AbstractFishery collapses frequently result from combined pressures of the environment and man, which are difficult to discern because of the complexities involved and our limited knowledge. Models to resolve this complexity often become too sophisticated, with too many assumptions and, consequently, with little capacity to predict beyond calibration data. In this paper we implement a different procedure where the model is kept simple and uncertainty accounts for the equation imperfectness to reproduce ecological complexity. Human and environmental forcing on an anchovy (Engraulis encrasicolus) stock are simulated with only six parameters plus their error terms, and the uncertainty is computed with Bayesian methods. The simple structure is able to reproduce the major dynamical features of this species in the Gulf of Cádiz, including data on life stages and age structure that had no contact with the model. This is a distinct performance for a frugal approach working on a mid‐trophic species and a positive instance where parsimony can simulate the interaction of man, fish and the environment, provided uncertainty is accounted for in the process.