Empirical regularities regarding start-ups that do not show prolonged longevity have been known for some time. In this chapter, the regularity is argued to originate from an interaction between cognition-laden characteristics of entrepreneurs and contextual conditions. Overconfidence is argued to cause miscalibration of objective probabilities of success, which in turn causes excess entry, but also to negatively affect survival rates. Moreover, overconfidence is argued to be an evolutionary mechanism that helps explain the distribution of entrepreneurs at the local level. It does so by advocating overconfident entrepreneurs to be more likely to beacon personal, but miscalibrated, beliefs to others and thereby set off spillover effects. The bias is therefore argued to be detrimental to actors at an individual level (as it negatively affects survival rates), but favorable at the system level (as it facilitates spillover effects). The concluding discussion of these matters is extended by a discussion of policy issues.