Acute effects of an individually timed lower-limb post-activation performance enhancement protocol on 50 m front crawl performance and race kinematics in national-level swimmers Authors: Arkadiusz Stanula, Marek Strzała, Wojciech Sadowski, Wojciech Rejdych, Katja Weiss, Jakub Karpiński, Robert Wilk, Thomas Rosemann, Beat Knechtle Published: 2026 ABSTRACT AbstractPost-activation performance enhancement (PAPE) has been proposed as an acute strategy to improve explosive performance, yet its effects on sprint-swimming performance remain unclear, particularly when individualized recovery timing is considered. This study examined the acute effects of a lower-limb PAPE protocol, applied using an individually determined recovery time (IRT), on 50 m front crawl performance and selected start-, race-, and turn-related kinematic variables in national-level male swimmers. Fourteen male national-level swimmers completed a randomized repeated-measures cross-over design comprising control (CON) and PAPE conditions. In the PAPE condition, swimmers performed a lower-limb conditioning protocol consisting of loaded lunge-position squats on a Keiser Squat machine, followed by a maximal 50 m front crawl trial at the end of the IRT. In the CON condition, swimmers completed the same swimming trial without the conditioning stimulus. Lower-limb power output increased significantly after the conditioning activity and reached its highest mean value at 6 min post-exercise (F4,52 = 97.14, p < 0.001). However, the PAPE protocol did not significantly improve overall 50 m performance (25.508 ± 0.856 s vs. 25.529 ± 0.773 s; mean difference = − 0.021 s, 95% CI: −0.148 to 0.107; p = 0.731) or most start-related variables. The only significant effect was observed for turn time, which was lower in the PAPE condition than in CON (7.707 ± 0.251 s vs. 7.771 ± 0.261 s; mean difference = − 0.064 s, 95% CI: −0.126 to − 0.001; p = 0.046). Responder analysis revealed substantial inter-individual variability across outcomes. In conclusion, a lower-limb PAPE protocol applied using the IRT did not improve overall 50 m front crawl performance in national-level male swimmers, although it produced a small improvement in turn time. These findings suggest that the effects of lower-limb PAPE in sprint swimming may be segment-specific rather than global, and that practical implementation should be individualized rather than applied uniformly across swimmers.Keywords: Warm-up strategies, Sprint swimming, Start performance, Turn performanceSubject terms: Health care, Medical research, Physiology Introduction Performance in sprint swimming is determined by the combined contribution of several race segments rather than by clean-swimming speed alone. In short-distance events, the start, underwater phase, turn, and finish can all meaningfully affect the final outcome, and race-analysis studies have shown that even small changes in these components may have practical consequences for overall performance1. This is especially relevant in sprint events, where outcomes are often separated by very small time margins. The importance of the start is further supported by biomechanical evidence showing that overall start performance depends on coordinated interactions between the on-block, flight, entry, and underwater phases2,3. Likewise, early start performance, including time to 5 m, has been shown to depend on specific kinetic and kinematic characteristics rather than on force production alone4. Among the acute strategies proposed to enhance explosive performance, post-activation potentiation (PAP) and post-activation performance enhancement (PAPE) have received considerable attention. Classical PAP refers to a transient physiological phenomenon typically assessed through electrically evoked twitch responses and commonly associated with mechanisms such as phosphorylation of myosin regulatory light chains5. By contrast, PAPE refers to the broader enhancement in voluntary performance observed after a conditioning activity in applied sport settings6. Contemporary literature emphasizes that these terms should not be treated as interchangeable, because improvements in voluntary performance may reflect not only potentiation, but also other mechanisms, including changes in muscle temperature, neural activation, and contractile readiness6,7. This distinction is particularly important in sport-specific studies, where the outcome of interest is actual performance rather than isolated twitch responses. Interest in PAPE has grown because conditioning activities can acutely improve subsequent performance in explosive tasks. However, the magnitude and direction of the response appear to depend strongly on the exercise mode, the athlete’s training background, and the recovery interval between the conditioning activity and the subsequent task6. Accordingly, individualized rather than fixed timing has been proposed, as the balance between fatigue and performance enhancement may differ substantially across athletes6,7. The practical importance of recovery timing is also supported by swimming studies showing that post-warm-up transition duration can meaningfully influence subsequent performance8,9. In the present study, this concept was operationalized as the individual