We used data from a cooperative angler tagging program to assess the potential benefit of leaving hooks in fish captured and released by anglers. We assembled 248,010 records of angler catches of 27 species of Australian fish. Hooks were left in only 1.1% of released fish, and the overall recapture rate was 8.8%. We used relative risk, that is, the probability of an event (recapture) in a treatment group (those with hooks not removed) divided by the probability of that event in a control group (those with the hook removed), to assess the potential effects of leaving hooks in released fish. Relative risk ranged from 0.30 to 7.6 and differed from 1.0 in only 3 of the 27 species studied. Thus, there was little evidence that hook removal affected recapture probability. Similarly, there was no evidence of any difference in relative risk among fish species grouped by habitat type. Pooling results across all species and habitats yielded an overall relative risk of 1.18 (95% confidence interval, 1.02-1.36), which suggests that the recapture rate of fish from which hooks were not removed before release was 18% greater than that for fish from which the hooks were removed. Overall, our results indicate that not removing hooks benefits recapture rate, which can be considered as a surrogate measure of the survival of released fish.