Long-Term Effects of Resisted Sled Sprint Training on Acceleration Performance in Female Professional Soccer Players

Featured Application

Resisted sled sprint training performed once per week improves acceleration performance and sprint mechanics in professional female soccer players, offering an efficient training strategy that can be safely integrated into competitive-season workloads.

Abstract

Resisted sprint training (RST) is widely used to enhance acceleration capacity; however, evidence concerning on long-term effects to RST in professional women’s remains limited. Methods: This study examined the chronic effects of a six-week resisted sled sprint training intervention using a single-group longitudinal pilot design in professional female soccer players. Twenty-two players were assessed at baseline (T1), with fourteen completing the post-intervention assessment (T2) and seven available at the two-month follow-up (T3). Athletes completed one weekly RST session with loading progressively increasing from 20% to 80% of body mass and total sprint volume ranging from 100 to 200 m per session. Sprint performance and kinematic variables of the first three acceleration steps for both limbs were assessed before the intervention, immediately after and at a two-month follow-up. Within-group changes across time were analyzed using a one-way ANOVA with Bonferroni post hoc comparisons, with the level of significance set at p ≤ 0.05. Results: Sprint times significantly improved following the intervention (T1-T2), with a 2.61% improvement in acceleration performance over the 0-20 m phase. This improvement was accompanied by increases in center of mass projection angle and toe-off distance, resulting in a more forward-oriented sprint posture. At follow-up, sprint performance showed partial retention of these changes. These adaptations were accompanied by greater hip and knee extension of the ipsilateral limb at toe-off, without evidence of adverse sagittal-plane kinematic alterations during the early acceleration phase. Conclusions: The results indicate that once-weekly RST was associated with preliminary improvements in acceleration performance in professional female soccer players and induces technical adaptations that did not appear to negatively affect sprint mechanics during the initial acceleration phase. Given the absence of a control group and the substantial attrition at follow-up, these findings should be interpreted as exploratory.

1. Introduction

Women’s soccer has experienced substantial growth worldwide [1]. Increases in professional leagues, improved player conditions, better infrastructure, sponsorships, and audience interest reflect the sport’s rising professionalism and popularity. This progress has translated into higher-paced gameplay and an elevated level of high-speed running by teams in international competitions, irrespective of ball possession. High-intensity running distance (>23 km/h) increased by 29% between the 2015 and 2019 Women’s World Cups [1].

Although high-speed actions are decisive in soccer [2], athletes in most sports, rarely attain maximum speed. Therefore, acceleration is a critical determinant of performance, particularly in soccer [3]. Soccer players perform frequent linear acceleration of varying durations, starting from different levels of initial velocity, and even after changes of direction [4].

The determinants of acceleration are complex and diverse [5]. During the acceleration phase, the total force is exerted in a direction that is not aligned with displacement, due in part to body positioning and gravitational constraints [3]. Thus, the initial steps of a sprint, where acceleration of body mass (BM) and forward velocity production occur in an unfavorable mechanical stance, require a substantial amount of horizontal force (FH) [3].

Accordingly, athletes’ technical ability to apply force effectively is a primary determinant of acceleration performance [3]. The ability to generate and apply optimally oriented forward force is crucial for peak performance [6]. In addition to the technical capacity to apply force, athletes’ specific neuromuscular capability plays a significant role. Musculotendinous mechanical properties, joint kinematic angles, and muscle groups such as hip extensors may facilitate better alignment of the resultant force [7].

Resisted sprint training (RST) has therefore been used as an overload method to enhance sprint acceleration capacity [8,9]. Performing sprint training under specific external load conditions may be an optimal training approach, as it can transfer to specific adaptations in unloaded sprinting [10]. Furthermore, evidence suggests that the acceleration phase can show positive adaptations with individually prescribed loads based on the force-velocity profile [11].

In professional male soccer players, very heavy sled sprint training has been shown to improve speed performance, mechanical effectiveness, and maximal horizontal force capability [9]. This training approach involves orienting the force horizontally, necessitating a slower speed and a greater inclination, a mechanical state unattainable in unloaded sprinting [9].

The impact of external load on acceleration phase kinematics in team sports is thus a topic of current interest in the literature. Evidence of potential alterations in sprinting technique comes mainly from studies evaluating the acute effects of RST [12] (immediate effects during training), these acute investigations consistently report that increasing external loads can amplify kinematic and mechanical differences between loaded and unloaded sprinting, which have been interpreted as negative technical adaptations. Such alterations include increases in sprint time, less effective force application patterns, changes in body posture such as greater forward trunk inclination, and modifications in hip, knee, and ankle kinematics that may lead to suboptimal sprint mechanics. Additionally, heavy external loads have been shown to affect load distribution during ground contact, potentially compromising the execution of sprinting movements [13,14,15,16,17,18].

Based on these acute observations, current guidelines often recommend caution when prescribing higher resisted sprint loads, and light loading strategies have therefore been most commonly adopted in research, as they are considered to minimally disrupt acceleration kinematics relative to free sprinting [13,14,15,16,17,18]. However, current guidelines draw primarily on acute experimental models and may not fully predict the long-term physiological or motor adaptations resulting from a progressive and sustained training program. In contrast, studies evaluating the chronic effect of resisted sprint acceleration training (long-term adaptations) on kinematics remain limited [19,20], particularly in elite team-sport settings. Moreover, all kinematic analyses of acceleration in team sports have been conducted exclusively with male participants [20,21].

Therefore, despite the continuous growth of women’s soccer, technological advances, and recent literature interest in sprint-training research, to date, no study has comprehensively assessed the chronic effect of progressive resisted sled sprint training across a broad loading spectrum while performing a detailed kinematic analysis in elite female athletes.

For this reason, this study analyzed the long-term effects of loaded RST on acceleration performance and kinematics in professional women’s soccer. We hypothesized that (1) chronic $\mathrm{RST}$ exposure would improve acceleration capacity in professional female soccer players, and (2) the six-week progressive RST program would not induce adverse alterations in sprint kinematics during the early acceleration phase (first three steps), specifically no reduction in step length, no increase in forward trunk inclination, and no changes in hip, thigh, knee, shank, ankle, or foot sagittal-plane angles that would move the players away from a conventional unloaded sprint technique.

2. Materials and Methods

2.1. Experimental Approach to the Problem

This study employed a single-group longitudinal pilot design to explore changes in sprint acceleration performance and associated kinematic variables following a resisted sled sprint training intervention in professional female soccer players. Because the players had no prior exposure to resisted sled sprinting, we implemented a two-week familiarization period preceding baseline testing to standardize execution and minimize potential learning effects.

Player were assessed at three time points during the competitive season: an initial assessment (T1) performed during one dedicated week after the familiarization period; a post-test at eight weeks (T2), conducted during the week following six consecutive weeks of training; and a third measurement (T3) taken eight weeks following T2 to assess short-term retention or decay of observed adaptations, during which time no specific intervention was performed.

