This paper reports on the first length based stock assessment for the Okavango delta fishery. This study was conducted against the background of friction between the different stakeholders of the Okavango Delta fishery, where the main contention was allegations of fish over-exploitation through commercial gill netting. The basic approach involved estimating the vital population parameters (growth and mortality) of selected fish stocks, running a length-based cohort analysis on the selected stocks, and then predicting their long term yield under different fishing scenarios (varying levels of effort) using the Thompson and Bell model. Fishermen were used in this study to collect some of the data, adopting a similar approach to what was done in the assessment of the Bangweulu (Zambia) fishery. The estimated mean exploitation rates from the length cohort analysis (LCA) for all the species is below 0.5, which suggests that the stocks are not biologically over-exploited. The Thompson and Bell yield prediction model indicated that the MSY level for the three-spot and red-breast Tilapia and Nembwe can be reached when the present effort is increased by a factor of 1.8, 2.1 and 3.5 respectively. MSY for the other species could not be identified within the range of possible effort levels. This also suggests that the fish stocks of the Okavango are not over-exploited. The Okavango delta fishery has the lowest yield and effort when compared to other African lakes and river systems using Bayley's logistic equation, which supports the findings that this fishery can be expanded without fear of biological over-exploitation.