On-site stormwater detention (OSD) is a conventional component of urban drainage systems, designed with the intention of mitigating the increase to peak discharge of stormwater runoff that inevitably results from urbanization. In Australia, singular temporal patts for design storms have goved the inputs of hydrograph generation and in t the design process of OSD for the last three decades. This paper raises the conc that many existing OSD systems designed using the singular temporal patt for design storms may not be achieving their stated objectives when they are assessed against a variety of altative temporal patts. The performance of twenty real OSD systems was investigated using two methods:(1) ensembles of design temporal patts prescribed in the latest version of Australian Rainfall and Runoff, and (2) real recorded rainfall data taken from pluviograph stations modeled with continuous simulation. It is shown conclusively that the use of singular temporal patts is ineffective in providing assurance that an OSD will mitigate the increase to peak discharge for all possible storm events. Ensemble analysis is shown to provide improved results. However, it also falls short of providing any guarantee in the face of naturally occurring rainfall.