All twenty-two professional female soccer players completed the familiarization period and baseline assessment (T1). Fourteen completed the intervention. Only players who completed all six training sessions were included in the intervention analyses, resulting in a training adherence of 100% and ensuring full exposure to the prescribed training stimulus. Owing to practical constraints inherent to professional soccer, including national team commitments, injury or illness, only seven players were able to undergo the final measurements (T3).

2.2. Subjects

Twenty-two female first-division soccer players (mean ± standard deviation: age: 24 ± 4.44 years, BM: 60.6 ± 5.62 kg, height: 166 ± 5.58 cm) participated in the study. All participants belonged to a first-division women’s soccer team in Spain. They had previous strength training experience ranging in duration from three to five years. The players were informed of the benefits and risks of the investigation before providing written informed consent to participate voluntarily. The research protocol was approved by the Autonomous University of Madrid Ethics Committee in accordance with the Declaration of Helsinki.

2.3. Procedures

All sessions began with 8 min of targeted mobility exercises, dynamic stretching, three acceleration exercises and three speed exercises. After warm-up, all participants performed one maximal 20 m sprint, 5 min before testing. The protocol consisted of two maximal sprints of 20 m with a 5 min rest between repetitions. Tests were conducted at the team’s usual training location on artificial turf. All participants completed the test on the same morning, after a rest day, and under consistent weather conditions.

Inclusion criteria required participants to be professional soccer players competing in the Spanish first division of women’s soccer; hence both goalkeepers and outfield players were included in the study. All players followed the same RST protocol regardless of playing position. The players’ overall routine included four to five weekly training sessions (including strength work preceding each field session) and one competitive match day.

2.4. Intervention Design

The intervention design consisted of six sessions of RST (

Table 1. Training design.

Session/Week	Load (%BM)			Volume (m)
1	3 × 20 m 20%	+	2 × 20 m 25%	100
2	4 × 20 m 30%	+	2 × 20 m 25%	120
3	5 × 20 m 40%	+	2 × 20 m 50%	140
4	6 × 20 m 60%	+	2 × 20 m 50%	160
5	7 × 20 m 70%	+	2 × 20 m 75%	180
6	8 × 20 m 80%	+	2 × 20 m 75%	200

Abbreviations. BM: body mass; m: meters. Load (%BM): two resisted sprint blocks per session, each performed with a different external load (%BM); repetitions × distance are specified for each load.

). Due to the sample characteristics of professional players without prior experience in loaded sprint training, a two-week familiarization period was completed before baseline testing to allow progressive exposure to resisted sprinting. During the first familiarization week, players performed two resisted sprints at 25% and two at 50% of body mass (BM). During the second week, the familiarization progressed to one sprint at 25% BM, one at 50% BM, and two at 75% BM. In addition, during the weeks in which baseline T1 and T2 assessments were conducted, players performed three resisted sprint efforts at 25%, 50%, and 75% BM following the unloaded sprint test. The intervention was designed to ensure a safe and gradual adaptation to high load stimuli.

The intervention was performed once per week. Load progressed from 20% to 80% of each player’s BM, and total sprint volume increased from 100 to 200 m across the training period. The initial number of sprints was five and increased to ten at the end of the six weeks of intervention. Implementation used six sleds. All players used a belt and a galvanized drag sled measuring 48 × 30 cm with a mass of 2.6 kg (Ranking, Madrid, Spain).

The training sessions were organized in small groups using six sleds, allowing players to rotate between sprint execution and sled retrieval, which facilitated a fluid, time-efficient, and competitive training dynamic in high-performance environments while maintaining the prescribed rest intervals between repetitions.

During the RST intervention, a standardized 3 min rest interval was provided between repetitions to minimize the influence of fatigue and ensure consistent execution of the training stimulus.

Changes in performance were assessed using kinematic and spatiotemporal variables and sprint times.

2.5. Kinematic Variables

This study focused on the early acceleration phase. The first three steps were analyzed, corresponding to the first 5 m of each sprint. Each step included two key events: toe-off (TO), defined as the last frame of ground contact during take-off, phase when the foot was in contact with the ground; and touchdown (TD), defined as the first frame in which the foot contacted the ground. Each moment included both the ipsilateral (IPSI) and contralateral (CONTRA) legs. The first TO and TD were identified as step 1, followed by step 2 and step 3 (Figure 1).

All three evaluations were video recorded. Video images were obtained at 240 Hz using a smartphone (iPhone 11 Pro) with a resolution of 1080p HD. The filming area covered a horizontal distance of 5 m along the midpoint of the camera arrangement. Using the starting line as reference point along the 0-20 m sprint line, the camera was positioned 1.5 m from the 0 m mark and perpendicularly 9 m from the 1.5 m mark. The camera was mounted on a tripod at a height of 1.20 m, with the optical axis oriented perpendicular to the direction of running. This sagittal-plane video-based approach has previously been implemented and validated in professional soccer environments to assess sprint running mechanics related to both performance and injury risk, as well as to examine sprint mechanical alterations under fatigue conditions, supporting its ecological validity for applied field-based assessments [19,22,23,24,25].

The main joint segments were identified using reflective markers placed at eight landmarks on the right side of the body. Image digitization was carried out manually following the 18-point model: vertex of the head, midpoint between suprasternal notch and the 7th cervical vertebra, center of shoulder, elbow, wrist, knee and ankle joint centers, hip landmark, head of the third metacarpal and the tip of the toe (Figure 2).

Kinematic variables of interest were determined from video frames identified as TO and TD, using the zoom tool at 600% in Kinovea (v.0.9.5). To minimize digitization variability, each trial was digitized two times by the same observer and the mean values were used for subsequent analyses.

Digitized coordinates were exported to Excel (Microsoft Office 2022), where a full kinematic analysis of the lower limb was conducted: hip, thigh, knee, shank, ankle, and foot angles, in addition to analysis of the trunk angle, knee separation distance, and center-of-mass (CoM) angle, defined as the angle between the athlete’s CoM and the ground contact point [26]. CoM location was calculated according to De Leva P. [27]. The calculation of the angles was performed with respect to the horizontal (Figure 2).

Conversely, the spatiotemporal variables selected were as follows: contact time (s) was calculated by subtracting the touchdown time from the subsequent toe-off time [28]. Flight time (s) was calculated by subtracting the toe-off time from the subsequent touchdown time [28]. Step length (m) was the horizontal displacement of the CoM between two consecutive TDs [28]. Step frequency (Hz) was calculated by dividing 1 by step time and toe-off distance (m) was the anteroposterior distance between the CoM at toe-off and the average toe position during contact [28]. All CoM distances were normalized with respect to each player’s anthropometry. This investigation used the same procedure as previous studies [19,21]. These kinematic and spatiotemporal variables were selected as explanatory descriptors of sprint acceleration mechanics and were not considered direct measures of sprint performance, which was primarily assessed through sprint time outcomes.

