Abstract

Purpose:

To compare the effects of maximum intended velocity (MIV), velocity-loss threshold (VLT), target velocity zone combined with velocity-loss threshold (TVZ+VLT), and traditional percentage-based training (PBT) on explosive performance and sprint outcomes in resistance-trained men.

Methods:

Fifty-two resistance-trained male athletes were randomly assigned to one of four groups and completed 12 supervised back squat sessions over 4 weeks. Primary outcomes were countermovement jump (CMJ) height, 0–30 m sprint time, and isometric mid-thigh pull rate of force development (IMTP RFD). Secondary outcomes included CMJ peak force and peak power, 0–10 m sprint time, IMTP peak force, relative IMTP peak force, and dynamic strength index (DSI).

Results:

Significant baseline-adjusted group effects were retained after Holm correction for CMJ height (F = 7.50, P = .001, ηp² = .324) and 0–30 m sprint time (F = 4.77, P = .011, ηp² = .233). MIV and TVZ+VLT produced greater improvements in CMJ height than PBT, and TVZ+VLT produced a greater improvement in 0–30 m sprint time than PBT. Corrected group effects were also observed for CMJ peak power and DSI, whereas IMTP RFD and IMTP peak force did not show prescription-specific differences.

Conclusions:

MIV and TVZ+VLT prescription packages were associated with more favorable short-term changes in selected explosive-performance and sprint-related outcomes than PBT. However, because the interventions differed in velocity zone, relative intensity, movement intent, feedback provision, and velocity-loss termination, and because session-level realized dose data were not retained, these findings should be interpreted as package-level effects rather than evidence for the independent superiority of any isolated VBT component.

1 Introduction

Muscular strength and the capacity to express force rapidly are important determinants of many sport actions, including jumping, sprint acceleration, change of direction, and repeated high-intensity mechanical output (Suchomel et al., 2016, 2018). Resistance training is therefore widely used to improve athletic performance, but the adaptive response to a resistance-training program depends not only on exercise selection and weekly frequency, but also on how load, effort, movement velocity, volume, and fatigue are prescribed within each session (Kraemer and Ratamess, 2004; Zourdos et al., 2016). In this context, the method used to regulate back squat training may influence whether adaptations are biased toward maximal force production, explosive power, or sprint-related performance.

Traditional percentage-based training (PBT) remains one of the most common resistance-training prescription models (Sánchez-Medina and González-Badillo, 2011; Weakley et al., 2021). In this approach, external load is typically anchored to a fixed fraction of one-repetition maximum (1RM), and training volume is controlled through planned sets, repetitions, and load progression. PBT is practical and easy to implement, but it assumes that a given percentage of 1RM represents a relatively stable training stimulus across sessions. In practice, daily fluctuations in fatigue, readiness, sleep quality, technical performance, and concurrent training stress may alter the actual stimulus imposed by the same relative load (Behm and Sale, 1993; González-Badillo and Sánchez-Medina, 2010). These limitations have increased interest in velocity-based training (VBT) as a method for monitoring and regulating resistance-training intensity and fatigue.

VBT uses barbell velocity to inform load selection, estimate relative intensity, provide feedback, and monitor within-set fatigue (González-Badillo and Sánchez-Medina, 2010; Sánchez-Medina and González-Badillo, 2011; Weakley et al., 2021). Because movement velocity is closely related to relative load during resistance exercises, and because velocity loss during a set reflects fatigue accumulation, VBT provides a practical framework for autoregulating training in applied sport settings. However, VBT should not be considered a single intervention. Different VBT-derived prescriptions may vary in their emphasis on movement intent, target velocity zone, feedback provision, and velocity-loss termination. Therefore, studies comparing VBT with PBT have provided useful evidence that velocity monitoring can improve load individualization or training outcomes, but they do not fully explain which type of velocity-informed prescription is most appropriate for a given performance goal (Dorrell et al., 2020).

Several VBT-related prescription models are particularly relevant to lower-body strength and power training. Maximum intended velocity (MIV) emphasizes the instruction to move every concentric repetition as fast as possible, even when external load constrains actual barbell velocity; this approach is grounded in the concept that intended movement velocity can influence velocity-specific training adaptations (Behm and Sale, 1993; González-Badillo et al., 2014). Velocity-loss threshold (VLT) training terminates sets once repetition velocity declines beyond a predetermined threshold, thereby regulating fatigue exposure and influencing the balance between mechanical volume, metabolic stress, and velocity quality (Sánchez-Medina and González-Badillo, 2011; Pareja-Blanco et al., 2017; Weakley et al., 2020). Target velocity zone combined with velocity-loss threshold training (TVZ+VLT) uses velocity to guide load selection within a specific velocity range while also limiting within-set velocity loss, which links training to a targeted region of the force-velocity continuum (Samozino et al., 2012; Conceição et al., 2016; Morin and Samozino, 2016). These models differ conceptually and practically from PBT, but they are often discussed under the broad label of VBT. This creates an important research gap, because different velocity-based prescription packages may produce different adaptations in explosive performance, sprint ability, and force-production characteristics.

