Effects of In-Season Velocity-Based vs. Traditional Resistance Training in Elite Youth Male Soccer Players

Abstract

The objectives of this study were to compare the effects of two in-season velocity loss training methods (VBT) on performance outcomes and to evaluate the effects of velocity-based training compared to traditional resistance training (TRT) on performance outcomes in young elite soccer players. VBT utilized the same relative load but varied in the extent of velocity loss during the set: 15% (VL15%) and 30% (VL30%). Thirty-four players were recruited and randomly distributed into three groups: the VL15% group (n = 12; age = 18.50 ± 0.67 years; stature = 183.41 ± 4.25 cm; body mass = 75.08 ± 5.57 kg), the VL30% group (n = 11; age = 17.91 ± 0.60 years; stature = 181.21 ± 6.56 cm, body mass = 73.58 ± 6.22 kg), and the traditional strength training group TRT (n = 11; age = 18.14 ± 0.74 years; stature = 182.17 ± 5.52 cm; body mass = 74.86 ± 6.68 kg). Alongside regular soccer sessions and matches, the groups underwent a four-week (2 sesions per week) resistance training intervention with back squats involved. Changes in leg strength (SQ1RM), 20 m sprint time (SPR 20 m), countermovement jump height (CMJ), reactive strength index (RSI), and change of direction (COD) from before and after were evaluated using a 3 × 2 ANOVA. While no significant interaction was found for SQ1RM and SPR20, all of the groups showed significant pre to post improvements. Significant interactions were observed for CMJ (F = 38.24, p = 0.000), RSI (F = 8.33; p = 0.001), and change of direction agility test (COD) (F = 3.64; p = 0.038), with a post hoc analysis showing differences between the VL15 (6.0%) and TRT (1.7%) groups (p = 0.034); VL15 (12.2%) and VL30 (3.2%) groups (p = 0.004); VL15 and TRT (0.4%) (p = 0.018); VL15 (2.4%) and VL30 (1.5%) (p = 0.049); and between the VL15 and TRT (0.4%) groups (p = 0.015). Four weeks of VL15% training during the season induced similar strength increases to VL30% and TRT, superior improvements in RSI and COD compared to VL30%, and superior improvements in CMJ, RSI, and COD tests compared to TRT. Thus, incorporating the VL15% training method may be recommended to improve power-related performance metrics in elite young soccer players.

1. Introduction

Resistance training (RT) is well established as an integral component of soccer training and as a powerful training stimulus for soccer players’ performance improvement [1]. It has been consistently reported that actions critical to winning in soccer are predominantly executed at a high intensity, encompassing sprints, changes of direction, jumps, tackles, and other movement patterns that demand increased lower body strength and power [2]. RT specific outcomes are largely influenced by several training variables, with relative intensity and volume considered to be determining factors of resulting neurophysiological adaptations [3]. Furthermore, optimizing the load and volume of time spent on resistance training is a crucial concern during the season, considering a wide range of fitness attributes and skills that must be consistently incorporated in the training process [4]. As a result, there is an ongoing demand for identifying and integrating effective yet efficient strength training methods into soccer training [5]. Various resistance training modalities aimed at enhancing strength and power have been introduced over the years, with traditional resistance training considered as a bencmark for prescribing training loads [6]. These traditional resistance training methods are prescribed based on individuals’ 1RM (one repetition maximum), not considering athletes’ daily fluctuations, and thus may lead to loading them inappropriately while decreasing training efficacy [7]. In this context, several resistance training methods, with the umbrella term, “autoregulative resistance training”, were developed recently to enable real-time adjustment of the training intensity according to an individual’s daily fluctuations in performance and training-induced fatigue [8]. The recent availability of kinematic measuring devices and the established correlation between relative load and barbell velocity—when the load is lifted at maximum intended velocity in a non-fatigued state [9,10]—have driven the adoption of barbell velocity as a metric for intensity; a practice referred to as velocity-based training. An increasing amount of evidence suggests that VBT methods to load and volume prescription can induce significant neuromuscular adaptations and physical performance improvements [11] in addition, the term covers variety of approaches with velocity loss thresholds [12] found to be superior to others in terms of strength and power performance outcomes [13]. The training load for each session is defined by the velocity of the fastest repetition, with repetitions continued until velocity falls below a predetermined threshold, indicating the percentage velocity loss (%VL) for the set.

