemble simulation strategy in a controlled model environment. We construct a model climatology using the ICON global numerical weather prediction (NWP) model, representing a wide range of realistic stratosphere–troposphere evolutions during winter months, but under controlled boundary conditions to exclude confounding factors like teleconnections of tropical origin. The simulations reproduce key aspects of observed stratosphere-troposphere coupling, providing a consistent framework to assess event-specific tropospheric responses. We produce spin-off ensembles for selected SSW events; the corresponding ensemble means help robustly quantify the tropospheric response to these SSWs. We find pronounced and robust event-to-event differences in the tropospheric response to SSWs. We further show that the flow anomalies in the lower stratosphere during the second week are well correlated with the surface response 3–7 weeks after the SSW. Moreover, our results indicate that the formation of wave reflection surfaces within the lower stratosphere may prevent the establishment of persistent lower-stratospheric anomalies. Overall, our controlled model simulations show that individual SSWs may differ significantly in their likelihood to induce a tropospheric response and that this likelihood is mainly determined by the post-SSW flow evolution within the stratosphere. These results may be relevant for sub-seasonal predictability of surface weather, especially given that the stratospheric part of the response to SSWs is highly predictable.