recovery time (IRT), defined as the post-activation time point at which each swimmer achieved the highest power output during a standardized testing procedure. Swimming provides a particularly relevant context in which to test such strategies. Lower-limb explosive force contributes substantially to the block phase, take-off, flight trajectory, water entry, and the subsequent underwater transition during the start2,3. In short-course swimming, turn performance is also a major success factor, with faster and more consistent turns distinguishing better-performing swimmers10. Reviews further suggest that dry-land strength and conditioning may positively influence starts and turns, although the available evidence remains heterogeneous with respect to protocols, populations, and outcome measures11,12. These considerations have led researchers to examine whether post-activation strategies can improve swimming-specific performance. Previous studies have reported beneficial effects of conditioning activities on swimming-start variables and, in some cases, sprint-swimming outcomes. Kilduff et al.13 observed improved swim-start performance following a PAP protocol in international sprint swimmers, while Cuenca-Fernández et al.14 showed that lower-limb activation protocols such as lunge and YoYo squat exercises could enhance start-related performance measures. Hancock et al.15 further reported improved 100-m freestyle performance in collegiate swimmers following a PAP intervention. More recently, Cuenca-Fernández et al.16 showed that PAP-based activation protocols may improve the first meters of a 50-m freestyle race without necessarily enhancing total performance time, and Đurović et al.17 reported that warm-up strategies including a PAPE component can affect swim-start performance. In national-level swimmers, Sarramian et al.18 also examined PAP-based warm-up strategies before 50 m freestyle, further highlighting the practical relevance of this topic. However, the available literature remains difficult to synthesize because studies differ substantially in the type of conditioning activity used, the muscle groups targeted, the competitive level of the swimmers, the outcome variables selected, and, crucially, the time allowed between the conditioning activity and the subsequent swimming task. Another important consideration is that the response to post-activation strategies may be task- or segment-specific rather than global. Swimming studies have shown that acute enhancement may differ depending on the movement examined, such as flutter-kick thrust or arm-pull thrust, rather than transferring uniformly across all components of race performance19,20. Therefore, the effects of a lower-limb PAPE stimulus on whole-race sprint performance cannot be assumed a priori from its effects on isolated physical qualities or specific movement tasks. A further methodological issue concerns the classification of swimmer level. Ruiz-Navarro et al.21 proposed a standardized model based on FINA points to improve comparability across swimming studies and to reduce ambiguity associated with labels such as “national-level” or “elite.” Within that framework, the present sample, with a mean of 764 FINA points and individual scores ranging from 667 to 920, can be characterized predominantly as national-standard swimmers, with some athletes approaching higher performance tiers. This distinction is relevant because responses to PAPE may depend on training status and competitive level. Therefore, despite promising previous findings, it remains unclear whether a lower-limb PAPE protocol applied using an individually determined recovery time (IRT), rather than a fixed recovery interval, can improve overall 50 m front crawl performance or whether its effects are confined to selected race segments and race-related kinematic parameters in national-level swimmers. Accordingly, the aim of the present study was to examine the acute effects of a lower-limb post-activation performance enhancement protocol, applied using IRT, on 50 m front crawl performance and selected start-, race-, and turn-related kinematic variables in national-level male swimmers. It was hypothesized that the protocol would affect race segments relying more strongly on lower-limb explosive force production, particularly the start and turn phases, whereas its effect on overall 50 m performance would be less certain. Methods Participants Fourteen male national-level competitive swimmers (age 20.1 ± 1.3 years, body mass 83.2 ± 9.1 kg, height 185.6 ± 6.5 cm) volunteered to participate in this study and provided written informed consent prior to participation. The study was approved by the University Bioethics Committee for Research at the Academy of Physical Education in Katowice (No. 8/2018), and all methods were performed in accordance with the relevant guidelines and regulations and in accordance with the approved study protocol. Participant characteristics are presented in Table 1. All swimmers belonged to the same university team and were ranked within the top 10 nationally. They had at least 10 years of training experience, and their average personal best performance corresponded to 764.1 ± 70.5 FINA points. During the experiment, all swimmers were in the