Numerous studies conducted over the past few decades have examined various percentages of velocity loss (%VL). Some research [14,15,16] indicates that a low %VL (5–15%) leads to comparable or greater improvements in strength, sprint time, and jumping ability than a higher %VL (greater than 20%), while not all studies have reached the same conclusion [17,18]. Interestingly, two recently published reviews comparing the effectiveness of velocity-based training (VBT) and traditional strength training presented conflicting results. One review [19] indicated that lower velocity loss positively impacts strength, jumping, and sprinting performance compared to traditional resistance training methods, while the other found no evidence of any differences between VBT and traditional strength training in these outcomes [20]. More studies about the topic are required.

Despite the current importance of resistance training prescription, only a few studies using the velocity loss approach in soccer players have been published. In a 6-week (2 sesions per week) study with professional players, Pareja-Blanco et al. [21] found that group training with a volume load of 15% (15% VL) resulted in significantly greater improvements in strength (estimated one-rep maximum) and power (vertical jump) compared to group training with 30% VL. However, no between-group differences were detected in terms of sprinting and endurance gains. Similarly, Rojas-Jaramilo et al. [22] proved 10% velocity loss training to be superior to 30% velocity loss training in improving strength, sprint, and vertical jump in non-trained young soccer players. Finally, in a sample of strength-naïve young female soccer players with three training sessions per week for 12 weeks during the preseason period, velocity-based training was proven to be superior compared to traditional strength training in countermovement jump and squat power but not maximal squat strength [23].

It is evident that a notable gap remains in the understanding of VBT application within soccer cohorts. Moreover, the lack of research focused on young elite soccer players precludes insights into how effective well-established training modalities can be for highly trained youth athletes. Understanding how different approaches impact strength and power adaptations specifically in this demographic could provide valuable insights to strength and conditioning coaches aiming to maximize performance in competitive settings. The objectives of this research are two-fold: 1. To compare the effects of two in-season velocity loss training methods on performance outcomes, and: 2. To evaluate the effects of velocity-based training in comparison to traditional resistance training on performance outcomes in young elite soccer players. According to a recent review [19], we hypothesized that low velocity-based training (15% VL) would lead to greater improvements in power-related fitness attributes.

2. Materials and Methods

2.1. Participants

Thirty-four elite youth soccer players volunteered to take part in this study and were randomly divided (lottery method) into 3 groups: velocity-based training group with 15% velocity loss alowed in each set—VL15 (n = 12; age = 18.50 ± 0.67 years; stature = 183.41 ± 4.25 cm; body mass = 75.08 ± 5.57 kg), velocity-based training group with 30% velocity loss allowed in each set—VL30 (n = 11; age = 17.91 ± 0.60 years; stature = 181.21 ± 6.56 cm, body mass = 73.58 ± 6.22 kg), and traditional resistance training group—TRT (n = 11; age = 18.14 ± 0.74 years; stature = 182.17 ± 5.52 cm; body mass = 74.86 ± 6.68 kg), which took part in a traditional resistance training program. Measurements were taken using an SECA measuring rod along with an SECA model scale (Seca GmbH, Hamburg, Germany) with 1 mm and 0.1 kg precision for height and body mass, respectively. Height was determined with the participant’s head held in the horizontal Frankfort plane position. The technical error of measurement (TEM) for height and body mass was under 0.02%.

All players were of an elite youth level, playing for two teams competing in the Serbia youth quality league, finishing second and third during the season when the investigation took place. The inclusion criteria for all players were as folows: (1) had at least 4 years of soccer training experience; (2) were free from lower limb injuries or illnesses in the 3 months leading up to the study; and (3) had a minimum of two years of resistance training experience from their time in previous age group squads. A typical in-season microcycle for the teams consisted of 5 soccer sessions (90 min per session), 2 strength and conditioning training sessions, and 1 game. The objectives, testing procedures, and responsibilities during this study were thoroughly communicated to all the participants. Each individual had the option to withdraw from the research at any point. All of the players were free from injury through the study duration, and all finished with over 90% attendance. During the study intervention, there were no reported musculoskeletal injuries or other medical issues that could interfere with the training, and none of the players used medications or dietary supplements. The procedures adhered to the Declaration of Helsinki, and the study protocol was reviewed and approved by the ethics committee of the University of Novi Sad, Serbia (Ref. No. 33-01-07/2021-3). All of the participants and/or their parents/legal guardians received verbal information about the study during one meeting and were provided with a detailed written document explaining this study. All of the participants or their parents or legal guardians voluntarily provided written consent for their participation.