2.6. Performance Variables

Photoelectric cells (Microgate, Bolzano, Italy) were placed at 0 m and 20 m to record 20 m sprint time. We selected the distance of 20 m for the test, with the objective of knowing the sprint times (performance variables).

2.7. Statistical Analyses

Statistical analyses were performed to examine within-subject changes across the different assessment time points in this single-group longitudinal pilot study. Data normality was assessed using the Kolmogorov-Smirnov test, and homogeneity of variances was examined with Levene’s test.

For variables meeting parametric assumptions, within-subject comparisons across time were conducted using one-way analysis of variance (ANOVA) with Bonferroni-adjusted post hoc comparisons. When normality assumptions were not met, equivalent non-parametric procedures were applied. Given the exploratory nature of the study and the limited sample size, analyses were conducted independently for each outcome variable to avoid overfitting and complex interaction effects.

Descriptive data are presented as mean ± standard deviation. Statistical significance was set at p < 0.05.

In addition, percentage change was calculated to describe performance adaptations using the following formula: [(final value − initial value)/initial value] × 100 [29]. Percentage change values were used to provide a practical representation of the magnitude of within-subject performance changes over time.

3. Results

The analysis of variance indicated a significant T1-T2 improvement in 0-20 m split time (p < 0.047) [ES: −0.5 (−1.12. −0.03)]. Significant differences were found in TO for the CoM angle (p < 0.001) at step 1 [T1 vs. T2: ES: 4.61 (3.66, 5.57); T1 vs. T3: 3.72 (2.81, 4.64)], step 2, [T1 vs. T2: ES: 4.97 (3.96, 5.98); T1 vs. T3: 4.48 (3.47, 5.49)] and step 3 [T1 vs. T2: 4.53 (3.58, 5.47); T1 vs. T3: 3.44 (2.56, 4.32)] (Figure 1). These results are presented in

Table 2. Results for kinematic variables: step 1.

				Test					p-Value			Cohen’s d [95% CI]
			n	T1 [SD]	n	T2 [SD]	n	T3 [SD]	1 vs. 2	1 vs. 3	2 vs. 3	1 vs. 2	1 vs. 3	2 vs. 3	F	η2
CM- ANGLE	TO	IPSI	22	53.39 ± 2.13	14	44.39 ± 1.64	7	46.12 ± 2.13	<0.001 ***	<0.001 ***	−1.73 *	4.61 [3.66, 5.57] β	3.72 [2.81, 4.64] β	−0.88 [−1.56, −0.21]	F[2.67] = 160	0.827
		CONTRA
	TD	IPSI	22	94.8 ± 2.99	14	92.4 ± 6.41	7	93.7 ± 4.43	0.196	0.697	0.732	0.48 [−0.05, 1.02] Δ	0.21 [−0.43, 0.87] Δ	0.26 [−0.91, 0.39] Δ	F[2.67] = 1.62	0.046
		CONTRA
TRUNK	TO	IPSI	22	47.42 ± 4.58	14	46.43 ± 4.01	7	45.47 ± 3.67	0.656	0.337	0.763	0.23 [−0.3, 0.76] Δ	0.46 [−0.19, 1.12] Δ	0.23 [−0.42, 0.88] Δ	F[2.67] = 1.06	0.031
		CONTRA	22	47.4 ± 4.58	14	46.4 ± 4.01	7	45.5 ± 3.67	0.656	0.337	0.763	0.23 [−0.3, 0.76] Δ	0.46 [−0.19, 1.23] Δ	0.23 [−0.4, 0.88] Δ	F[2.67] = 1.06	0.031
	TD	IPSI	22	48.7 ± 4.88	14	47.4 ± 4.09	7	48.5 ± 4.36	0.533	0.99	0.74	0.28 [−0.24, 0.82] Δ	0.04 [−0.6, 0.69]	−0.24 [−0.89, 0.41]	F[2.67] = 0.63	0.019
		CONTRA	22	48.71 ± 4.88	14	47.42 ± 4.09	7	48.51 ± 4.36	0.533	0.99	0.74	0.28 [−0.24, 0.82] Δ	0.04 [−0.6, 0.69]	−0.24 [−0.89, 0.41]	F[2.67] = 0.63	0.019
HIP	TO	IPSI	22	180.93 ± 6.98	14	185.31 ± 7.3	7	180.03 ± 7.5	0.067	0.924	0.073	−0.6 [−1.15, −0.06]	0.12 [−0.52, 0.77]	0.73 [0.06, 1.39] µ	F[2.67] = 3.6	0.097
		CONTRA	22	80.2 ± 11.95	14	76.4 ± 11.72	7	73.3 ± 7.49	0.396	0.144	0.679	0.34 [−0.18, 0.88]Δ	0.62 [−0.03, 1.28] µ	0.27 [−0.37, 0.93] Δ	F[2.67] = 1.99	0.056