same phase of seasonal preparation, namely the specific preparation subphase. Before the experimental trials, all participants completed a four-week strength-training period involving free weights (dumbbells and barbells) and Keiser pneumatic resistance machines (Keiser Corporation, Fresno, CA, USA). Table 1Physical characteristics of participants (n = 14).VariablesMean ± SDRange95% CIAge (y)20.1 ± 1.319.0–23.019.3–20.8Body weight (kg)83.2 ± 9.168.5–98.078.0–88.5Body height (cm)185.6 ± 6.5177.2–199.6181.9–189.4Percent body fat (%)7.9 ± 2.44.5–13.36.5–9.3Lean body mass (kg)76.6 ± 8.463.5–90.671.8–81.5FINA (points)764.1 ± 70.5667–920728.2–804.2 Physical characteristics of participants (n = 14). An a priori sample size calculation was performed using G*Power (Heinrich Heine University, Düsseldorf, Germany). Based on the primary within-subject comparison between the CON and PAPE conditions in a repeated-measures cross-over design, assuming a small-to-moderate effect size (ES = 0.45), an alpha level of 0.05, and statistical power of 0.80, the required sample size was estimated at 12 participants. To account for possible dropouts, 14 swimmers were recruited. Procedures To evaluate the effects of a lower-limb post-activation performance enhancement (PAPE) protocol on 50 m front crawl sprint performance, a randomized repeated-measures crossover design was used, comprising two experimental conditions: control (CON) and PAPE. All participants completed both experimental conditions, and the order of conditions was randomized. Before the main trials, anthropometric data, including height, body mass, and body composition, were collected. Height was measured using a stadiometer (Seca 213, Seca GmbH & Co., Hamburg, Germany) with an accuracy of 0.5 cm. Body mass and body composition were assessed by bioelectrical impedance analysis using the InBody 220 device (Biospace Co., Japan). All experimental tests were scheduled before the swimmers’ regular daily training. Participants were instructed to avoid physical exertion before testing, to maintain their habitual diet, and to refrain from caffeine and dietary supplements that could influence physical performance. Subsequently, in the Muscle Strength and Power Laboratory at the Academy of Physical Education in Katowice, one-repetition maximum (1RM) for the lower-limb conditioning exercise was determined in accordance with National Strength and Conditioning Association guidelines22. The test was performed on a Keiser Squat machine (Keiser Corporation, Fresno, CA, USA), with participants executing the movement in a lunge position (Fig. 1). This exercise was selected as a compromise between movement specificity and experimental control, because it allowed partial approximation of the stance and lower-limb force-production characteristics of the swimming start while also enabling standardized loading and power-output monitoring for IRT determination6,23. The foot position was individualized for each swimmer so that it replicated the stance adopted on the swimming starting block. The machine arms were adjusted to ensure that, at the lowest squat position (Fig. 1B), the rear knee joint reached 90° of flexion, as verified with a goniometer. After an individualized warm-up, each participant performed a set of 3 to 8 repetitions using a self-selected load. The 1RM was then estimated using the following equation: 1RM = load × [1 + (0.033 × number of repetitions)]24. Squat with the Keiser Squat machine. (A) Initial position; (B) Squat position with the legs setting in a stride; (C) Return to the initial position. One week later, each swimmer’s IRT was determined in order to identify the post-activation time point associated with the highest power output for that individual. For this purpose, participants performed 3 sets of 6 repetitions on the Keiser Squat machine with a load corresponding to 70% of their estimated 1RM, with 1 min of rest between sets. After the final repetition of the third set, participants performed one explosive repetition at 70% 1RM at 2, 4, 6, and 8 min post-exercise, while maintaining correct technique. The time point at which the highest power output was recorded was defined as the IRT. The next stage of the study was designed to examine the acute effects of the lower-limb PAPE protocol, applied using the IRT, on 50 m front crawl performance and selected start-, race-, and turn-related kinematic variables. All swimming trials were conducted in a 25-m indoor pool at the Academy of Physical Education in Katowice. A schematic overview of the experimental design is presented in Fig. 2. Timeline of the experimental procedures. Before each trial, all participants completed a standardized warm-up lasting approximately 15 min and consisting of short land-based and in-water components. The land-based warm-up lasted approximately 5 min and included general dynamic mobilization and low-intensity movement preparation. The in-water warm-up lasted approximately 10 min and consisted of 200 m easy front crawl, 4 × 25 m kick at low-to-moderate intensity, 4 × 25 m front crawl with a gradual increase in speed up to moderate intensity, 2 × 15 m controlled accelerations, and 50 m easy swimming. After the warm-up, participants assigned to the PAPE condition proceeded