2.2. Study Design

A longitudinal and experimental study employing a between-subjects design was conducted to evaluate the effects of load manipulation based on velocity loss or as a percentage of one-repetition maximum (1RM) within a 4-week resistance training program. This study took place during the latter half of the season (March–April 2021/2022). One week prior to the initial testing, the subjects became acquainted with all of the test procedures. After testing, the players from the VL15 and VL30 groups performed 3 to 4 sets of 3 to 4 repetitions of the squat exercise with varying weights to become familiar with the training protocol. Three days before the program commenced, both the VL15 and VL30 groups participated in an additional familiarization training session. Both the initial and final testing took place across two sessions. Prior to all tests, a standardized warm up was introduced, consisting of a light jogging session, dynamic stretching, and lower limb activation exercises. On day one, the subjects’ anthropometric characteristics were recorded before a 20 m sprint, and 1RM squat tests were conducted, respectively. Two days later, the subjects performed jumping tests, COD tests, and finished with an incremental load–velocity squat test, respectively. Supervised strength training sessions for the groups took place in the mornings at 9:00 a.m. at their respective clubs, equipped with all necessary materials, including GymAware, bars, plates, and elastic bands. Each session was led by three experienced strength and conditioning coaches with a strong background in velocity-based training (VBT) to ensure high-quality training. The sessions were held on Tuesdays and Thursdays, occurring 75–90 min before the regular soccer practice. Final testing was conducted no less than three days after the intervention period, mirroring the initial testing in terms of timing, order, protocols, and examiners. The participants were strongly advised to refrain from any strenuous activities for 24 h prior to testing.

2.3. Measurements

One repetition maximum—1RMSQ

The participants underwent one-repetition maximum (1RM) testing for the free-weight squat following guidelines established by the National Strength and Conditioning Association [24]. The process started with a standardized warm-up including dynamic stretching and preparatory exercises. This was followed by a series of warm-up sets: five repetitions at approximately 50% of their estimated 1RM, three repetitions at around 70% of 1RM, and two repetitions at about 80% of 1RM. Afterward, the participants attempted their 1RM with progressively heavier weights. A research team member monitored their squat depth to ensure that the participants reached a parallel position (thighs parallel to the floor). The participants were permitted up to five attempts, with their highest successful lift recorded as their one-rep max (1RM).

Individualized load–velocity relationships

A linear encoder was utilized to assess the mean velocity (MV) during free-weight back squats (GymAware (GYM), Kinetic Performance Technologies, Canberra, Australia). The participants carried out the half-squat exercise starting from an upright stance. They descended at a self-regulated average speed of approximately 0.50–0.70 m·s⁻¹ until the thighs were parallel with the floor. Then, they immediately reversed the movement to ascend back to the upright position at their maximum intended velocity. Before the main exercise, the participants completed a standardized warm-up like that performed in the one-repetition maximum (1RM) test. They then executed three repetitions at 30% of their baseline 1RM, followed by three at 60%, two at 80%, and one at 90%. Earlier studies verified that GYM provides excellent reliability of MV measurements within the range of 40–90% of 1RM [24]. The participants received verbal encouragement and velocity feedback to achieve maximal concentric velocity on each repetition. A three-minute rest period was allowed between each load. Only the quickest repetition achieved for each absolute load was retained for future analysis. Individualized load–velocity profiles were developed by plotting mean velocities (MVs) against their corresponding loads and applying a line of best fit. The MVs for 60% and 80% of the one-rep max (1RM) were subsequently used to modify the training loads in the velocity-based training (VBT) groups.

Drop jump—RSI index test

In the DJ test, thevparticipants descended from a 30 cm box using their preferred leg, landed on a contact mat with both feet, and then jumped as high as possible. They were instructed to maximize their jump height while minimizing the time spent in contact with the mat during the transition from landing to jumping. Each participant completed the test three times, taking a 30 s rest between jumps, and the highest result was noted for further analysis. The reactive strength index was determined by calculating the flight time/contact time (in seconds) ratio. The highest score was used for analysis.

