The article by Will G. Hopkins discusses the importance of reliability in sports medicine and science, focusing on the reproducibility of measurements across repeated trials on the same individuals. Key measures of reliability include within-subject random variation, systematic change in the mean, and retest correlation. The typical error of measurement, expressed as the standard deviation of repeated measurements, is a simple and adaptable measure of within-subject variation. The coefficient of variation, which expresses the typical error as a percentage of the mean, is often more useful for many measurements in sports medicine and science. Limits of agreement, which represent the 95% likely range of change between two trials, are another measure of within-subject variation but are biased by sample size and heterogeneity. Systematic changes in the mean, such as learning, motivation, or fatigue, need to be eliminated from estimates of within-subject variation. Retest correlation, while useful, is sensitive to participant heterogeneity and may not accurately reflect the reliability of a test or equipment. The article also discusses the practical applications of reliability, including decision-making, comparison of tests or equipment, estimation of sample size in experiments, and estimation of individual differences in response to treatments. It emphasizes the importance of high reliability in experimental research, as it directly affects the required sample size. The author provides formulas for estimating sample size based on typical error or retest correlation and highlights the need for adequate precision in reliability studies to ensure reliable estimates of outcomes. Finally, the article covers the design and analysis of reliability studies, emphasizing the importance of sufficient sample size and trials to achieve adequate precision. It also discusses the analysis of data, including the detection and handling of heteroscedasticity, and the use of transformations to improve the reliability estimates.The article by Will G. Hopkins discusses the importance of reliability in sports medicine and science, focusing on the reproducibility of measurements across repeated trials on the same individuals. Key measures of reliability include within-subject random variation, systematic change in the mean, and retest correlation. The typical error of measurement, expressed as the standard deviation of repeated measurements, is a simple and adaptable measure of within-subject variation. The coefficient of variation, which expresses the typical error as a percentage of the mean, is often more useful for many measurements in sports medicine and science. Limits of agreement, which represent the 95% likely range of change between two trials, are another measure of within-subject variation but are biased by sample size and heterogeneity. Systematic changes in the mean, such as learning, motivation, or fatigue, need to be eliminated from estimates of within-subject variation. Retest correlation, while useful, is sensitive to participant heterogeneity and may not accurately reflect the reliability of a test or equipment. The article also discusses the practical applications of reliability, including decision-making, comparison of tests or equipment, estimation of sample size in experiments, and estimation of individual differences in response to treatments. It emphasizes the importance of high reliability in experimental research, as it directly affects the required sample size. The author provides formulas for estimating sample size based on typical error or retest correlation and highlights the need for adequate precision in reliability studies to ensure reliable estimates of outcomes. Finally, the article covers the design and analysis of reliability studies, emphasizing the importance of sufficient sample size and trials to achieve adequate precision. It also discusses the analysis of data, including the detection and handling of heteroscedasticity, and the use of transformations to improve the reliability estimates.