An age-structured production model (ASPM) is used to assess the Carpenter resource off the south western coast of South Africa. The model is fitted to standardized CPUE and length frequency linefish data. Problems encountered when attempting to fit the model to these data are explored. Reasons may include conflicting data, or complexities in the resource dynamics that are not incorporated in the simple density-dependent ASPM dynamics. Plausible parameter estimates are only achieved when fixing or imposing penalty functions for key parameters. However, this leads to deterioration in the fit to the data with systematic trends in the residuals which render results suspect and management advice based on such assessments dubious. Indeed, a management procedure approach may be better suited to circumstances when a “best assessment” is problematic, as seems to be the case here.