Storm surge barriers are crucial for the flood protection of the Netherlands and other Deltas. In order to guarantee a high safety level, a high closure reliability is required. In the Netherlands, the reliability of flood defenses is typically assessed based on extreme water level and wave height statistics. Yet, in case of operated flood defenses, such as storm surge barriers, the temporal clustering of successive events may be just as important. This study investigates the evolution and associated flood risk of clusters of successive storm surge peaks at the Maeslant Storm Surge Barrier in the Netherlands. Two mechanisms are considered that may lead to a series of storm surge peaks. Multi-peaked storm surges, as a consequence of tidal movement on top of the surge, are studied by means of simplified storm patterns that are widely used within the Netherlands. Clusters of surges resulting from different, but related storms are investigated by means of time series analysis of a long sea-level record close to the Maeslant Barrier. We conclude that the tendency of storms to cluster and especially the occurrence of multi-peaked storms may substantially increase the flood risk in the area behind the Maeslant Storm Surge Barrier and we envision that the influence is likely to increase with sea-level rise. The numbers are however highly uncertain due to the strong sensitivity to assumptions, model choices and applied data set. More insight in the statistics of the time evolution of extreme sea water levels is needed to better understand and ultimately to reduce these uncertainties.