This thesis investigates the stochastic dynamics of popularity on Bluesky, a decentralized online social network, to challenge the idea that follower count defines influence. By analyzing the ecosystem, we quantify the continuous activity required to balance attention decay. We developed a custom crawler to reconstruct full user timelines from the public stream, compiling a longitudinal dataset of over 31 million interactions. We analyzed these dynamics using a Stochastic Mean-Field framework, which models popularity not as a static metric, but as a diffusive process governed by the opposing forces of content amplification and temporal decay. In this study, we leverage a recently developed stochastic model to characterize the platform's governing dynamics.