The Countermovement Jump Test (CMJ) was conducted following the Bosco protocol [25] using a contact platform (Just Jump, Probotics, USA). The participants were directed to place their hands on their hips and maintain an upright position as they dynamically lowered themselves to a depth of their choosing before jumping upward with maximal effort and landing with their knees fully extended. Each participant was allowed three attempts, with a passive rest period of 45 s between repetitions. The highest jump performance recorded from these attempts was used for subsequent analysis.

20 m Sprint Test (SPR20)

The participants underwent a 20 m sprint test, which was timed using light gates (Witty; Microgate, Bolzano, Italy). After a designated warm-up that featured two submaximal efforts, each subject completed two trials. They began from a crouched position with their front foot positioned 0.5 m before the first timing gate, and they started voluntarily to eliminate reaction time. Throughout the test, the subjects received verbal encouragement to exert maximum effort. A two-minute passive recovery period was implemented between trials. The best performance from the trials was selected for subsequent statistical analysis.

In the 505 change of direction tests, light gates were positioned 5 m from a specified turning point. The players began from a starting position 10 m away from the timing gates (15 m from the turning point). Each subject started with their front foot 0.5 m before the timing gate. They were instructed to accelerate through the timing gates as quickly as possible, decelerate at the 15 m mark, and then re-accelerate to return through the timing gates with maximum intention. Two trials were completed with 120 s of passive rest, with the better result used for further analysis.

2.4. Training Interventions

The participants took part in two resistance training sessions to familiarize themselves with the training method, thereby optimizing their training adaptations. All three groups completed a total of 4 weeks of individually supervised strength training, consisting of 2 sessions per week (on Tuesdays and Thursdays), which amounted to 8 training sessions overall. All of the groups also had their usual weekly competitive microcycle, consisting of a recovery day, four soccer field practices, game day, and a day off (

Table 1. Microcycle.

	Mon	Tue	Wed	Thu	Fri	Sat	Sun
AM		RT Soccer field practice	Soccer field practice	RT Soccer field practice	Soccer field practice
PM	Off	Off	Off	Off	Off	Match day	Recovery

AM—morning; PM—afternoon; RT—resistance training.

).

During the 4-week intervention, the resistance training loads (expressed as a percentage of 1RM), the number of sets, and the rest periods between sets were consistently maintained across all groups. The TRT group performed back squats at 80% and 60% of their baseline 1RM during the first and second sessions of the week, respectively. These loads were chosen because they are commonly used in strength training programs, effectively target specific physical qualities along the strength–velocity continuum, and yield reliable velocity data [24]. For the VL15 and VL30 groups, relative loads were established based on the individual squat load–velocity relationship, as recent research has shown a strong correlation between %1RM and mean velocity (MV). Consequently, a target MV for the first repetition of the first exercise set in each session was determined as an estimate of %1RM. The targeted MV for the VL15% group was 0.729 ± 0.075 m/s at 80% of 1RM and 0.861 ± 0.099 m/s at 60% of 1RM. For the VL30% group, the targeted MPV was 0.724 ± 0.048 m/s at 80% of 1RM and 0.831 ± 0.06 m/s at 60% of 1RM.

The absolute load (in kg) was adjusted for each participant to match the desired velocity, maintaining a tolerance of ±0.06 m·s⁻¹ in relation to the targeted %1RM for each session. If the maximum movement velocity (MV) during a set of 5 repetitions deviated by ±0.06 m/s from the target velocity, the barbell load was modified by ±5% of 1RM for the next set. Alongside the back squat as the main compound movement, supplementary exercises were incorporated into the training program (

Table 2. Descriptive characteristics of the base training program.

	Exercise	Sets × Reps	Load
Training session 1	BB back squat	3 × 6	PBT 80% 1RM, VBT: load that aligns with a 80% of 1RM movement velocity.
	Bench press	3 × 6	Body weight
	Deadlift	3 × 6	2 RIR
	BB bent-over row	3 × 6	1 RIR
	OHP	3 × 6
	Lunge †		BM+
Training session 2	BB back squat	3 × 8	PBT 60% 1RM, VBT: load that aligns with 60% of 1RM movement velocity
	Bench press	3 × 10	2 RIR
	Deadlift	3 × 10	2 RIR
	BB bent-over row	3 × 10	1 RIR
	OHP	3 × 10	2 RIR
	Hip thrust	3 × 10	2 RIR

RM—one repetition maximum; BB—barbell; BM—body mass; OHP—overhead press; RIR—repetitions in reserve; † lunge load calculated [16]: 0.6 (6RM squat [kg; 0.52] + 14.82 kg).