to the Strength and Power Laboratory, where they performed the lower-limb conditioning exercise on the Keiser Squat machine using the load established during the earlier testing sessions. Immediately after the conditioning activity, the researcher initiated timing of the pre-determined IRT, which continued uninterrupted during the participant’s transition from the laboratory to the pool. The laboratory and swimming pool were located in the same building, and the transfer from the laboratory to the pool deck involved a walking distance of approximately 50 m, including ascent to the first floor via a staircase of 24 steps, and took about 60 s. At the end of the IRT, the swimmer performed a start from the starting block and completed a maximal 50 m front crawl trial under race-like conditions. In the CON condition, participants performed the same standardized warm-up but did not complete the conditioning activity. Instead, they performed the 50 m front crawl trial exactly 8 min after completion of the in-water warm-up. This interval was selected to approximate the average duration of the recovery period applied in the PAPE condition, while excluding the conditioning stimulus itself. After a 1-h passive recovery period, all participants repeated the same land- and water-based warm-up. In accordance with the randomized cross-over design, the experimental conditions were then reversed. Thus, participants who had previously completed the PAPE condition performed the CON trial, whereas those who had previously completed the CON condition performed the PAPE trial. All trials were performed at maximal effort, and standardized verbal encouragement was provided throughout testing. To verify that the 1-h passive recovery period was sufficient to avoid carry-over fatigue between trials, two 50 m front crawl trials were conducted one week before the main experiment under identical conditions, separated by a 60-min interval. The reliability analysis showed excellent reproducibility of performance, with an intraclass correlation coefficient of 0.983 (95% CI: 0.977–0.989). Kinematic measurements During the swimming trials, selected kinematic variables related to the start, race, and turn were recorded. The analyzed variables were defined as follows: dive distance (DD), the distance from the pool wall beneath the starting block to the point of first finger contact with the water (cm); flight time (FT), the time between the last foot contact with the starting block and first finger contact with the water (s); jump time (JT), the time from the start signal to first finger contact with the water (s); horizontal velocity (HV), the horizontal velocity of the hips during the flight phase, calculated as horizontal hip displacement divided by FT (m·s⁻¹)14; time to 15 m (T15), the time from the start signal until the swimmer’s head reached the 15-m mark (s); time to 20 m (T20), the time from the start signal until the swimmer’s head reached the 20-m mark (s); time to 50 m (T50), the official 50-m race time recorded by the Omega timing system (s); and turn time (TT), the time from the moment the swimmer’s head reached the 5-m mark before the turning wall until it reached the 10-m mark after the wall (s). Each 50-m trial was recorded using four high-speed digital video cameras (Fig. 3). The first camera (JVC GC-PX100BE, Japan), operating at 150 Hz, was positioned on a tripod 1.5 m above the water surface and 2 m from the start line along the poolside. This camera was used to analyze DD, FT, JT, and HV. The remaining three cameras (SONY FDR-X3000R, Japan) were mounted on tripods 1 m above the water surface and positioned at 5 m, 15 m, and 20 m from the start line along the poolside. These cameras were used to identify the moments when the swimmer’s head crossed the 15-m and 20-m reference marks and to support determination of turn-related timing. Camera setup for video analysis. All trials were performed in accordance with the start procedure defined by World Aquatics (WA). The Omega starting system emitted synchronized acoustic and light signals. To synchronize the start signal with all video recordings, an additional light device connected to the main starting system was positioned at the 17-m mark and flashed simultaneously with the primary start signal. Video recordings were analyzed using Kinovea software (v. 0.8.26, Kinovea, Paris, France) by two independent researchers experienced in motion digitization. Reference points for analysis were based on calibrated markers placed along the lane in which the trials were performed. To assess inter-rater reliability, six trials were independently digitized and evaluated using intraclass correlation coefficients (ICCs). The ICC values ranged from 0.978 (95% CI: 0.975–0.981) to 0.992 (95% CI: 0.983–0.996). Statistical analysis Means and standard deviations (SD) were used to describe all measured variables. The assumption of normality was assessed using the Shapiro-Wilk test; therefore, parametric statistical analyses were applied. Changes in lower-limb power output measured at baseline and at 2, 4, 6, and 8 min after the conditioning activity were analyzed using a one-way repeated-measures analysis of variance (ANOVA), with time as the within-subject factor. When a