Abstract Mid-water pair trawling (PTM) targeting small pelagic resources represents a key fishing activity in the Adriatic Sea. This fishery is experiencing a long period of crisis due to resource depletion and the lack of appropriate market strategies, and vessels spend most of the time searching for fishing schools. The searching strategy largely depends on the interaction between vessels: the captains of the {PTM} units take their decision also checking the position and the fishing status of other vessels. Understanding this strategy represents a key step towards a more effective resource management, since strategies directly determine the pattern of fishing effort. A Conditional Logit model has been devised to analyze fishermen's strategy as a non-cooperative game. This category of games is characterized by the existence of (at least) one equilibrium point - a Nash Equilibrium - in which each player plays his strategy, that is a Best Response to the strategies of the other players. This equilibrium point was estimated for the different scenarios defined by environmental (sea surface temperature and atmospheric pressure) and economic (fuel and fish prices at market) variables. Vessel Monitoring System data were used to capture fleet activity, while different datasets were collected to reconstruct environmental and economic drivers. Results indicate a good predictive power of the model, and suggest that the equilibrium strategy that guides units' behaviour is invariant with respect to environmental conditions, whereas it is largely influenced by economic factors. These latter, via strategies, may determine important consequences on the resources in terms of exploited areas and the impact of fishing activity. In particular, a low fuel price when fish price is high leads to higher values of CPUE, and then to a more efficient but also impacting fishing activity.