).

To maintain consistency across the groups in these exercises, the sets and repetitions were standardized, with the load calculated using specific equations related to body mass or by using a repetitions-in-reserve method (refer to Table 2). All of the participants were given strong verbal encouragement throughout their repetitions to inspire them to give their best effort.

2.5. Statistical Analysis

All of the tested variables were expressed as means ± standard deviations (SDs). The assumption of normality was evaluated using the Shapiro–Wilk test, while Levene’s test was utilized to assess homoscedasticity. Baseline differences between the groups were determined through a univariate analysis of variance (ANOVA), considering the factor group (VL15, VL30, and TRT). The differences between measurements taken before and after the intervention were analyzed using paired samples t-tests for each group individually, and between-group comparisons under the influence of the experimental treatment were assessed using a two-way ANOVA (3x2). The level of significance was established as p ≤ 0.05. A post hoc Bonferroni test following the ANOVAs was performed to determine the significance of the interactions between factors. The mean difference effect size was determined using Cohen’s d, which was calculated by subtracting the means and dividing the outcome by the pooled standard deviation. The Cohen’s d values were interpreted according to Hopkins et al. [26] (small, 0.2; medium, 0.5; large, 0.8; very large, 1.3). Data analysis was executed using the SPSS statistical software package, version 20 (Chicago, IL, USA).

3. Results

No significant differences among the groups were found in the pretest for any of the variables analyzed. Additionally, no significant interaction was identified for SQ1RM (F = 1.96; p = 0.158) and SPR20 (F = 0.542; p = 0.578) (

Table 3. Pre–post and between-group differences between selected variables with % improvement and Cohen’s effect size (d).

	VL15				VL30				TRT
	IN	FIN	%	d	IN	FIN	%	d	IN	FIN	%	d	p
SQ1RM	106.83 ± 12.23	115.83 ± 12.50 *	8.4	0.72	101.09 ± 9.89	111.45 ± 9.07 *	10.2	1.09	105.00 ± 7.51	116.45 ± 7.64 *	10.9	1.49	0.000
CMJ	56.63 ± 7.94	60.04 ± 8.28 *	6.0	0.42	50.51 ± 4.72	53.65 ± 4.97 *	6.2	0.66	50.98 ± 4.77	51.85 ± 4.71 *	1.7	0.18	0.000 ‡
RSI	2.45 ± 0.21	2.75 ± 0.26 *	12.24	1.42	2.18 ± 0.23	2.25 ± 0.29	3.2	0.26	2.27 ± 0.33	2.28 ± 0.29	0.4	0.03	0.001 †‡
SPR20m	3.00 ± 0.76	2.95 ± 0.89 *	1.6	0.06	3.07 ± 0.43	3.02 ± 0.33 *	1.6	0.11	3.09 ± 0.50	3.05 ± 0.44 *	1.2	0.08	0.587
COD	2.20 ± 0.09	2.10 ± 0.06 *	2.4	1.11	2.27 ± 0.10	2.22 ± 0.10 *	1.5	0.50	2.29 ± 0.09	2.24 ± 0.11	0.6	0.55	0.038 †‡

SQ1RM—1RM squat test; CMJ—countermovement jump test; RSI—reactive strength index test SPR20m—20 m sprint test; 505 test—change of direction test; IN—initial tests result ± standard deviation; FIN—final test result; %—percent improvement; p—level of statistical significance; * statistically significant difference pre vs. post p < 0.05; †—statistically significant difference between VL15 and VL30 group; ‡—statistically significant difference between VL15 and TRT group.

). Pre vs. post comparisons showed significant improvements for SQ1RM in all groups, with 8,4% (moderate effect size), 10.2% (moderate effect size), and 10.9% (large effect size) for the VL15, VL30, and TRT groups, respectively. In addition, for SPR20m, significant differences from preintervention to postintervention were seen for all the groups, although with a small improvement percentage (1.6% for VL15 and VL30 and 1.2% for TRT) and trivial effect size for all the groups.