	TD	IPSI	22	93.3 ± 13.31	14	99.1 ± 12.93	7	95.2 ± 12.53	0.227	0.897	0.634	−0.44 [−0.98, 0.09]	−0.14 [−0.79, 0.5]	0.29 [−0.35, 0.95] Δ	F[2.67] = 1.41	0.041
		CONTRA	22	170.03 ± 8.12	14	167.56 ± 13.04	7	169.2 ± 10.53	0.67	0.97	0.888	0.22 [−0.3, 0.76] Δ	0.07 [−0.57, 0.73]	−0.15 [−0.8, 0.5]	F[2.67] = 0.37	0.011
THIGH	TO	IPSI	22	46.48 ± 4.68	14	41.12 ± 6.56	7	45.43 ± 5.66	0.003 **	0.822	0.087	0.94 [0.38, 1.49] µ	0.18 [−0.47, 0.83]	−0.75 [−1.42, −0.09]	F[2.67] = 6.66	0.166
		CONTRA	22	147.2 ± 9.33	14	150.1 ± 10.38	7	152.2 ± 5.29	0.518	0.081	0.663	−0.31 [−0.85, 0.21]	−0.54 [−1.2, 0.11]	−0.22 [−0.88, 0.42]	F[2.67] = 1.54	0.044
	TD	IPSI	22	135.4 ± 10.81	14	128.3 ± 12.07	7	133.3 ± 10.01	0.054	0.836	0.369	0.63 [0.08, 1.17] µ	0.18 [−0.46, 0.84]	−0.44 [−1.1, 0.21]	F[2.67] = 2.88	0.079
		CONTRA	22	58.68 ± 7.14	14	59.86 ± 11.88	7	59.31 ± 7.57	0.886	0.978	0.982	−0.12 [−0.65, 0.4]	−0.06 [−0.72, 0.58]	0.05 [−0.59, 0.71]	F[2.67] = 0.11	0.003
KNEE	TO	IPSI	22	171.14 ± 5.8	14	177.25 ± 8.75	7	173.32 ± 6.39	0.01 **	0.54	0.239	−0.84 [−1.39, −0.29]	0.3 [−0.95, 0.35]	0.54 [−0.11, 1.2] Δ	F[2.67] = 5.06	0.131
		CONTRA	22	92.9 ± 11.38	14	90.3 ± 11.26	7	87.5 ± 6.96	0.622	0.27	0.704	0.24 [−0.28, 0.78] Δ	0.51 [−0.14, 1.17] Δ	0.26 [−0.39, 0.91] Δ	F[2.67] = 1.27	0.037
	TD	IPSI	22	107.2 ± 11.08	14	112.4 ± 9.32	7	108.9 ± 10.36	0.149	0.878	0.543	−0.5 [−1.04, 0.03]	−0.15 [−0.81, 0.49]	0.34 [−0.3, 1] Δ	F[2.67] = 1.84	0.052
		CONTRA	22	138.75 ± 13.58	14	131.25 ± 20.25	7	129.34 ± 13.01	0.213	0.197	0.933	0.45 [−0.08, 0.99] Δ	0.57 [−0.09, 1.23] Δ	0.11 [−0.53, 0.76]	F[2.67] = 2.1	0.059
SHANK	TO	IPSI	22	37.63 ± 3.13	14	38.37 ± 3.75	7	38.76 ± 1.77	0.704	0.31	0.89	−0.23 [−0.76, 0.3]	−0.35 [−1, 0.3]	−0.12 [−0.77, 0.53]	F[2.67] = 0.68	0.02
		CONTRA	22	60.1 ± 5.98	14	60.4 ± 7.69	7	59.7 ± 7.82	0.99	0.982	0.954	−0.03 [−0.57, 0.49]	0.05 [−0.59, 0.71]	0.09 [−0.55, 0.75]	F[2.67] = 0.04	0.001
	TD	IPSI	22	62.7 ± 5.25	14	60.8 ± 6.25	7	62.2 ± 4.61	0.423	0.966	0.718	0.33 [−0.2, 0.87] Δ	0.08 [−0.57, 0.73]	−0.25 [−0.9, 0.4]	F[2.67] = 0.83	0.024
		CONTRA	22	17.43 ± 9.12	14	11.12 ± 11.75	7	8.65 ± 9.22	0.063	0.029 *	0.744	0.61 [0.07, 1.15] µ	0.85 [0.18, 1.52] µ	0.24 [−0.41, 0.89] Δ	F[2.67] = 4.33	0.114
ANKLE	TO	IPSI	22	139.03 ± 6.59	14	139.87 ± 7.52	7	137.7 ± 7.26	0.899	0.834	0.621	−0.11 [−0.65, 0.41]	0.18 [−0.46, 0.84]	0.3 [−0.35, 0.96] Δ	F[2.67] = 0.43	0.013
		CONTRA	22	90.2 ± 6.08	14	91.2 ± 5.69	7	92.3 ± 4.26	0.781	0.495	0.826	−0.17 [−0.71, 0.35]	−0.37 [−1.02, 0.28]	−0.19 [−0.84, 0.46]	F[2.67] = 0.67	0.02
	TD	IPSI	22	94.8 ± 6.96	14	97.2 ± 7.61	7	96.3 ± 6.94	0.427	0.798	0.923	−0.33 [−0.87, 0.2]	−0.21 [−0.86, 0.44]	0.12 [−0.52, 0.77]	F[2.67] = 0.79	0.023
		CONTRA	22	135.57 ± 5.84	14	134.29 ± 8.62	7	134.84 ± 7.55	0.796	0.951	0.972	0.17 [−0.36, 0.7]	0.09 [−0.55, 0.75]	−0.07 [−0.72, 0.57]	F[2.67] = 0.2	0.006
FOOT	TO	IPSI	22	78.59 ± 6.91	14	78.5 ± 7.03	7	81.06 ± 7.55	0.996	0.783	0.719
		CONTRA	22	149.9 ± 8.77	14	149.2 ± 9	7	147.4 ± 6.57	0.942	0.64	0.802	0.08 [−0.44, 0.62]	0.29 [−0.36, 0.95] Δ	0.2 [−0.44, 0.86] Δ	F[2.67] = 0.4	0.012
	TD	IPSI	22	147.9 ± 8.49	14	143.6 ± 9.65	7	145.9 ± 8.24	0.178	0.778	0.709	0.48 [−0.05, 1.02] Δ	0.22 [−0.43, 0.87] Δ	−0.25 [−0.91, 0.39]	F[2.67] = 1.62	0.046
		CONTRA	22	61.87 ± 9.82	14	56.83 ± 15.55	7	53.81 ± 12.07	0.312	0.141	0.753	0.39 [−0.14, 0.93] Δ	0.62 [−0.03, 1.29] µ	0.23 [−0.41, 0.89] Δ	F[2.67] = 2.12	0.06