significant main effect was detected, pairwise comparisons were performed using Bonferroni-adjusted post hoc tests. The Greenhouse-Geisser correction was applied when the assumption of sphericity was violated. Comparisons between the CON and PAPE conditions for all recorded kinematic variables were performed using two-tailed paired-samples t-tests. The primary outcome of the study was 50 m front crawl performance (T50), whereas the remaining kinematic variables were treated as secondary exploratory outcomes. Percentage changes between conditions were calculated as follows: Δ% = 100 × (PAPE − CON)/CON. The smallest worthwhile change (SWC) was calculated as 0.2 × SD of the CON condition and was used to classify individual responses as improved, unchanged, or impaired. The magnitude of differences between conditions was expressed as standardized effect sizes (ES). Effect sizes were interpreted as trivial (< 0.2), small (0.2–0.6), medium (0.6–1.2), and large (> 1.2) according to Hopkins25(2004). Mean differences and their 95% confidence intervals (CI) were calculated using the convention PAPE − CON. Statistical significance was set at p ≤ 0.05. All statistical analyses were performed using STATISTICA version 13.3 (TIBCO Software Inc., Palo Alto, CA, USA). Figures were prepared using RStudio (Version 2024.04.2 + 764)26; Posit team, 2026), R (Version 4.4.1)27; R Core Team, 2025), and the R packages ggplot228 and Durga29. Results Power output A one-way repeated-measures ANOVA revealed a significant main effect of time on lower-limb power output following the conditioning activity performed on the Keiser Squat machine (F4,52 = 97.14, p < 0.001). Compared with baseline, lower-limb power increased significantly during the recovery period and reached its highest mean value at 6 min, corresponding to a 23.6% increase relative to baseline (2048.2 W vs. 1656.9 W, p < 0.001). Power output decreased slightly at 8 min compared with 6 min, although this difference was not statistically significant. The largest number of swimmers reached their peak power at 6 min (n = 6), whereas 4 swimmers peaked at 4 min, 3 at 8 min, and 1 at 2 min (Fig. 4). Power output during lower-limb exercise before and after the post-activation performance enhancement (PAPE) stimulus. Kinematic variables Table 2 summarizes the effects of the lower-limb PAPE protocol on the selected start-, race-, and turn-related kinematic variables, and the individual response patterns are illustrated in Fig. 5. T50 was considered the primary performance outcome, whereas the remaining start-, race-, and turn-related kinematic variables were analyzed as secondary exploratory outcomes. Table 2Effects of the lower-limb post-activation performance enhancement (PAPE) protocol on selected kinematic variables of the swimming start, race, and turn under control (CON) and PAPE conditions. Note: Values are presented as mean ± SD. Mean differences (△), percentage changes (△ %), and effect sizes (ES) were calculated as PAPE − CON. Thus, negative values indicate lower values in the PAPE condition than in the control condition. CI, confidence interval.VariableCONPAPE△ (△ %)(PAPE − CON)p value95% CIESDD (cm)308.19 ± 16.24311.26 ± 18.313.079 (0.99%)0.137−1.119 to 7.2760.178FT (s)0.26 ± 0.040.26 ± 0.040.000 (0.00%)1.000−0.017 to 0.0170.000JT (s)1.06 ± 0.051.06 ± 0.060.001 (0.16%)0.885−0.020 to 0.0220.027HV (m·s⁻¹)4.54 ± 0.304.40 ± 0.28−0.139 (− 2.86%)0.097−0.307 to 0.029−0.485T15 (s)6.68 ± 0.236.68 ± 0.25−0.003 (− 0.04%)0.886−0.045 to 0.039−0.012T20 (s)9.39 ± 0.289.40 ± 0.320.015 (0.15%)0.582−0.042 to 0.0720.050T50 (s)25.53 ± 0.7725.51 ± 0.86−0.021 (− 0.09%)0.731−0.148 to 0.107−0.025TT (s)7.77 ± 0.267.71 ± 0.25−0.064 (− 0.81%)0.046−0.126 to − 0.001−0.248 Effects of the lower-limb post-activation performance enhancement (PAPE) protocol on selected kinematic variables of the swimming start, race, and turn under control (CON) and PAPE conditions. Note: Values are presented as mean ± SD. Mean differences (△), percentage changes (△ %), and effect sizes (ES) were calculated as PAPE − CON. Thus, negative values indicate lower values in the PAPE condition than in the control condition. CI, confidence interval. Effects of the lower-limb post-activation performance enhancement (PAPE) protocol on selected start-related variables in the control (CON) and PAPE conditions. Panels A, C, E, and G show box-and-whisker plots with individual paired raw data connected by lines between CON and PAPE. The annotation above each comparison bracket indicates the standardized effect size (ES) and its 95% confidence interval. Adjacent panels B, D, F, and H present responder analysis based on the smallest worthwhile change (SWC), expressed as the percentage of swimmers classified as improved, impaired, or unchanged. A small, non-significant increase in DD was observed under the PAPE condition, whereas FT and JT were essentially unchanged between conditions. HV tended to be lower after PAPE, although this difference did not reach statistical significance (Table 2; Fig. 5). Responder analysis showed substantial inter-individual variability across all start-related variables. No significant differences were observed for T15 or T20, and both variables showed mixed responder profiles (Table 