Leadership emergence (Taggar, Hackew, & Saha, 1999) is a dynamic process shaped by multiple factors, including performance expectations (Kalish & Luria, 2016) and individuals’ social capital (Maak, 2007). This study pits the predictions of the Expectations States Theory (Correll & Ridgeway, 2003) against those of the Social Capital Theory (Adler & Kwon, 2002). Using data from 164 individuals (14 teams) undergoing a two-day assessment boot camp in an elite military unit, we explore the complex interplay between performance, friendship networks, and leadership emergence, measured three times. Analysis using Stochastic Actor Oriented models for the co-evolution of networks and behavior (Snijders, Steglich, & Schweinberger, 2007) reveals that Expectations States Theory has a stronger hold, demonstrating that high-performing individuals are more likely to emerge as leaders and that leadership further enhances their performance. Implications for leadership theory, modeling of social networks, and leadership emergence in high-stress, short-duration teams are discussed.