Batterham and Hopkins have proposed a new approach for reporting the statistical findings from research studies. Their technique combines information on the magnitude of the estimate of the effect (e.g., mean difference), the degree of imprecision about that effect (e.g., the confidence interval), and the smallest difference that has real-world (or clinical) meaning. This information is combined into an overall set of likelihood statistics and a set of short descriptors (likely beneficial, etc.) are proposed. In this commentary, I address three issues: Is this approach better than using p-values? Is this approach useful with observational data? What are the drawbacks of this approach? I particularly comment on the usefulness of their approach for epidemiologists and other researchers who work with observational data.