The primary objective of competitive swimming is to efficiently traverse a set distance in the shortest time possible. Therefore, improving muscular strength for increased propulsion, maintaining streamlined alignment, and minimizing drag through optimal body positioning are vital for enhancing athletic performance in swimmers (Fig, 2005; Kibler et al., 2006). Proper alignment of the head, shoulders, trunk, pelvis, and lower limbs forms the technical foundation of swimming, and aligning these body segments in a nearly straight line reduces hydrodynamic resistance and enhances swimming efficiency (Willardson, 2007; Jia et al., 2022). Trunk muscles play a crucial role in maintaining body posture and providing active stabilization in the unstable aquatic environment (Patil et al., 2014). Insufficient strength in trunk muscles can lead to energy wastage due to compromised stabilization (Martens et al., 2013; Khiyami et al., 2022). Furthermore, maintaining a stable body position during swimming is crucial for optimizing power output from both the upper and lower limbs (Willardson, 2007; Strzala et al., 2012). Trunk muscle training is considered beneficial as it enhances stabilization, resulting in increased force production by the limbs and efficient force transmission between the trunk and extremities (Hibbs et al., 2008; Dingley et al., 2014; Weston et al., 2015). In sprint swimmers, trunk strength training is thought to induce neuromuscular adaptations such as enhanced neural activation, improved motor unit synchronization, optimized recruitment patterns, and reduced inhibitory reflexes (Strzala et al., 2012; Dingley et al., 2014), and these adaptations directly contribute to enhancing stroke efficiency and athletic performance.

However, empirical evidence supporting trunk strength’s direct impact on swimming performance is conflicted. While core training improves swim efficiency in adolescents (Karpiński et al., 2020), elite studies show minimal transfer (Keiner et al., 2021). This may stem from methodological limitations: most assessments use isometric/endurance tests ill-suited to capture velocity-specific strength adaptations crucial for sprint swimming. Trunk strength is typically measured by the number of repetitions and the load lifted (Faries and Greenwood, 2007). The endurance of the anterior, posterior, and lateral trunk muscles is assessed using trunk flexion, trunk extension, and right and left side bridge tests (Reed et al., 2012; Shamsi et al., 2016). Most studies have evaluated the maximal isometric strength and endurance of the trunk, with little attention paid to the load, force, and power-velocity relationships (Zemkova and Zapletalova, 2022). Therefore, further research is needed to address this gap in the literature and investigate strength and power-related measures within cross-sectional and intervention studies. The lack of specificity in many dryland strength training programs is frequently cited as a reason for their limited effectiveness in improving swimming performance (Girold et al., 2007). Critically, no study has examined isokinetic trunk strength which quantifies torque-velocity profiles in relation to segmental swimming performance, despite its potential to reveal sport-specific neuromuscular adaptations.

This gap is compounded by a lack of sport-specific validation. Static endurance testing is well-suited for assessing postural stability in endurance-oriented sports, while Isokinetic assessmen may be more appropriate for disciplines emphasizing strength (Pérez-Olea et al., 2018). Isokinetic testing provides precise measurements of muscle strength under controlled conditions, including angular velocities, contraction types, and motion ranges. It is a well-established method for assessing trunk strength, with testing protocols in kneeling (Palmer and Uhl, 2011), standing (Andre et al., 2012), and seated positions (Juan-Recio et al., 2017). The seated test protocol specifically isolates lumbar motion and reduces hip force interference. Research has shown that trunk endurance and balance control may not be constraining factors for exceptional performance (Wirth et al., 2022; Keiner et al., 2021). However, some studies have highlighted that the intensity of trunk rotation significantly impacts athletic performance, such as its effect on baseball pitching speed (Taniyama et al., 2021), canoe sprint force (Zinke et al., 2019), and golf swing speed (Gordon et al., 2009). Its role in freestyle swimming remains untested where transverse-plane rotation dominates the roll mechanism (Andersen et al., 2021).

This study examines freestyle swimming (the dominant competitive stroke) using isokinetic dynamometry to evaluate trunk strength-performance relationships. It aims to: (1) quantify trunk strength characteristics across velocities (60°/s, 120°/s) in elite freestyle swimmers; (2) Analyze associations between strength metrics and 100 m sprint performance segments (start, turn, overall). Given the velocity-specific nature of force production in aquatic environments, we hypothesize that (1) peak torque at 120°/s would correlate more strongly with sprint performance than 60°/s, reflecting the high-velocity force demands of the catch phase (Wirth et al., 2022); (2) trunk rotation torque will show a stronger relationship to performance than flexion/extension torque in freestyle, analogous to transverse-dominant sports (Zinke et al., 2019), due to shared roll mechanics.

