Sprint swimming performance, particularly in the 50-meter freestyle event, represents the ultimate expression of alactic and lactic anaerobic capacities in competitive swimmers (1, 2). This event, typically lasting 21–28 s depending on athlete level, primarily demands immediate and short-term energy systems, with an estimated 65%–75% contribution from anaerobic metabolism (3). Within this specific physiological context, upper-body muscular power and the capacity to sustain a high velocity in the face of progressive lactate accumulation and neuromuscular fatigue emerge as critical performance determinants (4).

The scientific literature clearly established the fundamental role of upper-body muscular power in sprint performance. A recent meta-analysis by Crowley et al. (5) demonstrated significant correlations (r = −0.68 to −0.79) between bench press power measurements and 50-meter freestyle performance in competitive swimmers. More specifically, the force-velocity profile approach, validated by Samozino et al. (6), enables more refined characterization of athletes’ neuromuscular qualities. Previous studies have shown that elite swimmers exhibit maximum muscular power (Pmax) values 18%–25% higher than their lower-level counterparts (7), underscoring the importance of this physical quality for high-level performance. The transfer of this Pmax to effective propulsion in water constitutes a complex process that remains poorly quantified (8). This methodological limitation represents a significant gap in our understanding of sprint swimming performance.

Concurrently, the assessment of anaerobic capacities through ecological tests like the Repeated Sprint Swimming Test has gained popularity for its predictive validity. The work of Bishop et al. (9) found that mean time during an 8 × 15 m protocol correlated at r = −0.81 with 50-m performance time, while the fatigue index (calculated as the percentage decrement in time across sprints) showed a positive correlation (r = 0.62) with performance degradation at the end of the race. These results suggest that the capacity to maintain high velocity across multiple repetitions is a more relevant indicator of sprint performance than isolated maximum velocity (10). Nevertheless, understanding the interactive relationship between both approaches (i.e., dry-land muscular power and aquatic anaerobic capacities) and their relative contribution to swimming performance remains necessary.

However, despite these advances, several limitations persist in the current literature. First, few studies have simultaneously integrated dry-land power measurements and aquatic anaerobic capacities into a comprehensive predictive model. For instance, Garrido et al. (11) examined the effects of combined dry-land and aerobic training on performance but did not develop a statistical model to quantify their joint predictive power. Similarly, Loturco et al. (12) reported correlations between tethered swimming force, dry-land power, and sprint performance, yet their analysis stopped short of constructing a multivariate model to determine the relative contribution of each measure to competitive race times. Second, the interactive relationship between these different variables and their relative contribution to performance remains insufficiently quantified (13). Finally, existing studies often feature limited sample sizes or heterogeneous populations, compromising the generalizability of results. These gaps highlight the need for an integrative model, the distinct roles of neuromuscular power and metabolic capacity, into actionable insights for coaching practice. Such a framework would empower coaches to prioritize training interventions based on evidence of what matters most for sprint performance.

The present study therefore aimed to combine these different methodological approaches to develop an integrative model of sprint swimming performance. More specifically, our objectives were: (i) to quantify the relationships between upper-body muscular power, RSS variables, and 50-meter freestyle performance; (ii) to establish a multiple regression model predicting performance from these variables; (iii) to identify the relative contributions of each determinant to performance variance. Based on existing data, we hypothesize that the combination of dry-land power measurements and RSS variables would explain more than 75% of the variance in 50-meter freestyle performance.

The statistical analysis revealed a strong relationship between maximum muscular power and swimming performance. Pearson's correlation showed a very strong negative association between power output and 50-m swimming time (r = –0.86, p < 0.001), indicating that swimmers with higher power levels achieved faster 50-m performance (Figure 1, Table 1).

Correlations between the RSS variables and 50-m performance showed significant positive relationships for fastest time (r = 0.83, p < 0.001), mean time (r = 0.78, p < 0.001), and total time (r = 0.78, p < 0.001). Fatigue index did not show a significant correlation (r = 0.05, p = 0.78) (Figure 2, Table 1).

The multiple regression model including Pmax, mean and fastest time, fatigue index, and total time explained 86% of the variance in 50-m performance [R² = 0.862, adjusted R² = 0.844, F_(5,31) = 48.37, p < 0.001]. Among the predictors, only Pmax was a significant independent determinant (β = –0.027, p < 0.001). The other RSS variables did not significantly contribute to the prediction once Pmax was included in the model (Table 2).

The main objective of this study was to examine the combined influence of upper-body muscular power and in-water RSS variables on 50-m freestyle performance in competitive swimmers. The results provide strong evidence that maximum muscular power is the primary determinant of sprint swimming performance, while repeated sprint swimming variables contribute secondarily. Together, these parameters explained 86% of the variance in 50-m performance, emphasizing the decisive role of neuromuscular power in short-distance swimming (22–24).