2; Fig. 6). The primary performance outcome, overall T50, was unchanged between conditions. Swimmers recorded 25.508 ± 0.856 s in the PAPE condition and 25.529 ± 0.773 s in CON, corresponding to a mean difference of − 0.021 s (95% CI: −0.148 to 0.107; p = 0.731; ES = − 0.025). The responder analysis showed improvement in 5 swimmers (35.7%), no meaningful change in 5 (35.7%), and impairment in 4 (28.6%) (Fig. 6E, F). In contrast, TT was significantly lower in the PAPE condition than in CON (7.707 ± 0.251 vs. 7.771 ± 0.261 s), with a mean difference of −0.064 s (95% CI: −0.126 to − 0.001; p = 0.046; ES = − 0.248). The responder analysis showed that 9 swimmers (64.3%) improved, 3 (21.4%) remained unchanged, and 2 (14.3%) were impaired (Fig. 7A, B). Effects of the lower-limb post-activation performance enhancement (PAPE) protocol on selected race-related variables in the control (CON) and PAPE conditions. Panels A, C, and E show box-and-whisker plots with individual paired raw data connected by lines between CON and PAPE. The annotation above each comparison bracket indicates the standardized effect size (ES) and its 95% confidence interval. Adjacent panels B, D, and F present responder analysis based on the smallest worthwhile change (SWC), expressed as the percentage of swimmers classified as improved, impaired, or unchanged. Effects of the lower-limb post-activation performance enhancement (PAPE) protocol on turn time in the control (CON) and PAPE conditions. Panel A shows a box-and-whisker plot with individual paired raw data connected by lines between CON and PAPE. The annotation above the comparison bracket indicates the standardized effect size (ES) and its 95% confidence interval. Panel B presents responder analysis based on the smallest worthwhile change (SWC), expressed as the percentage of swimmers classified as improved, impaired, or unchanged. Discussion The main finding of the present study was that the primary performance outcome, 50 m front crawl time (T50), was not significantly improved by a lower-limb post-activation performance enhancement (PAPE) protocol applied using IRT in national-level male swimmers. Most start-related variables were also unchanged, whereas turn time showed a small but statistically significant improvement. In addition, responder analysis revealed substantial inter-individual variability across the measured outcomes. The lack of a significant effect on T50 should be interpreted directly rather than softened into a trend-based conclusion. Although acute enhancement after a conditioning activity is often reported in explosive tasks, the magnitude and direction of the response depend on several interacting factors, including fatigue-potentiation balance, conditioning-exercise characteristics, recovery interval, and athlete-specific characteristics5,6,30. In applied sport settings, PAPE does not necessarily transfer to the final outcome of a complex competitive task, particularly when performance depends on the interaction of several race segments rather than on isolated force production alone7. A similar pattern has been reported in swimming-specific studies, where PAP/PAPE-based warm-up strategies improved early race segments without necessarily enhancing total 50 m freestyle time16, and where in national-level swimmers a traditional race-specific warm-up even outperformed some PAP-based alternatives18. More recently, Bufalo et al.31 also reported no improvement in 50 m freestyle performance after a resisted sprint warm-up protocol, suggesting that additional activation may not always translate into better race outcomes. One plausible explanation is that any acute enhancement induced by the lower-limb conditioning activity was too small to influence total race time once embedded within a 50 m sprint. This may be particularly true in well-trained swimmers, in whom margins for improvement are limited and outcome variability may reflect not only physical readiness but also technical stability and race execution. In the present study, the sample was composed predominantly of national-standard swimmers according to the classification model proposed by Ruiz-Navarro et al.21, although some athletes approached higher performance tiers, which may also have contributed to heterogeneous responsiveness. A particularly noteworthy finding was the significant improvement in turn time. Although the conditioning stimulus was originally selected with the expectation that it might benefit start performance through enhanced lower-limb explosiveness, the turn may have been more sensitive to this intervention. Although statistically significant, the magnitude of the observed change (~ 0.06 s) was small, and its practical relevance in competitive settings remains uncertain, particularly because it did not translate into an improvement in overall 50 m performance. In addition, because multiple secondary outcomes were examined without formal correction for multiple comparisons, the statistically significant effect observed for turn time may reflect a Type I error and should therefore be interpreted cautiously. In short-course swimming, turn performance relies heavily on rapid body positioning, effective wall contact, and a forceful push-off, followed by an efficient underwater