We present a dynamic model of venture investment with endogenous fund heterogeneity and deal flows that produces performance persistence without innate skill differences. Investors work with multiple funds and use tiered contracts to manage dynamic moral hazard. Recently successful funds receive continuation contracts that encourage greater innovation, and subsequently finance innovative projects through assortative matching. Initial luck, therefore, exerts an enduring impact on performance by altering funds' future investment opportunities. The model generates implications broadly consistent with empirical findings, such as that persistently outperforming funds encourage greater innovation and attract better entrepreneurs even with worse terms. The model further predicts ``incumbent bias'' in investing in funds, compensation improvement for recently successful managers, backloading across contracts, and amplification of skill differences.