The effects of barotrauma on the short-term mortality and physical condition of Australian bass Macquaria novemaculeata were investigated after being: (1) conventionally angled and released during two tournaments in deep impoundments and (2) released untreated or subjected to venting or recompression during a manipulative experiment. All fish were released into two 20 m deep bathy-cages and monitored for 3 days. Of 238 M. novemaculeata angled during the tournaments, 43 (18·1%) had clinical signs of barotrauma or were vented and five of these later died (11·6% mortality). Catch histories varied significantly between both barotrauma and non-barotrauma fish and tournaments, but only hook ingestion significantly influenced mortality (P < 0·05). During the manipulative experiment, venting significantly influenced mortality (13·3%) compared to no treatment or recompression (no deaths). Magnetic resonance images and dissections of barotrauma fish indicated large variation among clinical signs. On the basis of these results, wherever possible M. novemaculeata suffering barotrauma should be immediately released with no treatment. Fish that are unable to resubmerge should be recompressed, while those held in live wells and released in shallow water should be vented.