Agility is a significant determinant of success in soccer; however, studies have rarely presented and evaluated soccer-specific tests of reactive agility (S_RAG) and non-reactive agility (change of direction speed – S_CODS) or their applicability in this sport. The aim of this study was to define the reliability and validity of newly developed tests of the S_RAG and S_CODS to discriminate between the performance levels of junior soccer players. The study consisted of 20 players who were involved at the highest national competitive rank (all males; age: 17.0 ± 0.9 years), divided into three playing positions (defenders, midfielders, and forwards) and two performance levels (U17 and U19). Variables included body mass (BM), body height, body fat percentage, 20-m sprint, squat jump, countermovement jump, reactive-strength-index, unilateral jump, 1RM-back-squat, S_CODS, and three protocols of S_RAG. The reliabilities of the S_RAG and S_CODS were appropriate to high (ICC: 0.70 to 0.92), with the strongest reliability evidenced for the S_CODS. The S_CODS and S_RAG shared 25–40% of the common variance. Playing positions significantly differed in BM (large effect-size differences [ES]; midfielders were lightest) and 1RM-back-squat (large ES; lowest results in midfielders). The performance levels significantly differed in age and experience in soccer; U19 achieved better results in the S_CODS (t-test: 3.61, p < 0.05, large ES) and two S_RAG protocols (t-test: 2.14 and 2.41, p < 0.05, moderate ES). Newly developed tests of soccer-specific agility are applicable to differentiate U17 and U19 players. Coaches who work with young soccer athletes should be informed that the development of soccer-specific CODS and RAG in this age is mostly dependent on training of the specific motor proficiency.

Purpose: To compare the effects of velocity-based training (VBT) and 1-repetition-maximum (1RM) percentage-based training (PBT) on changes in strength, loaded countermovement jump (CMJ), and sprint performance. Methods: A total of 24 resistance-trained males performed 6 weeks of full-depth free-weight back squats 3 times per week in a daily undulating format, with groups matched for sets and repetitions. The PBT group lifted with fixed relative loads varying from 59% to 85% of preintervention 1RM. The VBT group aimed for a sessional target velocity that was prescribed from pretraining individualized load–velocity profiles. Thus, real-time velocity feedback dictated the VBT set-by-set training load adjustments. Pretraining and posttraining assessments included the 1RM, peak velocity for CMJ at 30%1RM (PV-CMJ), 20-m sprint (including 5 and 10 m), and 505 change-of-direction test (COD). Results: The VBT group maintained faster (effect size [ES] = 1.25) training repetitions with less perceived difficulty (ES = 0.72) compared with the PBT group. The VBT group had likely to very likely improvements in the COD (ES = −1.20 to −1.27), 5-m sprint (ES = −1.17), 10-m sprint (ES = −0.93), 1RM (ES = 0.89), and PV-CMJ (ES = 0.79). The PBT group had almost certain improvements in the 1RM (ES = 1.41) and possibly beneficial improvements in the COD (ES = −0.86). Very likely favorable between-groups effects were observed for VBT compared to PBT in the PV-CMJ (ES = 1.81), 5-m sprint (ES = 1.35), and 20-m sprint (ES = 1.27); likely favorable between-groups effects were observed in the 10-m sprint (ES = 1.24) and nondominant-leg COD (ES = 0.96), whereas the dominant-leg COD (ES = 0.67) was possibly favorable. PBT had small (ES = 0.57), but unclear differences for 1RM improvement compared to VBT. Conclusions: Both training methods improved 1RM and COD times, but PBT may be slightly favorable for stronger individuals focusing on maximal strength, whereas VBT was more beneficial for PV-CMJ, sprint, and COD improvements.

No abstract

Jukic, I, García-Ramos, A, Malecek, J, Omcirk, D, and Tufano, JJ. Validity of load–velocity relationship to predict 1 repetition maximum during deadlifts performed with and without lifting straps: The accuracy of six prediction models. J Strength Cond Res 36(4): 902–910, 2022—This study aimed to compare the accuracy of six 1 repetition maximum (1RM) prediction models during deadlifts performed with (DLw) and without (DLn) lifting straps. In a counterbalanced order, 18 resistance-trained men performed 2 sessions that consisted of an incremental loading test (20-40-60-80-90% of 1RM) followed by 1RM attempts during the DLn (1RM = 162.0 ± 26.9 kg) and DLw (1RM = 179.0 ± 29.9 kg). Predicted 1RMs were calculated by entering both group and individualized mean concentric velocity of the 1RM (V1RM) into an individualized linear and polynomial regression equations, which were derived from the load–velocity relationship of 5 ([20-40-60-80-90% of 1RM], i.e., multiple-point method) or 2 ([40 and 90% of 1RM] i.e., 2-point method) incremental warm-up sets. The predicted 1RMs were deemed highly valid if the following criteria were met: trivial to small effect size, practically perfect r, and low absolute errors (<5 kg). The main findings revealed that although prediction models were more accurate during the DLn than DLw, none of the models provided an accurate estimation of the 1RM during both DLn (r = 0.92–0.98; absolute errors: 6.6–8.1 kg) and DLw (r = 0.80–0.93; absolute errors: 12.4–16.3 kg) according to our criteria. Therefore, these results suggest that the 1RM for both DLn and DLw should not be estimated through the recording of movement velocity if sport professionals are not willing to accept more than 5 kg of absolute errors.