The very strong negative correlation between muscular power and 50-m time (r = –0.86) confirms the dominant contribution of upper-limb propulsion to sprint swimming performance. This observation aligns with biomechanical evidence showing that over 70% of total propulsive force in front crawl originates from upper-limb actions (8, 25, 26). Comparable studies by Morouço et al. (4) reported similar associations between dry-land power and sprint times (r = –0.82), supporting the ecological validity of load–velocity profiling for assessing sprint swimming potential. Unlike those investigations, which relied on single-load power assessments, the present study used a comprehensive profiling method, allowing more precise quantification of the mechanical capacity underlying performance (6, 15). The current regression coefficient (β = –0.027, p < 0.001) further demonstrates that improvements in muscular power translate directly into faster swimming, highlighting the practical importance of power-oriented dry-land training (7, 25).

Strong positive correlations were observed between RSS time variables and 50-m performance, particularly for the fastest 15-m sprint time (r = 0.83) and the mean 15-m sprint time (r = 0.78), indicating that swimmers who performed better in the repeated sprint swimming test also achieved faster 50-m race times, in agreement with previous studies linking repeated-sprint ability to anaerobic power and sprint performance (9, 27). However, despite these strong bivariate relationships, RSS variables did not remain a significant independent predictor in the multiple regression model, likely due to the substantial shared variance with dry-land muscular power measures. Both RSS performance and muscular power rely on common neuromuscular and anaerobic mechanisms, such as maximal force production, rate of force development, and phosphagen system efficiency (12, 13), and when entered simultaneously into the model, muscular power appears to capture a greater proportion of the variance in 50-m performance, thereby attenuating the unique predictive contribution of RSS variables.

Interestingly, the fatigue index from the repeated sprint test showed no significant relationship with 50-m performance (r = 0.05). This suggests that the specific capacity to resist fatigue across repeated bouts (as measured by the RSS decrement) is a distinct construct that may be less determinant for a single, brief maximal effort. While anaerobic lactic metabolism undoubtedly contributes to 50-m performance (3), our finding indicates that inter-individual differences in the rate of fatigue development during a repetitive test are not a key discriminator of 50-m time. This aligns with energy system models (28, 29) where the anaerobic alactic system provides the critical immediate power for the start and initial high velocity, and where a high anaerobic lactic capacity may support power maintenance without necessarily being reflected in a gross fatigue index from a different task structure. Thus, for the 50-m event, the ability to generate maximal instantaneous power (largely alactic) and to sustain a high power output (supported by lactic capacity) appears paramount, whereas the fatigue resistance specific to repeated efforts with short recovery may be a more specialized quality for different competitive demands.

The integration of dry-land and in-water measures offers a comprehensive view of performance determinants. While both types of measures are related to sprint success, the maximal muscular power derived from the force-velocity profiling method remained the only significant predictor after accounting for RSS variables. This suggests that this specific assessment captures the fundamental neuromuscular capacity (Pmax, F0, V0) underpinning overall sprint ability more precisely than simpler, single-load power measurements. This finding contrasts with earlier skepticism regarding the transfer of dry-land power to aquatic performance (30), demonstrating instead that dry-land power assessment provides essential information for individualized training prescription and talent identification (11).

This study is limited by its cross-sectional design, which prevents causal inference, and by its homogeneous sample of national-level male swimmers, reducing generalizability to other groups. Furthermore, the absence of direct metabolic measurements (e.g., blood lactate) prevents precise quantification of anaerobic metabolic involvement, requiring interpretation based on performance indicators alone. Moreover, technical, anthropometric, and start/turn parameters were not directly analyzed. Future research should employ longitudinal designs to confirm the causal link between increases in dry-land power and improvements in sprint performance (7). Including female swimmers, younger athletes, and different stroke specialists would enhance external validity. Integrating biomechanical and kinematic analyses with physiological testing could also clarify how muscular power improvements translate into propulsive efficiency in water (28, 31).

The findings of this study confirmed that maximum muscular power where assessed from the force-velocity profile in the bench press, is the primary determinant of 50-m freestyle performance, accounting for the majority of the observed variance, while repeated sprint swimming variables play a secondary role. These results emphasize the importance of prioritizing dry-land strength and neuromuscular power development, complemented by in-water training targeting sprint performance and anaerobic capacity. For coaches and sport scientists, integrating load–velocity profiling with repeated-sprint testing provides a reliable framework for performance monitoring and individualized training prescription in competitive swimmers.