Note: TO: toe-off; TD: touchdown; IPSI: ipsilateral; CONTRA: contralateral; SD: standard deviation; CI: confidence interval. Significant post-hoc [p < 0.05] = *; [p < 0.01] = **; [p < 0.001] = ***; Effect size [Small: 0.2–0.59 = ^Δ; Moderate: 0.60–1.19 = µ; Large: >1.19 = β].

and

Table 3. Results for kinematic variables: step 2.

				Test					p-Value			D‘Cohen’s d [95% CI]
			n	T1 [SD]	n	T2 [SD]	n	T3 [SD]	1 vs. 2	1 vs. 3	2 vs. 3	1 vs. 2	1 vs. 3	2 vs. 3	F	η2
CM- ANGLE	TO	IPSI	22	57.57 ± 2.27	14	46.6 ± 2.05	7	47.69 ± 2.33	<0.001 ***	<0.001 ***	0.295	4.97 [3.96, 5.98] β	4.48 [3.47, 5.49] β	−0.49 [−1.15, 0.16]	F[2.67] = 196	0.854
		CONTRA
	TD	IPSI	22	96.9 ± 2.03	14	98.5 ± 3.49	7	97.3 ± 2.92	0.096	0.844	0.512	−0.56 [−1.1, −0.02]	−0.16 [−0.82, 0.48]	0.39 [−0.25, 1.05] Δ	F[2.67] = 2.32	0.065
		CONTRA
TRUNK	TO	IPSI	22	51.36 ± 5.58	14	51.09 ± 6.83	7	52.12 ± 4.88	0.985	0.921	0.921	0.04 [−0.48, 0.57]	−0.12 [−0.78, 0.52]	−0.17 [−0.82, 0.48]	F[2.67] = 0.13	0.004
		CONTRA	22	51.4 ± 5.58	14	51.1 ± 6.83	7	52.1 ± 4.89	0.985	0.921	0.861	0.04 [−0.48, 0.57]	−0.12 [−0.78, 0.52]	−0.17 [−0.82, 0.48]	F[2.67] = 0.13	0.004
	TD	IPSI	22	52 ± 5.32	14	53.4 ± 5.72	7	54.6 ± 3.87	0.559	0.274	0.76	−0.27 [−0.81, 0.26]	−0.5 [−1.16, 0.15]	−0.23 [−0.88, 0.42]	F[2.67] = 1.3	0.037
		CONTRA	22	51.97 ± 5.32	14	53.41 ± 5.72	7	54.63 ± 3.87	0.559	0.274	0.76	−0.27 [−0.81, 0.26]	−0.5 [−1.16, 0.15]	−0.23 [−0.88, 0.42]	F[2.67] = 1.3	0.037
HIP	TO	IPSI	22	180.98 ± 8.19	14	183.64 ± 11.21	7	183.34 ± 9.45	0.566	0.742	0.995	−0.27 [−0.8, 0.26]	−0.24 [−0.89, 0.41]	0.03 [−0.62, 0.68]	F[2.67] = 0.58	0.017
		CONTRA	22	82 ± 13.02	14	76.2 ± 15.42	7	82.5 ± 10.44	0.262	0.992	0.341	0.42 [−0.11, 0.96] Δ	−0.04 [−0.69, 0.61]	−0.46 [−1.12, 0.19]	F[2.67] = 1.6	0.046
	TD	IPSI	22	99.4 ± 12.06	14	102.3 ± 13.42	7	104.9 ± 7.16	0.666	0.163	0.694	−0.24 [−0.78, 0.28]	−0.46 [−1.12, 0.19]	−0.21 [−0.87, 0.43]	F[2.67] = 1.09	0.031
		CONTRA	22	163.39 ± 7.8	14	166.06 ± 8.99	7	167.31 ± 8.42	0.465	0.335	0.893	−0.31 [−0.85, 0.21]	−0.46 [−1.12, 0.19]	−0.14 [−0.8, 0.5]	F[2.67] = 1.23	0.035
THIGH	TO	IPSI	22	50.37 ± 4.61	14	47.45 ± 5.78	7	48.78 ± 6.36	0.921	0.647	0.739	0.53 [−0, 1.07] Δ	0.29 [−0.36, 0.94] Δ	−0.24 [−0.89, 0.41]	F[2.67] = 2.01	0.057
		CONTRA	22	149.4 ± 8.81	14	154.9 ± 10.57	7	149.6 ± 6.95	0.07	0.99	0.19	−0.59 [−1.13, −0.04]	−0.02 [−0.67, 0.63]	0.56 [−0.09, 1.22]Δ	F[2.67] = 2.87	0.079
	TD	IPSI	22	132.6 ± 8.41	14	131.1 ± 10.11	7	129.7 ± 4.26	0.82	0.315	0.808	0.17 [−0.36, 0.7]	0.33 [−0.31, 0.99] Δ	0.16 [−0.49, 0.81]	F[2.67] = 0.56	0.016
		CONTRA	22	68.58 ± 8.03	14	67.35 ± 6.5	7	67.32 ± 8.37	0.816	0.866	1	0.16 [−0.37, 0.69]	0.16 [−0.48, 0.82]	0 [−0.64, 0.65]	F[2.67] = 0.22	0.007
KNEE	TO	IPSI	22	167.45 ± 5.74	14	170.89 ± 6.53	7	170.38 ± 6.8	0.109	0.335	0.966	−0.54 [−1.08, −0]	−0.46 [−1.12, 0.19]	0.08 [−0.57, 0.73]	F[2.67] = 2.3	0.064
		CONTRA	22	96 ± 13.27	14	91.1 ± 14.1	7	94.2 ± 10.26	0.353	0.911	0.751	0.37 [−0.16, 0.9]Δ	0.13 [−0.51, 0.78]	−0.23 [−0.89, 0.41]	F[2.67] = 0.97	0.028
	TD	IPSI	22	117.4 ± 8.69	14	119.2 ± 9.69	7	119.3 ± 2.32	0.762	0.532	0.996	−0.2 [−0.74, 0.32]	−0.22 [−0.88, 0.42]	−0.01 [−0.67, 0.63]	F[2.67] = 0.39	0.011
		CONTRA	22	117.4 ± 16.54	14	116.75 ± 11.14	7	114.48 ± 9.63	0.984	0.752	0.775	0.04 [−0.48, 0.58]	0.21 [−0.43, 0.87]Δ	0.17 [−0.48, 0.82]	F[2.67] = 0.22	0.007
SHANK	TO	IPSI	22	37.83 ± 2.22	14	38.34 ± 2.42	7	39.16 ± 1.97	0.675	0.18	0.518	−0.22 [−0.76, 0.3]	−0.58 [−1.24, 0.07]	−0.36 [−1.01, 0.29]	F[2.67] = 1.61	0.046
		CONTRA	22	65.4 ± 8.51	14	66 ± 8.51	7	63.8 ± 8.62	0.96	0.845	0.72	−0.07 [−0.6, 0.46]	0.18 [−0.47, 0.83]	0.25 [−0.4, 0.9]Δ	F[2.67] = 0.3	0.009
	TD	IPSI	22	70 ± 4.53	14	70.2 ± 4.92	7	69 ± 4.19	0.979	0.795	0.7	−0.05 [−0.58, 0.48]	0.21 [−0.44, 0.86] Δ	0.26 [−0.39, 0.91] Δ	F[2.67] = 0.33	0.01
		CONTRA	22	5.98 ± 10.71	14	4.11 ± 7.21	7	1.8 ± 4.74	0.687	0.293	0.684	0.22 [−0.31, 0.75] Δ	0.49 [−0.16, 1.15] Δ	0.27 [−0.38, 0.92] Δ	F[2.67] = 1.16	0.034
ANKLE	TO	IPSI	22	138.94 ± 6.25	14	140 ± 5.14	7	139.12 ± 3.22	0.74	0.994	0.871	−0.19 [−0.73, 0.33]	−0.03 [−0.68, 0.61]	0.16 [−0.49, 0.81]	F[2.67] = 0.29	0.009
		CONTRA	22	93.5 ± 6.37	14	94.7 ± 5.85	7	97.2 ± 4.41	0.725	0.13	0.381	−0.2 [−0.73, 0.33]	−0.64 [−1.3, 0.02]	−0.43 [−1.09, 0.22]	F[2.67] = 1.92	0.054
	TD	IPSI	22	98.2 ± 6.39	14	100.8 ± 6.5	7	100.5 ± 5.12	0.268	0.474	0.994	−0.41 [−0.95, 0.12]	−0.38 [−1.04, 0.27]	0.03 [−0.61, 0.68]	F[2.67] = 0.04	0.04
		CONTRA	22	135.05 ± 8.47	14	134.28 ± 8.04	7	137.43 ± 5.21	0.927	0.62	0.435	0.09 [−0.43, 0.63]	−0.3 [−0.96, 0.34]	−0.4 [−1.06, 0.25]	F[2.67] = 0.77	0.023
FOOT	TO	IPSI	22	78.89 ± 6.31	14	78.34 ± 4.87	7	80.03 ± 2.67	0.93	0.691	0.327	0.1 [−0.42, 0.63]	−0.22 [−0.87, 0.43]	−0.32 [−0.98, 0.33]	F[2.67] = 0.49	0.015
		CONTRA	22	151.9 ± 8.7	14	151.3 ± 8.24	7	146.6 ± 6.87	0.963	0.128	0.192	0.06 [−0.46, 0.6]	0.64 [−0.01, 1.3] µ	0.57 [−0.08, 1.23] Δ	F[2.67] = 2.12	0.059
	TD	IPSI	22	151.8 ± 7.1	14	149.5 ± 6.54	7	148.5 ± 4.35	0.361	0.253	0.881	0.36 [−0.17, 0.9] Δ	0.52 [−0.13, 1.18] Δ	0.15 [−0.49, 0.81]	F[2.67] = 1.59	0.045
		CONTRA	22	50.92 ± 11.97	14	49.82 ± 10.36	7	44.37 ± 5.26	0.916	0.135	0.246	0.1 [−0.42, 0.64]	0.63 [−0.02, 1.29] µ	0.52 [−0.13, 1.18] Δ	F[2.67] = 1.98	0.056

