The Bayesian paradigm is increasingly being used to overcome the limitations of frequentist approaches in statistical inference. We specified a weakly informative prior to compute the coefficient of the linear models using the Bayesian statistical framework for the African obtuse snakehead, Parachanna obscura. Fish samples were collected monthly for six months from artisanal fishermen from Eleiyele Reservoir and measured for Total Length (L) to the nearest cm and weighed (W, wet weight) to the nearest g. A fitted Bayesian linear model (estimated using MCMC sampling with four chains of 2000 iterations and a warmup of 1000) indicates positive allometric growth (median b = 3.20, 95% Credible Interval (CI) [2.93, 3.49]) with substantial explanatory power (R2 = 0.91, 95% CI [0.87, 0.94] based on the posterior predictive model. The Scale Reduction Factor was less than 1, and the effective sample size was larger than 1000 for all parameters: reliable evidence of convergence and typically a sufficiently large effective sample size. The African obstuse snakehead, Parachanna obscura thrived in Eleiyele Reservoir because they were plump. Bayesian analysis could be a reliable alternative to the classical ordinary least square methods for linear regression models in fisheries.