A significant interaction was observed for the Countermovement Jump (CMJ) test (F = 38.24, p = 0.000), with the post hoc analysis revealing differences between the VL15 and TRT groups (p = 0.034). When comparing the initial and final measurements, the VL15 group demonstrated a 6.0% improvement (moderate effect size), while the VL30 group showed a 6.2% improvement (moderate effect size). In contrast, the TRT group had a smaller improvement of 1.7% (trivial effect size). For the Reactive Strength Index (RSI), a significant interaction was noted (F = 8.33; p = 0.001), with the post hoc analysis revealing differences between the VL15 and VL30 groups (p = 0.004) as well as between the VL15 and TRT groups (p = 0.018). The VL15 and VL30 groups achieved improvements of 12.2% (large effect size) and 3.2% (small effect size), respectively, while the TRT group saw a minimal improvement of 0.4% (trivial effect size). The COD test also showed a significant interaction (F = 3.64; p = 0.038). The post hoc analysis indicated significant differences between the VL15 and VL30 groups (p = 0.049) and between the VL15 and TRT groups (p = 0.015), but no significant difference was found between the VL30 and TRT groups (p = 0.980). In terms of percentage improvements from pre- to post-testing, the VL15 group improved by 2.4% (moderate effect size), the VL30 group improved by 1.5% (moderate effect size), and the TRT group improved by 0.6% (trivial effect size).

4. Discussion

Research indicates that velocity-based training can effectively improve various fitness attributes in athletes. However, there is a scarcity of studies focused on its effectiveness in soccer. Thus, this investigation aimed to compare the impacts of two in-season velocity loss methods (VL15 and VL30) with traditional strength training (TRT) on leg strength, change of direction ability, jumping, and sprinting performance in elite youth soccer players. The present study results confirmed that all the groups significantly increased their strength level, with no significant differences in strength-enhancing capacity between the groups. VL15 was proved to be superior to both VL30 and TRT in developing the reactive strength index (RSI) and COD test. In addition, all three groups significantly increased in their countermovement jump pre vs. post (p < 0.05), only VL15 significantly improved in RSI, and both VL 15 and VL30 significantly improved in their COD performance. In accordance with the initial hypothesis, it seems that adding two sessions a week for four weeks of velocity-based training with a relative load of 60–80% and velocity loss threshold of 15% appears to be a more robust strategy than VL30 or traditional strength training for lower body strength, jumping, and change of direction performance enhancement during the competitive period in elite youth soccer players. Ultimately, neither training method was shown to be effective in improving 20 m sprint performance.

A growing body of studies have recently emerged evaluating the impacts of different %VLs on the set on performance outcomes in athletic populations, with the presented data generally in line with this study’s findings. In an 8-week study involving 16 training sessions conducted by Pareja-Blanco et al. [15], it was found that velocity-based training (VBT) with a 20% velocity loss produced squat strength gains similar to those achieved with a 40% velocity loss (18% and 13.4%, respectively). Additionally, the 20% velocity loss group demonstrated significantly greater improvements in counter-movement jump (CMJ) performance, showing an increase of 9.5% compared to just 3.5% in the 40% velocity loss group (p < 0.05) among resistance-trained young athletes. The outcomes of the 8 weeks of Rodriguez-Rosell et al.’s study [27] demonstrate that VL10% training conducted twice a week resulted in larger percentage increases in Countermovement Jump (CMJ) performance (9.2% compared to 5.4%) and sprint performance (−1.5% versus 0.4%) compared to VL30% in resistance-trained young males. Three velocity-based groups (VL10%, VL30%, and VL45%) underwent eight weeks of velocity-based training (VBT) with consistent training parameters: a load of 55–70% of one-repetition maximum (1RM), a training frequency of two sessions per week with three sets per session, and a four-minute recovery period between sets. All of the groups demonstrated significant improvements in muscle strength (VL10%: 6.4–58.6%; VL30%: 4.5–66.2%; VL45%: 1.8–52.1%; p < 0.05–0.001). Notably, the VL10% group experienced a significantly greater improvement in counter-movement jump (CMJ) performance (11.9%), as well as a larger percentage change in sprint performance compared to the other two groups (VL10%: −2.4%; VL30%: −1.8%; VL45%: −0.5%). Finally, two recently published reviews dealing with the effects of VBT on the performance outcomes of elite athletes [28] and trained individuals [29] suggest that VBT may effectively improve maximum leg strength, countermovement jump, and sprint ability. Furthermore, it appears that implementing smaller velocity losses (up to 20%) promotes more beneficial neuromuscular adaptations, decreases neuromuscular fatigue, and ensures a higher quality of performance with a significantly lower total volume of training. Overall, the previously mentioned study supports this study’s findings that low velocity loss training leads to superior power-related performance while yielding similar strength outcomes compared to high velocity loss training (30%) in athletes.