The data presented in Table 2 indicate significant sex-based differences in peak torque (PT) for trunk extension and rotation. Trunk extension PT was significantly greater than flexion PT at both high and low angular velocities (p < 0.01). Although left rotation PT was marginally higher than right rotation PT, this difference was not statistically significant. High-speed conditions elicited greater PT in trunk extension and rotation compared to low-speed conditions (p < 0.05).

Table 3 displays the peak torque angle distribution of the trunk, showing that peak torque consistently occurs at mid-range angles across all movements, with minimal variability at 60°/s. Notably, there is a significant angle reduction in extensors and rotators at 120°/s, while the flexor angle remains stable but with high standard deviations.

Table 4 presents the mean test scores, standard deviations, and reliability metrics, including the coefficient of variation (CV) and intraclass correlation coefficient (ICC). All tests demonstrated acceptable absolute reliability, with CV values under 8%. Relative reliability was excellent, as indicated by ICC values ranging from 0.968 to 0.998.

Table 5 presents correlation coefficients between trunk strength and swimming performance segments. Combined sample correlations are provided for reference only and may be confounded by sex differences. After Benjamini–Hochberg FDR correction (q < 0.05), no significant correlations were identified. In male swimmers, trunk strength generally exhibited non-significant or positive correlations with performance metrics showed mixed correlation directions (e.g., 60°/sF-15 m: r = −0.055; 120°/sE-15 m: r = 0.21). The strongest association was 60°/sRR-100 m (r = 0.465, p = 0.045), but it did not survive FDR correction. In female swimmers, trunk strength metrics did not show any statistically significant correlations with performance segments. However, early-phase performance (15 m) exhibited moderate negative trends with flexion torque: at 60°/s flexion (r = −0.479) and 120°/s left rotation (r = −0.177), in contrast to the positive correlations observed in males.

This study utilized isokinetic dynamometry to assess trunk strength at varying angular velocities and explored its relationship with 100 m freestyle swimming performance. Results indicated that sprint freestyle swimmers exhibited higher peak torque in trunk rotation and flexion during high-speed movements compared to low-speed ones. While swimmers generated greater rotational torque at high speeds (p < 0.05), this mechanical advantage did not enhance performance. The lack of FDR-significant associations suggests trunk strength, as measured by isokinetic dynamometry, may not directly determine swimming performance in elite athletes. Training approaches should prioritize inter-segmental coordination, with a focus on optimizing torque transfer from the trunk to the extremities. These findings offer valuable insights for refining training strategies in sports.

The data aggregated across protocols (Table 2) show trunk extension peak torque (PT) significantly exceeded flexion PT at all velocities (p < 0.01), with the disparity increasing at higher speeds (e.g., extension-flexion difference: 67.2 Nm at 60°/s vs 101.3 Nm at 120°/s). Consistent with the rapid torque-generation demands in swimming, rotational PT at 120°/s surpassed 60°/s values (p < 0.05, Table 2), contrasting the typical velocity-force relationships observed in non-aquatic sports. The angle distribution patterns (Table 3) further revealed that the extensors/rotators exhibited velocity-dependent shifts toward earlier ranges of motion (ROM), with peaks at 5.6°–6.6° at 120°/s compared to 12.9°–17.4° at 60°/s. In contrast, the flexor angles remained stable but with increased variability at 120°/s (SD = 11.29° vs 3.84° at 60°/s), suggesting compromised measurement consistency under high-velocity conditions. These torque-angle profiles align with the biomechanics of freestyle swimming, where rapid torso rotation coordinates with limb propulsion (Ruiz-Navarro et al., 2025), but caution is warranted when interpreting high-velocity data due to the reduced isokinetic phase duration.