Note: TO: Toe-off; TD: touchdown; IPSI: ipsilateral; CONTRA: contralateral; SD: standard deviation; CI: confidence interval. Significant post-hoc [p < 0.001] = ***; Effect size [Small: 0.2–0.59 = ^Δ; Moderate: 0.60–1.19 = µ; Large: >1.19 = β].

. Similarly, toe-off distance differed significantly at TO in both analyzed steps: step 1 (p < 0.001) and step 2 (p < 0.001) (

Table 4. Results for kinematic variables: step 3.

				Test					p-Value			d‘Cohen’s d [95% CI]
			n	T1 [SD]	n	T2 [SD]	n	T3 [SD]	1 vs. 2	1 vs. 3	2 vs. 3	1 vs. 2	1 vs. 3	2 vs. 3	F	η2
CM-ANGLE	TO	IPSI	22	58.59 ± 1.91	14	48.74 ± 2.52	7	51.1 ± 1.88	<0.001 ***	<0.001 ***	0.004 **	4.53 [3.58, 5.47] β	3.44 [2.56, 4.32] β	−1.08 [−1.76, −0.40]	F[2.67] = 151	0.819
		CONTRA
TRUNK	TO	IPSI	22	56.12 ± 4.78	14	57.53 ± 5.13	7	57.3 ± 5.13	0.546	0.752	0.989	−0.28 [−0.81, 0.25]	−0.23 [−0.89, 0.41]	0.04 [−0.6, 0.69]	F[2.67] = 0.6	0.018
		CONTRA	22	56.1 ± 4.79	14	57.5 ± 5.13	7	57.3 ± 5.13	0.546	0.752	0.989	−0.28 [−0.81, 0.25]	−0.23 [−0.89, 0.41]	0.04 [−0.6, 0.69]	F[2.67] = 0.6	0.018
HIP	TO	IPSI	22	186.67 ± 7.25	14	193.32 ± 12	7	185.53 ± 11.67	0.041 *	0.94	0.12	−0.64 [−1.19, −0.1]	0.11 [−0.54, 0.76]	0.75 [0.09, 1.42] µ	F[2.67] = 3.98	0.106
		CONTRA	22	86.5 ± 9.88	14	87.3 ± 12.39	7	81.6 ± 11.51	0.965	0.375	0.273	−0.06 [−0.6, 0.46]	0.44 [−0.21, 1.09] Δ	0.50 [−0.15, 1.16] Δ	F[2.67] = 1.29	0.037
THIGH	TO	IPSI	22	49.45 ± 4.72	14	44.21 ± 7.54	7	51.77 ± 7.46	0.009 **	0.55	0.013 *	0.80 [0.25, 1.35] µ	−0.35 [−1.01, 0.3]	−1.15 [−1.84, −0.47]	F[2.67] = 7.7	0.187
		CONTRA	22	149.6 ± 7.31	14	150.3 ± 9.15	7	155.8 ± 8.1	0.955	0.067	0.111	−0.07 [−0.61, 0.45]	−0.74 [−1.41, −0.07]	−0.66 [−1.33, −0]	F[2.67] = 2.83	0.078
KNEE	TO	IPSI	22	169.98 ± 5.1	14	176.35 ± 8.07	7	167.99 ± 8.43	0.003 **	0.699	0.013 *	−0.89 [−1.45, −0.34]	0.28 [−0.37, 0.93] Δ	1.17 [0.49, 1.86] µ	F[2.67] = 8.59	0.204
		CONTRA	22	97.5 ± 11.07	14	95.8 ± 10.43	7	90.3 ± 10.31	0.832	0.105	0.256	0.15 [−0.37, 0.68]	0.67 [0.01, 1.33] µ	0.52 [−0.13, 1.18] Δ	F[2.67] = 2.17	0.061
SHANK	TO	IPSI	22	39.44 ± 2.42	14	40.57 ± 2.51	7	39.76 ± 2.08	0.193	0.914	0.563	−0.46 [−1, 0.07]	−0.13 [−0.78, 0.52]	0.33 [−0.32, 0.99] Δ	F[2.67] = 1.6	0.045
		CONTRA	22	67.1 ± 7.55	14	66.1 ± 9	7	66 ± 8.52	0.894	0.919	0.0618	0.12 [−0.41, 0.65]	0.12 [−0.52, 0.78]	0 [−0.64, 0.66]	F[2.67] = 0.12	0.004
ANKLE	TO	IPSI	22	139.55 ± 6.86	14	139.47 ± 6.08	7	137.1 ± 6.09	0.999	0.477	0.502	0.01 [−0.52, 0.54]	0.38 [−0.27, 1.03] Δ	0.36 [−0.28, 1.02] Δ	F[2.67] = 0.79	0.023
		CONTRA	22	93.9 ± 4.89	14	96.8 ± 5.33	7	95.8 ± 4.61	0.092	0.495	0.822	−0.56 [−1.11, −0.02]	−0.37 [−1.02, 0.28]	0.19 [−0.45, 0.85]	F[2.67] = 2.3	0.064
FOOT	TO	IPSI	22	79.88 ± 6.55	14	81.1 ± 5.97	7	82.65 ± 5	0.735	0.348	0.713	−0.2 [−0.73, 0.33]	−0.45 [−1.11, 0.2]	−0.25 [−0.91, 0.39]	F[2.67] = 0.99	0.029
		CONTRA	22	153.2 ± 8.2	14	149.3 ± 10.58	7	150.2 ± 8.15	0.268	0.596	0.95	0.41 [−0.12, 0.95] Δ	0.31 [−0.33, 0.97] Δ	−0.09 [−0.75, 0.55]	F[2.67] = 1.29	0.037
Results for spatiotemporal variables: step 1–2 and result for performance variable.
				Test					p-value			d‘Cohen’s d [95% CI]
			n	T1 [SD]	n	T2 [SD]	n	T3 [SD]	1 vs. 2	1 vs. 3	2 vs. 3	1 vs. 2	1 vs. 3	2 vs. 3	F	η2
TOE-OFF DISTANCE		Step 1	22	0.37 ± 0.01	14	0.48 ± 0.02	7	0.46 ± 0.02	<0.001 ***	<0.001 ***	0.05 *	−4.63 [−5.59, −3.67]	−3.73 [−4.65, −2.8]	0.89 [0.22, 1.57] µ	F[2.67] = 161	0.828
		Step 2	22	0.37 ± 0.02	14	0.46 ± 0.03	7	0.43 ± 0.02	<0.001 ***	<0.001 ***	0.02 *	−3.22 [−3.99, −2.45]	−2.31 [−3.08, −1.55]	0.9 [0.23, 1.58] µ	F[2.67] = 75.5	0.693
STEP VELOCITY		Step 1	22	3.51 ± 0.34	14	4.62 ± 0.56	7	4.16 ± 0.33	<0.001 ***	<0.001 ***	0.007 **	−2.51 [−3.19, −1.82]	−1.48 [−2.18, −0.77]	1.03 [0.35, 1.7] µ	F[2.67] = 44.4	0.57
