The study aimed to investigate the effect of combined balance and plyometric training on dynamic balance and quickness performance of elite badminton athletes. Sixteen elite male badminton players volunteered to participate and were randomly assigned to a balance-plyometric group (PB: n = 8) and plyometric group (PT: n = 8). The PB group performed balance combined with plyometric training three times a week over 6 weeks (40 min of plyometrics and 20 min of balance training); while the PT group undertook only plyometric training for the same period (3–4 sets × 8–12 reps for each exercise). Both groups were given the same technical training (badminton techniques for 6 days a week). The dynamic stability and quick movement ability were assessed at baseline and after the intervention by measuring the performance of dynamic posture stability test (DPSI and COP), T-running test and hexagon jump test. The results showed that compared to PT, PB induced significantly greater improvements in F-DPSI, L-DPSI (p = 0.003, 0.025, respectively), F-COP_AP, F-COP_ML, F-COP_PL, L-COP_PL (p = 0.024, 0.002, 0.029, 0.043, respectively), T-running test and hexagon jump test (p < 0.001). The change in L-DPSI, L-COP_AP, L-COP_ML did not differ between PB and PT (p > 0.907). The findings suggest that combined training holds great promise of improving the dynamic balance and quickness performance in elite badminton athletes.

Background: Short-term (5 min) heart rate variability (HRV) analysis is widely used in assessing autonomic nervous system activity during exercise. While shortening the HRV measurement duration can help improve its application efficiency, its accuracy needs to be verified. This study investigated the agreement between ultra-short-term (UST) HRV (3 or 4 min) and standard 5 min HRV and explored the optimal recording duration under resting and post-exercise conditions. Methods: Fourteen participants exercised on a cycle ergometer at 60% of their maximum peak power. Data were collected during the rest condition (Pre-E) and three post-exercise conditions (Post-E1, Post-E2, and Post-E3), with indicators of the standard deviation (SDNN) of the ultra-short and short-term RR intervals and the root mean square (RMSSD) of the continuous difference between RR intervals. Repeated measures ANOVA, Cohen’s d statistic, Bland–Altman analysis, and interclass correlation coefficients (ICC) assessed the agreement between UST-HRV and ST-HRV. Results: The consistency results of SDNN and RMSSD in resting and post-exercise were different. At the Pre-E, Post-E2, and Post-E3 phases, no statistical differences for SDNN and RMSSD were observed, with ICCs surpassing 0.9, indicating a high level of agreement. However, at Post-E2, there was a significant difference between 3 min RMSSD and 5 min RMSSD (p < 0.05), as well as between 3 min SDNN, 4 min SDNN, and 5 min SDNN (p < 0.05). Furthermore, the limits of agreement were observed to decrease as the time duration increased in Bland–Altman plots. Conclusions: UST-HRV analysis is a reliable substitute for standard 5 min HRV assessment, particularly during resting conditions. For post-exercise measurements, assessing the appropriateness of a 3- or 4 min duration based on the exercise’s length is recommended to ensure accuracy and reliability.

Quantifying external load using kinematic variables from inertial devices provides crucial insights into player performance. This study analyzed load variables in men's singles badminton matches, differentiating between set outcomes (winners vs. losers) across different score gaps (0-5, 6-10, >10 points). Data were collected from 18 highly trained players (110 sets) using the Catapult Vector S7 microtechnology units, housed with a 100hz accelerometer, gyroscope and magnetometer. The results indicated that set winners exhibited a lower player load (estimated difference [ED] and 95% CI = -14.1 [-26.20, -1.78], p = 0.023), covered less distance (ED = -134.84 [-248.99, -17.01], p = 0.02), performed fewer explosive efforts (defined as the sum of high-intensity accelerations, decelerations, and changes of direction [COD]) (ED = -19.75 [-31.85, -7.33], p = 0.002), CODs (ED = -21.02 [-34.18, -7.67], p = 0.003), and accelerations (ED = -7.03 [-13.16, -0.73], p = 0.03) than set losers. Notably, when the score gap was narrow (0–5 points), set winners performed more explosive efforts and CODs than set losers (adjusted p = 0.0412 and 0.0499, respectively). However, as the score gap widened (6–10 and >10 points), set winners exhibited fewer explosive efforts and CODs (all adjusted p < 0.05). Furthermore, when the score gap exceeded 10 points, set winners demonstrated a lower player load, covered less distance, and performed fewer right-side CODs (all adjusted p < 0.05). These findings suggest that winners generally have a lower external load in men's singles badminton matches. However, when opponents are evenly matched, on-court movement may play a pivotal role in determining the outcome.

