Understanding variability in fish production, biomass, production/biomass (P/B) ratios, and their relationship to exploitation is central to fisheries sustainability. At Escanaba Lake, Wisconsin, USA, data from a compulsory creel census (1965?2009) were combined with survey data on fish populations to test for empirical relationships between annual production and exploitation rates of walleye (Sander vitreus). Empirical estimates of walleye production were relatively high and temporally variable in Escanaba Lake. Annual production, biomass, and P/B ratios ranges were 2.4?11.3 kg·ha?1·year?1, 9.1?49.4 kg·ha?1, and 0.15?0.30 year?1, respectively. Walleye production rates were significantly and positively correlated with adult (i.e., age 3+) walleye density ? the core metric used to manage walleye populations in the Ceded Territory of Wisconsin. However, adult walleye density correlated best with ages 4?6 production. Annual exploitation based on the percentage of individuals removed from the population was linearly related to annual exploitation based on the percentage of production removed. On average, 100% of annual production was harvested at ?20% annual exploitation of individuals; thus, higher sustained exploitation rates would deplete walleye biomass in Escanaba Lake over time. A sustainable annual exploitation rate of 20% annually is lower than the 35% limit reference point currently used for walleye management in the Ceded Territory of Wisconsin. Nevertheless, our estimate is strongly concordant with two recent and independent modeling efforts addressing this same topic. We suggest empirical estimates of fish production are a vastly underutilized tool for better understanding sustainable exploitation rates in walleye populations and, likely, fisheries in general.