		Step 2	22	4.16 ± 0.39	14	5.20 ± 0.44	7	4.92 ± 0.27	<0.001 ***	<0.001 ***	0.081	−2.64 [−3.34, −1.94]	−1.92 [−2.65, −1.19]	0.71 [0.05, 1.38] µ	F[2.67] = 51	0.6
CONTACT TIME		Step 1	22	0.21 ± 0.02	14	0.23 ± 0.17	7	0.2 ± 0.01	0.408	0.164	0.862
		Step 2	22	0.18 ± 0.01	14	0.17 ± 0.01	7	0.17 ± 0.01	0.025 *	0.014 *	0.751	0.71 [0.16, 1.26] µ	0.95 [0.27, 1.62] µ	0.23 [−0.41, 0.89] Δ	F[2.67] = 5.54	0.142
FLIGHT TIME		Step 1	22	0.05 ± 0.01	14	0.06 ± 0.01	7	0.07 ± 0	0.1	<0.001 ***	0.075	−0.55 [−1.1, −0.01]	−1.28 [−1.97, −0.59]	−0.72 [−1.39, −0.06]	F[2.67] = 7.84	0.19
		Step 2	22	0.06 ± 0.01	14	0.07 ± 0.01	7	0.07 ± 0	0.074	<0.001 ***	0.094	−0.59 [−1.14, −0.05]	−1.28 [−1.98, −0.59]	−0.69 [−1.36, −0.02]	F[2.67] = 7.96	0.192
STEP LENGHT		Step 1	22	0.93 ± 0.08	14	1.19 ± 0.1	7	1.13 ± 0.11	<0.001 ***	<0.001 ***	0.17	−2.56 [−3.25, −1.87]	−1.97 [−2.7, −1.23]	0.59 [−0.06, 1.26] Δ	F[2.67] = 48.8	0.593
		Step 2	22	1.02 ± 0.09	14	1.27 ± 0.12	7	1.23 ± 0.11	<0.001 ***	<0.001 ***	0.553	−2.18 [−2.83, −1.52]	−1.84 [−2.56, −1.11]	0.34 [−0.31, 0.99] Δ	F[2.67] = 36.5	0.521
STEP FREQUENCY		Step 1	22	3.79 ± 0.35	14	3.89 ± 0.49	7	3.7 ± 0.24	0.628	0.091	0.211	−0.26 [−0.8, 0.27]	0.22 [−0.42, 0.88] Δ	0.49 [−0.16, 1.15] Δ	F[2.67] = 1.21	0.035
		Step 2	22	4.06 ± 0.32	14	4.1 ± 0.27	7	4 ± 0.25	0.851	0.837	0.572	−0.14 [−0.67, 0.38]	0.18 [−0.46, 0.84]	0.33 [−0.32, 0.98] Δ	F[2.67] = 0.522	0.015
KNEE DISTANCE		Step 1	22	0.49 ± 0.04	14	0.65 ± 0.07	7	0.62 ± 0.04	<0.001 ***	<0.001 ***	0.304	−2.68 [−3.39, −1.98]	−2.2 [−2.95, −1.45]	0.48 [−0.17, 1.15] Δ	F[2.67] = 54.7	0.62
		Step 2	22	0.5 ± 0.03	14	0.63 ± 0.04	7	0.6 ± 0.04	<0.001 ***	<0.001 ***	0.079	−2.92 [−3.66, −2.19]	−2.2 [−2.96, −1.45]	0.71 [0.05, 1.38] µ	F[2.67] = 63.1	0.653
SPRINT TIME		0–20 m	22	3.44 ± 0.16	14	3.35 ± 0.17	7	3.45 ± 0.14	0.047 *	0.947	0.102	−0.5 [−1.12. −0.03] µ	0.1 [−0.55, 0.75]	0.68 [0.01, 1.34] µ	F[2.67] = 3.2	0.087

Note: TO: toe-off; IPSI: ipsilateral; CONTRA: contralateral; SD: standard deviation; CI: confidence interval. Significant post-hoc [p < 0.05] = *; [p < 0.01] = **; [p < 0.001] = ***; Effect size [Small: 0.2–0.59 = ^Δ; Moderate: 0.60–1.19 = µ; Large: >1.19 = β].

).

Trunk, ankle, and foot angles remained unchanged at both TO and TD on the IPSI and CONTRA sides. Thigh (p < 0.0029) (p < 0.001) and knee angles (p < 0.008) (p < 0.001) changed significantly at TO in the IPSI leg at steps 1 and 3, respectively, and hip angle (p < 0.048) [ES: −0.64 (−1.19, −0.1)] changed significantly at TO in the IPSI leg at step 3 (Table 2 and Table 4) (Figure 1).

At TD, the only significant angular change was observed in the CONTRA shank angle at step 1 (p < 0.029); no other TD variables changed significantly across the three steps. At TO, only the CONTRA hip angle at step 2 (p < 0.04) changed significantly (Table 2 and Table 4) (Figure 1).

For the spatiotemporal variables, step length (p < 0.001) and step velocity (p < 0.001) changed significantly from T1 to T2 in both steps 1 [ES: −2.56 (−3.25, −1.87)] and 2 [ES: −2.18 (−2.83, −1.52)] (Table 4). However, no differences were found in contact time, flight time, or step frequency for these steps from T1 to T2 (Table 4).

4. Discussion

The purpose of this study was to investigate the long-term effects of RST and to determine which kinematic variables could explain changes in acceleration performance. The main findings were that RST improved acceleration capacity and did not induce adverse alterations in early acceleration kinematics. Specifically, no reductions in step length, increases in forward trunk inclination, or sagittal-plane changes in hip, thigh, knee, shank, ankle, or foot angles were observed, supporting our hypotheses.

