In addition to its empirical results, this paper makes three distinct contributions to the literature on valuing fishing recreation by stated preference methods. In particular, the paper proposes a new approach to designing a choice experiment, called 'approximate minimax design'. It employs the latest advances in optimal design theory and is much easier to compute than the ordinary minimax design. Second, the choice-experiments are embedded in a trip frequency model, such that the changes in trip frequency can be predicted from improved fish management. Most importantly, the approach proposed in the paper provides a first step in developing dynamic fish management models, because the policies analysed may change the age-structure of the stock. The paper also contributes with new results on anglers' valuations of angling site characteristics. They conform to traditionally obtained results but also show some unexpected results. In particular, the result does not indicate a difference in the valuation of the species caught. As expected, the application shows that the marginal value of trophy-fish is high.