VBT appears to be an effective method for enhancing performance attributes in well-trained individuals. This is widely supported by various studies that compare the effects of VBT with traditional resistance training on performance outcomes. In a six-week (two sessions/week) study of trained men by Dorelli et al. [16], the VBT group achieved similar improvements in back squats (9% vs. 8%) and a significantly better jump height (5% vs. 1%) when compared with a TRT approach, despite a significant 9% reduction in the total training volume. Additionally, a comparison of six weeks of velocity-based training (VBT) versus traditional resistance training (TRT), conducted with resistance-trained individuals [30], demonstrated more favorable training outcomes for the VBT group in jump height (effect size = 1.81), sprint performance (effect size = 1.27–1.35), and change of direction (effect size = 0.67–0.97). However, it remained unclear if either training method was superior for enhancing maximal strength (effect size = −0.57). The observed training effects were greater than those reported in the present study, which could be attributed to the longer study duration (6 weeks compared to 4 weeks) as well as the higher volume of lighter loads, faster training repetitions, and consequently, more pronounced adaptations in the velocity aspect of the force–velocity curve and high-speed actions such as the countermovement jump, sprint, and change of direction [14]. Furthermore, the increases in maximal strength observed for both groups in this study were nearly identical to the present study’s findings for the VL30% and TRT groups (10.2%, ES = 1.09 and 10.9%, ES = 1.49, respectively). These results were also slightly higher than those reported by Dorrell et al. (~9%, ES = 0.59 vs. ~8%, ES = 0.44 for VBT and TRT, respectively) and our VL15% group (8.4%, ES = 0.72). The six-week study comprised two weekly sessions dedicated to free weight squats, with a progressively increasing load that ranged from 65% to 90% of the 1RM, aimed to evaluate the impact of velocity-based resistance training compared to traditional resistance training on the athletic performance of college female basketball players [31]. Despite similar strength gains (22.3%, ES = 1.39 vs. 19.4, ES = 3.09), velocity-based training was proved to produce more favorable improvements in CMJ (7.3%, ES = 0.53), RSI (23.9%, ES = 0.85) and the COD test (4.6%, ES = 0.87). In addition, no group-by-time interaction and no significant improvements were found for sprint performance, altogether corroborating the present study’s findings.

Interestingly, only a few studies have compared the effects of VBT with different %VL and/or traditional resistance training on performance within a population of soccer players [21,22,23]. Two previous studies, Pareja-Blanco et al. (2017) and Rojas-Jaramillo et al. (2024), compared resistance training programs in male soccer players using similar %VL in the set (10 vs. 30% and 15% vs. 30%, respectively). Although the range of relative loads used was lower (50–70% 1RM and 45–60% 1RM), the results obtained were generally in line with the present study’s findings: groups the lower velocity loss in both studies showed larger improvements in countermovement jump (5.3%; ES: 0.40 and 8.0%, ES: 0.51) and similar in 1RM strength (5.3%; ES: 0.45 and 22.2, ES: 1.65). In addition, while sprint time showed no significant improvements pre vs. post in the Parejo-Blanco study (0.4%: ES 0.10), being practically identical to the present study’s findings, Rojas-Jamarillo’s study reported significant and large sprint time pre vs. post improvements (11.3%, ES: 1.85). This difference could likely be attributed to both the younger age of this study’s participants compared to both ours and Pareja-Blanco’s study, as well as the resistance training history (resistance training naïve cohort). Finally, to the best of our knowledge, only one study compared the effects of velocity-based training and traditional strength training. Ortega et al. [23] examined the effects of VBT (20% velocity drop threshold with load equivalent to 65% of 1RM) vs. TRT (load equivalent to 80% of 1RM) on 30 sprint time, countermovement jump, lower body maximum strength, and maximal squat power. The presented data indicate that the VBT group, with a 42% lower volume of training, produced significant increases in maximal strength (p < 0.000), squat power (p < 0.000), 30m sprint time (p < 0.000), and countermovement jump (p< 0.001), with squat power and CMJ being meaningfully better than the results of the TRT group (p < 0.008). Together, the aforementioned data support the effectiveness of velocity-based training in enhancing strength and power performance attributes in well-trained youth soccer players. However, it is essential to exercise caution regarding the threshold load required to achieve the desired improvements. Clearly, further research on this topic is needed.