Objective: The aim of this study was to develop and evaluate binary classification models to detect physical fatigue in badminton athletes using inertial measurement unit (IMU) data and machine learning algorithms. Methods: Thirty-two collegiate badminton athletes participated in this study. Movement data of these participants were collected using multi-sensor IMUs placed on key body regions and a single forearm IMU sensor before and after fatigue induction. Feature selection was performed using Lasso regression to identify the most relevant kinematic features. Six machine learning models—support vector machine (SVM), random forest (RF), logistic regression (LR), k-nearest neighbors (KNN), decision tree (DT), and naïve Bayes (NB)—were applied to construct binary classification models to detect fatigue. Models’ performances were evaluated based on area under the curve (AUC), accuracy, sensitivity, precision, and F1 score. Results: Overall performance of the SVM model was the best across both multi-sensor and single-sensor configurations, with AUC values of 0.89 and 0.95. The single-sensor forearm IMU model achieved high predictive accuracy, underscoring the feasibility of using simplified setups for fatigue monitoring. Feature selection by Lasso regression revealed key fatigue indicators, including forearm kinematic features, contributing significantly to model accuracy. Conclusion: IMU data combined with machine learning models can reliably be used to assess physical fatigue in badminton athletes. High performances of multi-sensor and single-sensor configurations suggest flexibility in model application, supporting real-time, field-based monitoring for fatigue management, performance optimization, and injury prevention in sports. Future research should validate these models in broader athletic populations and explore additional data sources to enhance their predictive abilities.

Background: Basketball requires repeated jumping, short-distance acceleration, rapid deceleration, and frequent changes of direction, yet it remains unclear whether French contrast training (FCT) produces greater adaptations than equal-load traditional training (ELT) in female collegiate basketball players. This study examined the effects of a 10-week FCT program on lower-limb power, short-distance acceleration, change-of-direction performance, and one-repetition maximum (1RM) back squat performance. Methods: In this randomized controlled trial, twenty-five female collegiate basketball players were randomly allocated to an FCT group (n = 13) or an ELT group (n = 12). Jump, sprint, change-of-direction, and 1RM back squat outcomes were assessed before and after training and analyzed using two-way repeated-measures ANOVA. Results: Significant group × time interactions favored FCT for squat jump, countermovement jump, drop-jump height, reactive strength index (all p < 0.001), and eccentric utilization ratio (p = 0.003). Significant group × time interactions were also observed for 10-m sprint time, 10-m maximal velocity, 10-m maximal acceleration, lane agility, and basketball-specific change-of-direction performance (all p < 0.001). Post hoc analyses showed significant pre-to-post improvements in the FCT group for all jump-related, sprint, and change-of-direction outcomes (p = 0.035 for eccentric utilization ratio; all other p < 0.001). The ELT group also improved in these outcomes, although the magnitude of improvement was generally smaller. For 1RM back squat, there was a significant main effect of time (p < 0.001), with significant improvements in both groups (both p < 0.001), whereas the group × time interaction was not significant (p = 0.106). Conclusions: FCT may provide greater benefits than ELT for improving explosive and basketball-specific rapid movement outcomes in female collegiate basketball players. These findings may help strength and conditioning coaches select FCT when the goal is to improve explosive and rapid multidirectional movement performance rather than maximal strength alone.