transition. Previous work has shown that turn consistency and speed are important success factors in short-course events, with better swimmers tending to display faster and less variable turn performances10. In addition, systematic-review evidence suggests that dry-land strength and conditioning may positively affect turn performance, likely because the turn includes a relatively direct force application against a solid surface12. This is also consistent with previous work showing that turning movements are sensitive to race constraints and should not be viewed as purely stereotyped actions32. In this context, the present results may indicate that the applied lower-limb PAPE protocol enhanced a race component in which lower-limb force production can be expressed more directly than during the start. The absence of significant changes in start-related variables contrasts with some previous studies. Cuenca-Fernández et al.33, Đurović et al.17, Hancock et al.15 and Kilduff et al.13 reported beneficial effects of PAP/PAPE-based warm-up strategies on swim-start or sprint-swimming performance. However, the available literature is not fully consistent, and discrepancies likely reflect differences in swimmer level, conditioning stimulus, total warm-up structure, performance task, and especially the timing of the post-activation window. This interpretation is further supported by findings showing that acute enhancement may be more readily expressed in start kinetics than in broader race outcomes33, while other studies have reported no improvement in swim-start or 50 m performance after pre-race activation strategies34,35. Therefore, the present data do not necessarily contradict earlier reports, but rather support the view that the ergogenic value of PAPE in swimming is conditional rather than universal. The present start-related findings may also be explained by the biomechanical complexity of the swimming start. Start performance depends not only on force production on the block, but also on the organization of the on-block, flight, entry, and underwater phases, as well as on the interaction between impulse orientation, body configuration, and transition mechanics1,2. Likewise, the determinants of early start performance, such as time to 5 m, include both kinetic and kinematic factors rather than lower-limb power output alone4. Therefore, an acute enhancement in lower-limb readiness may not translate automatically into better global start performance unless it is integrated into the broader technical execution of the movement. The selected conditioning exercise was also intended as a compromise between start-related specificity and methodological control, rather than as a full simulation of the swimming start6,23. This interpretation is consistent with the present lack of meaningful changes in dive distance, flight time, jump time, horizontal velocity, and time to 15 m. Another important finding was the substantial inter-individual variability observed across the tested variables. For several outcomes, improved, unchanged, and impaired responses coexisted within the same sample. This is highly relevant in the context of PAPE, because contemporary literature emphasizes that post-activation responses vary considerably between athletes and are likely influenced by factors such as training status, strength level, exercise familiarity, recovery kinetics, and the interaction between fatigue and potentiation6,7. From an applied perspective, the responder analysis in the present study suggests that the lack of a group-level effect should not be interpreted as evidence that no swimmer benefited, but rather that the protocol did not produce a sufficiently robust and consistent response across the sample to improve performance at the group level. This observation is especially relevant because the present study addressed one of the central methodological issues in PAPE research, namely the use of individualized rather than fixed recovery intervals. Although IRT was determined from the time point of highest post-activation power output, the present findings suggest that identifying the recovery window alone may be insufficient and that other features of the protocol, including movement specificity, load characteristics, and transfer of acute neuromuscular readiness to a technically complex swimming task, are also critical6. This interpretation is reinforced by previous swimming studies showing that post-warm-up recovery duration can meaningfully affect sprint performance and that additional dryland activation during transition may sometimes enhance outcomes8,9,36. The present findings also fit within the broader literature on dry-land strength and conditioning in swimmers. Systematic reviews have concluded that resistance training and dry-land strength training can positively influence swimming performance, starts, and turns, but they also emphasize substantial heterogeneity in interventions and outcomes11,23. Against this background, the present results suggest that a lower-limb dry-land activation strategy may influence selected race segments, particularly those involving force application against a fixed surface, without necessarily improving overall sprint performance. From a practical perspective, the