It is noteworthy that we did not find significant effects of either velocity-based training or traditional resistance training on the 20 m sprint time among our participants. This aligns with the findings of Pareja-Blanco et al. (2017), who reported no change in 20 m sprint times following lower load velocity-based training in professional soccer players. It can be speculated that the obtained results may be partly due to the training status of the current study’s participants (well-trained), as research has shown that trained adolescents tend to improve less in sprint outcomes with resistance training than their untrained peers [32]. Furthermore, the training and testing specificity may influence outcomes, as the upward force vector applied during this study training sessions is likely crucial for inducing specific functional adaptations [33].

The present study results demonstrated significant changes in the performance parameters when utilizing velocity loss methods, with smaller velocity losses usually eliciting greater neuromuscular adaptations compared to larger velocity losses. Although this study does not address them, it is worth briefly hypothesizing the mechanisms that could explain the observed improvements in the performance outcomes. Clearly, reducing velocity losses allows for higher-quality repetitions, leading to greater speeds attained during the exercise session.

Parejo-Blanco et al. [18] explored the effects of applying velocity losses of 20% and 40% in young males on performance enhancements, changes in muscle fiber characteristics, and cross-sectional area. Their findings revealed that the group experiencing a high velocity loss of 40% exhibited a significant decrease in type IIX fibers, which could adversely affect strength and power development as well as lengthen recovery times [34]. In addition, Rodríguez-Rosell et al. [14] demonstrated that their low velocity loss group (VL10%) experienced significant improvements in strength, jumping, and sprinting performance, which were accompanied by increased neural activation of the agonist muscles involved in the exercises. In contrast, neural activation remained unchanged for the VL30% and VL45% groups. Collectively, these findings suggest that establishing a low velocity loss limit (5–15%) during each exercise set leads to both effective and efficient training stimuli, promoting significant neuromuscular adaptations while requiring fewer repetitions and inducing lower levels of fatigue (mechanical and physiological stress) compared to training volumes associated with higher velocity loss thresholds. This could be of particular importance during soccer season, when the general objective of resistance training is to induce long-term benefits on performance outcomes while maintaining peak performance during subsequent field soccer practice. Indeed, resistance training with heavy loads and near failure repetitions per set has been shown to induce greater short-term deterioration in jump, sprint, and change of direction compared to lower repetitions or %VL in the set [35,36]. These physiological differences reinforce the present study’s findings and highlight the advantageous use of velocity-based training to optimize strength and power adaptations in elite youth soccer players.

Several limitations of this study should be noted. The participant sample is limited to players from Serbia, and we do not know whether athletes from other countries have the same characteristics. The training load was manipulated solely for the back squat; however, this approach is consistent with previous velocity-based training (VBT) studies. Adjusting the load for other lower-body exercises in the training routine, such as the deadlift, forward lunge, and hip thrust, using velocity thresholds was not deemed appropriate, as maximizing concentric velocity was not the primary training objective for these movements. Additionally, we did not monitor the load used during regular soccer practice sessions conducted among all three groups with their respective coaches, which could have impacted the training adaptations. Lastly, this study only lasted four weeks, highlighting the need for comparative investigations using strength modalities over longer durations.

5. Conclusions

It seems that adding two sessions a week for four weeks of velocity-based training with a relative load of 60–80% and velocity loss threshold of 15% appears to be a more robust strategy than VL30% or traditional strength training for jumping and change of direction performance enhancement during the competitive period in elite youth soccer players. In addition, all three resistance training modalities were comparably effective in maximal strength gains. Strength and conditioning specialists may consider implementing low velocity loss training modalities during the competitive season in elite youth soccer players to achieve substantial improvements in the mentioned performance outcomes.