In this paper, we argue that the temporal dynamic patterns of entrepreneurial activities affect the likelihood of new venture creation. We put forward theoretical model based on complexity science, and test these hypotheses using longitudinal data from Chinese Panel Study of Entrepreneurial Dynamics. This paper reaches the following findings: ① Rate and new ventures' initial performance shows an inverse U-shaped curvilinear relationship; ② A lower concentration is associated with a higher likelihood of new ventures' initial performance. This paper concludes with some suggestions for the role of temporal dynamic patterns of entrepreneurial activities in new venture creation. A complexity science approach holds valuable potential for entrepreneurship scholars as it provides a useful way to frame and explain new firm formation. We expect for furthering insights into the process of new venture creation.