findings do not support uniform implementation of this lower-limb PAPE protocol as a general pre-race strategy to improve 50 m front crawl performance. However, the heterogeneity of individual responses and the isolated improvement in turn time suggest that this type of strategy may still warrant individualized consideration when the goal is to target selected race segments rather than overall race time. This interpretation is consistent with race-analysis evidence indicating that segment-level data can help identify meaningful performance targets even when whole-race changes are small1. Several limitations should be acknowledged. First, the sample size was modest, although justified by the a priori power analysis and partly offset by the repeated-measures cross-over design. Accordingly, some non-significant findings should be interpreted with caution, as the possibility of Type II error cannot be excluded. Second, only male swimmers were included, limiting generalizability to female athletes. Third, only one lower-limb conditioning protocol was tested, and therefore the results should not be generalized to other PAPE stimuli. Fourth, the study focused on performance and kinematic outcomes and did not include direct mechanistic measurements such as electromyography, muscle temperature, or force-platform-derived variables during the actual start or turn. Fifth, although the experimental design had clear applied relevance, the standardized warm-up itself may have induced a degree of neuromuscular activation, thereby reducing the contrast between conditions. This possibility is consistent with the findings of Bufalo et al.31, who suggested that a conventional warm-up may itself induce priming effects and thereby reduce the likelihood of detecting an additional benefit from a more specific activation stimulus. Sixth, the transition from the laboratory to the pool, which included walking and stair ascent, represented a systematic procedural difference between the PAPE and CON conditions. This transition may have introduced additional physiological activation, such as increased muscle temperature or arousal, which cannot be fully separated from the effect of the conditioning stimulus itself. Therefore, the observed effects cannot be attributed exclusively to the conditioning stimulus alone. Finally, although the reliability data supported the use of a 1-h interval between trials, the within-session cross-over format may still be considered a design limitation. Moreover, because several secondary exploratory kinematic outcomes were examined without formal adjustment for multiple comparisons, the statistically significant effect observed for turn time may reflect a Type I error and should therefore be interpreted cautiously. These limitations mean that the present results should be interpreted as evidence about this specific protocol in this specific population rather than as a general statement about the efficacy of PAPE in swimmers. Future studies should standardize or replicate transition demands in the control condition, compare individualized and fixed recovery intervals directly, examine more task-specific dry-land and in-water conditioning stimuli, and include swimmers of different sexes and performance levels. Greater use of force, kinetic, and underwater kinematic measurements would also help identify whether acute enhancements are expressed preferentially in the push-off, transition, or clean-swimming phases. Finally, given the responder profile observed here, future work should explicitly address which athletes are most likely to benefit from a given protocol and whether those benefits are segment-specific rather than global. These directions are consistent with current calls for more individualized and mechanistically informed PAPE research in sport settings6,7. Conclusion A lower-limb PAPE protocol applied using IRT did not improve overall 50 m front crawl performance in national-level male swimmers. However, it was associated with a small and isolated improvement in turn time, which should be interpreted cautiously given the exploratory nature of the analysis. Taken together, these findings suggest that the acute effects of lower-limb PAPE in sprint swimming may be segment-specific rather than global, and that practical implementation should be individualized rather than applied uniformly across swimmers. Therefore, this protocol does not appear to support general use as a universal pre-race strategy for 50 m front crawl performance, although it may warrant individualized consideration when the aim is to optimize specific race segments such as the turn. Author contributions Conceptualization: A.S., M.S., W.S., W.R., J.K., R.W. and B.K.; Methodology: A.S., M.S. and W.S.; Investigation: A.S., W.S., W.R., J.K. and R.W.; Data analysis: A.S., M.S. and W.S.; Visualization: A.S.; Writing - original draft preparation: A.S., M.S., W.S. and B.K.; Writing - review and editing: A.S., M.S., W.S., W.R., J.K., R.W., T.R., and B.K.; Supervision: A.S. and W.S.; Project administration: A.S. and W.S. Data availability The datasets analyzed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets analyzed during the current study are available from the corresponding author on reasonable request.