Abstract

PURPOSE: To assess the validity of a submaximal protocol designed to predict the one repetition maximum (1-RM) in the free-weight back squat. Undertaking a 1-RM carries a risk of injury due to the maximal nature of the assessment, however, if a protocol and equipment can be used to accurately predict an individual’s strength through a submaximal assessment then this will negate the need to take athletes to maximal levels.

METHODS: Fifteen resistance trained individuals (mean ± SD, stature: 178.0 ± 6 cm, body mass: 85.0 ± 11.3 kg 1-RM: 148.5 kg), were recruited. Participants completed two visits, the first a habituation of the protocol and the second, testing of a submaximal predication of 1-RM using the Flex laser optic device (www.flexstronger.com). The Flex software sets a series of progressive incremental loads based on a participant’s estimated 1-RM. At each load, the device captures concentric movement displacement and time, enabling the calculation of mean concentric velocity (MCV). Linear regression is then applied to the captured data between absolute load and attained MCV, facilitating the prediction of 1-RM. Following the Flex prediction, participants were blinded to the result and a standardised 1-RM protocol was followed, loading the bar in increments until the participant could no longer complete an unaided repetition. For all repetitions, participants were instructed to maintain eccentric control before completing the concentric phase as explosively as possible. Least products regression (LPR) were used to assess linearity and proportional bias, enabling quantification of validity between the two measures. Systematic and random error of the Flex were assessed by quantifying the 95% limits of agreement (LOA) between predicted and actual 1-RM data.

RESULTS: Back squat mean and standard deviation for the predictive and actual 1-RM was 166.9 ± 52.1 kg and 1-RM was 148.5 ±46.1 kg, respectively. LOA from -21 to 52 kg indicate that on average the Flex submaximal protocol has a systematic error over predicting an individual's 1-RM by 15.5 kg, with a random error of 36.8 kg. Whilst the LPR between the predicted and actual measures resulted in an R2 ≥ 0.89 demonstrating a degree of similarity in the two values, the intercept, y = 16.01+(0.81* x)demonstrates that there appears to be proportional bias i.e. as strength increases the error also increases.

CONCLUSION: The preliminary findings suggest that whilst there is a degree of correlation between the Flex predicted value and actual 1-RM, it does not provide a sufficiently valid measure of 1-RM strength for the free-weight back squat in trained individuals, generally over predicting 1-RM values. In addition, the random bias of the device increases in proportion to maximal strength. The sample size of 15 participants in the current study was small and therefore there is a need for further research using a larger participant pool in addition to the examination of the validity for the deadlift and bench-press lifts before full conclusions are drawn.

PRACTICAL APPLICATION: At present, practitioners wanting to examine maximal strength levels for the free-weight back squat should continue to use conventional 1-RM protocols.