Over the past decade, acoustic telemetry has become commonplace in studies on fish movement and behaviour. Over small spatial scales, arrays of acoustic receivers can be used to estimate movement paths in 2- or 3-dimensions with high temporal resolutions. Despite the growing prevalence of acoustic telemetry arrays, guidelines on how to generate robust position estimates—and further utilize this data in animal movement models such as hidden Markov models or step selection functions—are sparse. As animal movement models generally require either true positions or accurately specified spatial error distributions, understanding positioning error is crucial for behavioural inference. Here, current methods of telemetry positioning are reviewed. Simulated case studies are used to highlight the effect of state space model parameter selection on positioning accuracy, and in turn, the fitting of animal movement models.