The study investigated the relationship between trunk strength and sprint performance. Contrary to hypotheses 1-2, which peak torque at 120°/s shows a stronger correlation with sprint performance than at 60°/s and rotational torque is more closely linked to performance than flexion/extension metrics, no statistically significant correlations were found after correcting for multiple comparisons. The observed nominal associations (p < 0.05 uncorrected) were likely the result of Type I errors due to the large number of statistical tests conducted. While trunk rotation strength has been shown to predict performance in terrestrial rotation-dominant sports like baseball (Taniyama et al., 2021), its transfer to aquatic environments appears limited. Similarly, weak trunk strength-performance links have been reported in soccer (Keiner et al., 2015) and swimming (Prieske et al., 2014), underscoring the context-dependent nature of physical adaptations. Methodological and biomechanical factors may explain the limited relationship between dry-land trunk strength and swimming performance. Isokinetic tests conducted on land cannot replicate the triaxial torque demands experienced during aquatic propulsion (Pérez-Olea et al., 2018). Fluid dynamics favor energy transfer via coordinated body roll over maximal torque output (Zamparo et al., 2012; Ruiz-Navarro et al., 2025). Furthermore, Andersen et al. (2021) found that torso muscles are more crucial for postural stability and control during front crawl swimming than for torso rotation. In conclusion, swimming performance appears to depend more on the technical integration of movement patterns than on isolated trunk strength. The context-specific nature of physical adaptations should be considered when evaluating the relevance of strength measures to aquatic sports performance.

Despite the absence of significant within-sex associations after FDR correction, visual inspection of sex-stratified scatterplots (Figure 2) revealed nominal trends that merit cautious interpretation. Stratified data suggested potential sex-specific patterns: males exhibited a tendency toward positive associations between rotation torque and performance, while females showed a slight trend linking flexion to acceleration and passive relationships between rotation torque and performance. Although these observations are not statistically significant, they suggest the hypothesis of differing kinetic strategies: males might use rotation to compensate for limited sagittal-plane force transmission, while females may focus on flexion-driven streamline control (Andersen et al., 2020). It is critical to emphasize that these patterns did not survive multiple comparisons correction and should be considered strictly exploratory. They serve primarily to highlight the complexity of trunk biomechanics and the necessity for sex-specific investigations in future research with larger samples. Methodologically, pooled correlations were confounded by biological dimorphism, such as males’ generally higher strength and performance. The lack of within-sex associations after stratification confirms that initial combined-sample trends reflected sex differences rather than causal relationships.

The present study challenges the utility of isolated trunk strength as a predictor of elite swimming performance. The non-significant outcomes observed underscore a fundamental disconnect between conventional strength metrics and the biomechanical demands of aquatic locomotion, where hydrodynamic efficiency likely supersedes raw torque output (Zamparo et al., 2012). Within a multifactorial training framework (Wirth et al., 2022), trunk strength should be integrated with technical skills (e.g., stroke efficiency) and physiological capacities, rather than being considered an isolated performance predictor. Importantly, individualized dryland training programs (e.g., velocity-specific loads) can further modulate strength adaptation (Izquierdo et al., 2002; Amaro et al., 2018), potentially explaining the null correlations observed. Several limitations of the current study warrant cautious interpretation. First, peak torque measurements at 120°/s occurred at low joint angles (5.6°–6.5°), which may have captured acceleration artifacts rather than true trunk strength. Second, the male-skewed sampling (M:22, F:12) obscured potential sex-specific differences in the relationships examined. Third, the study did not quantify body roll kinematics or the periodization of dryland training, both of which may have influenced the observed outcomes. Future research should employ instrumented tethered swimming systems to directly quantify trunk force production during aquatic locomotion, synchronizing these measurements with 3D motion capture of roll mechanics. Additionally, sex-stratified analyses in balanced cohorts are needed to identify potential dimorphic strategies. Integrating longitudinal training metrics (e.g., velocity-specific dry-land loads) with comprehensive biomechanical profiling may further elucidate the complex relationships between trunk strength, technical skills, and swimming performance.

The present study utilized isokinetic dynamometry at angular velocities of 60°/s and 120°/s to investigate the relationship between trunk strength and 100 m performance. The findings did not reveal any statistically significant correlations, challenging the notion that isolated trunk strength is a reliable predictor of swimming performance. This suggests that hydrodynamic efficiency in swimming may depend more on inter-segmental coordination and technical efficiency than on maximal torque output alone. Consequently, our study does not support the prioritization of isolated trunk strength training as a performance-enhancing strategy for swimmers. The observed trends between sexes, while intriguing, are exploratory in nature and underscore the need for sex-specific investigations with larger sample sizes in future research.