In the present study, sprint acceleration performance during the 0-20 m phase improved by 2.61%. Petrakos et al. [30] suggest that a 1% improvement in acceleration may improve a rugby or soccer player’s chances of reaching the ball before an opponent. These findings support the potential benefit of RST for soccer players.

The statistically significant changes observed in the CoM angle at TO during the first three steps following the intervention (Table 2, Table 3 and Table 4) were associated with improved mechanics during the first steps (Figure 1), resulting in a greater forward body orientation. Kugler et al. [6] established a relationship between faster athletes and a constant forward orientation during the acceleration phase. Furthermore, the progressive increase in values that occurred step by step aligns with Hans et al. [28] who explain that this is inevitable due to the increase in CoM-height (CoM-h), trunk angle, and the decrease in contact time that occur from the first step onwards. The follow-up assessment conducted two months after suggested partial retention of this the effect, although the T3 values indicated attenuation after RST was discontinued (Table 4).

The CoM angle depends on CoM-h and toe-off distance [28]. In the present study, toe-off distance increased significantly in the two analyzed steps. This technical characteristic is shared by both sprinters and rugby players [21]. This biomechanical adjustment that occurred after the training is related to increased performance in our study, as also observed by Kugler et al. [6]. Therefore, the strategy for placing the supporting foot during the delayed TO moment, increasing the distance from the CoM, implies a more forward-inclined orientation. Although our research did not evaluate FH, following Morin et al. [3] we established that this increased inclination may favor a greater application of FH, a fundamental aspect for improving acceleration. After concluding that “it seems that the importance is not so much the amount of total force produced, but the way it is oriented onto the supporting ground during the acceleration phase of the sprint”, Morin et al. [3] stated in 2011 the need for research to see whether “pushing forward to achieve greater distance” could be trained or improved, and what practical means or exercises could lead to it. Thus, the present findings suggest that RST may enhance propulsion by increasing the distance between the support foot and the CoM, thereby promoting a more effective TO during early acceleration in team sports. The follow-up data also suggested that the effect diminished after eight weeks without RST.

The toe-off distance depends in turn on the orientation of the segments of the supporting leg and the trunk [28]. Because thoracic and pelvic movements can occur independently, changes in toe-off distance may reflect altered pelvic positioning rather than trunk inclination alone [31]. Unlike studies that reported increased trunk inclination [20], our changes suggest an adaptation primarily driven by altered pelvic positioning and lower limb kinematics. In our case, the change in toe-off distance was related to a different positioning of the pelvis, with a greater hip extension and a different orientation of the lower limb segments.

After training, the stance foot was positioned farther from the CoM and was accompanied by greater triple extension at TO, reflected by increased hip, thigh, and knee extension (Figure 1). This result differs from studies that analyzed the acute effects of loaded sleds, which highlighted reduced knee and hip extension as the load increased [12].

At TD, significant differences were observed in the CONTRA shank angle in step 1; a smaller angle, indicating a more horizontal tibial orientation (Figure 1). It is possible that the new body arrangement caused a change in the coordination pattern, a more synchronous timing between the thigh and tibia, and as a result, the players exhibited a more compact recovery phase. This theory could be further corroborated through a more exhaustive analysis in future studies, following the theory and methodology outlined in Donaldson et al. [32].

In terms of spatiotemporal variables, we analyzed step frequency and step length, which are key parameters for describing sprint acceleration mechanics and understanding sprint technique [33]. We found a significant increase in step length, indicating a greater horizontal displacement of the CoM between two consecutive TDs in the steps analyzed [28]. In contrast, step frequency which is the difference between TD and subsequent TO times, [28] did not show any changes. Our findings align with Gleadhill et al. [34] who suggest that an increase in step length from the first step, coupled with an unchanged step frequency, may be indicative of improved acceleration performance in women. These findings highlight the need to interpret spatiotemporal sprint data in a sex-specific context in these variables, highlighting that while it is beneficial for female sprinters to focus on achieving a greater step length from the first support, the opposite is true for their male counterparts, where increasing this value could be associated with a decrease in performance. These results led to reflection on the interpretation of spatiotemporal data based on sex and underscored the need for further research in this regard.

Although few studies have evaluated contact-time changes after resisted sprint training with loads, the increase in contact time values in acute interventions seems consistent [12]. In our case, contact time did not change from T1 to T2 in steps 1 or 2 (Table 4). This may be related to a greater use of time to achieve a more forward position and a greater force production [6].

Our findings suggest that RST can be integrated into routine training in professional soccer settings. This study highlights that this type of training can be carried out in the context of professional soccer.

Once-weekly resisted sled sprint training, lasting approximately 20-30 min and progressing from 20% to 80% BM, improved acceleration performance in professional women’s professional players.

Consequently, this study supports incorporating resisted sprint loading prescriptions into periodized physical preparation model for elite soccer team. Our findings support the idea that strength and conditioning coaches need not be concerned about adverse effects on sprint technique.

The primary limitations of this study is the absence of an appropriate control group, which limits the ability to establish causal relationships between the intervention and the observed adaptations.

It should be noted that the results were obtained from a 2D analysis and a study focused only on the initial three steps. Although this approach allows the assessment of sprint kinematics in an applied and ecologically valid field-based context, the use of 2D analysis may limit the detection of out-of-plane movements and the interpretation of small kinematic changes.

We agree with Osterwald et al. [12] that changes may be present throughout the entire phase. Therefore, it would be beneficial for future studies to conduct a more comprehensive analysis, encompassing angular velocity and avoiding load standardization based on a percentage of the one repetition maximum.

Another notable limitation is the small sample size. We cannot determine whether the patterns observed in T3 would be replicated in other contexts, as the statistical power decreased due to a reduced sample size. Therefore, while we observed a clear tendency toward attenuation of performance following $\mathrm{RST}$ cessation (T3) these directional trends must be interpreted cautiously due to the small final sample size.

In addition, although digitization variability was minimized by repeating the digitization procedure and using mean values, no formal intra-rater or inter-rater reliability analysis was performed, and the kinematic findings should therefore be interpreted as exploratory.

5. Conclusions

A six-week progressive RST intervention was associated with improvements in acceleration performance in professional female soccer players, without inducing adverse alterations in sprint kinematics during the early acceleration phase. Specifically, the training program did not result in reductions in step length, increases in forward trunk inclination, or changes in hip, thigh, knee, shank, ankle, or foot sagittal-plane angles that would indicate a deviation from a conventional unloaded sprint technique. However, after an eight-week period without specific RST, the observed performance improvements tended to attenuate, suggesting that continued exposure to the stimulus may be necessary to maintain these adaptations. Given the pilot nature of the study and the absence of a control group, these findings should be interpreted as exploratory and should be confirmed in future controlled studies.