Validity and within-device reliability of the Metric application for assessing barbell velocity and displacement

This study evaluated the validity and within-device reliability of the Metric smartphone application compared to a linear position transducer (LPT). Twenty-five resistance-trained participants (19 male, 6 female) completed incremental loading one-repetition maximum (1RM) assessments in the back squat, bench press, and deadlift. Mean concentric velocity (MCV) and displacement (CD) were measured simultaneously by two Metric devices and one GymAware LPT. The Metric application detected 94.3% of repetitions (F1-score: 0.971) compared to GymAware's 99.5% (F1-score: 0.997). Within-device reliability of Metric varied by exercise: highest in squat (MCV: ICC = 0.994, CV = 3.83%, SDC = 0.05 m/s [4.27% of 1RM]; CD: ICC = 0.991, CV = 1.53%, SDC = 0.02 m), then bench press (MCV: ICC = 0.981, CV = 7.57%, SDC = 0.08 m/s [6.38% of 1RM]; CD: ICC = 0.974, CV = 3.53%, SDC = 0.02 m), and lowest in deadlift (MCV: ICC = 0.941, CV = 13.29%, SDC = 0.13 m/s [14.05% of 1RM]; CD: ICC = 0.753, CV = 27.16%, SDC = 0.10 m). However, Metric did not meet our predetermined reliability criteria for velocity trackers (ICC ≥ 0.997, CV ≤ 3.5%). Substantial-to-perfect agreement between Metric and GymAware was found for squat MCV (CCC = 0.981–0.993), but only poor-to-moderate agreement for bench press and deadlift MCV (CCC = 0.866–0.946), and poor agreement across all exercises for CD (CCC < 0.768). While Metric offers a cost-effective and user-friendly alternative to LPTs, it cannot currently be recommended for research or practice where high precision is required.