Recent evidence has increasingly emphasized that velocity- and intent-related variables are not merely monitoring tools, but may shape the adaptive profile of resistance training. Studies in elite or chronically strength-trained athletes have shown that maximal-intended-velocity or ballistic back-squat training can preferentially improve movement velocity, propulsive power, and rapid force-production characteristics, even when maximal strength gains are comparable between training approaches (Lecce et al., 2025a, 2025b). Similarly, work in sprinters suggests that autoregulated VBT may provide advantages over fixed percentage-based loading for selected neuromuscular and sprint-related outcomes (Guo et al., 2026). Collectively, these findings indicate that movement intent, execution velocity, and autoregulated loading are practically relevant prescription factors. However, existing studies have generally compared one velocity-informed approach with a conventional or controlled-velocity condition, rather than directly comparing several applied prescription packages that differ in target velocity zone, velocity-loss termination, feedback, and load-regulation strategy within the same experimental framework. The present study addresses this gap by comparing MIV, VLT, TVZ+VLT, and PBT as integrated back-squat prescription packages, while assessing jump, sprint, and isometric force-production outcomes together.

Therefore, the purpose of this study was to compare the short-term effects of four integrated back-squat prescription packages, namely MIV, VLT, TVZ+VLT, and PBT, on explosive performance, sprint-related outcomes, and isometric force-production characteristics in resistance-trained men. We hypothesized that MIV and TVZ+VLT would produce greater improvements in explosive-performance and sprint-related outcomes than PBT and VLT, whereas maximal isometric force would improve across all conditions.

2 Materials and methods2.1 Participants

A repeated-measures analysis-of-variance power calculation was conducted using G*Power. Based on a medium effect size (f = 0.25), α = .05, power = .80, 4 groups, and 2 repeated measurements, a minimum sample size of 48 participants was required. To allow equal group allocation and potential attrition, 52 resistance-trained male athletes were recruited. Participants were resistance-trained male athletes (age: 21.5 ± 1.9 y; relative back squat 1RM: 1.42 ± 0.19 kg·kg−1), with ≥1 year of systematic resistance-training experience and technical proficiency in the free-weight back squat and bench press. Participants were excluded if they reported recent musculoskeletal injury, contraindications to maximal testing or resistance training, use of substances affecting neuromuscular performance, or additional structured training during the intervention. The study was approved by the Institutional Review Board of Ankang University (Approval No.: XIPE-2024-ETH-03-018) and complied with the principles of the Declaration of Helsinki, with written informed consent obtained from all participants.

2.2 Experimental design

Our intention was to assess potential differences among 4 back squat prescription models on measures of explosive performance, sprint performance, and isometric force-production characteristics. Fifty-two resistance-trained male athletes were randomly assigned to 4 groups: maximum intended velocity (MIV; n = 13), velocity-loss threshold (VLT; n = 13), target velocity zone combined with velocity-loss threshold (TVZ+VLT; n = 13), and traditional percentage-based training (PBT; n = 13). Group allocation was performed using a computer-generated randomization sequence with equal allocation to the four groups. Baseline characteristics of each group are presented in Table 1.

VariableUnitMIV (n=13)VLT (n=13)TVZ+VLT (n=13)PBT (n=13)pAgeyears21.5 ± 1.921.8 ± 2.021.2 ± 1.821.6 ± 2.10.83Heightcm176.3 ± 5.2175.8 ± 6.1177.1 ± 5.7176.6 ± 5.90.91Body masskg75.4 ± 7.676.3 ± 7.974.9 ± 8.274.1 ± 7.10.88Body mass indexkg·m-²24.2 ± 1.824.1 ± 2.024.3 ± 1.923.8 ± 1.70.94Training experienceyears2.4 ± 0.92.6 ± 1.02.3 ± 0.82.5 ± 1.10.79Relative back squat 1RMkg·kg-¹1.43 ± 0.181.40 ± 0.201.45 ± 0.171.41 ± 0.190.86

Participant baseline characteristics by group.

Values are mean ± SD. These baseline characteristics were retained from the original dataset documentation and did not differ significantly between groups.

Measurements of countermovement jump (CMJ) height, CMJ peak force and peak power, 0–10 m and 0–30 m sprint performance, isometric mid-thigh pull (IMTP) peak force and rate of force development, relative IMTP peak force, dynamic strength index, and load–velocity profiling were taken after familiarization. All participants completed 12 supervised back squat training sessions over a 4-week period, with 3 sessions performed each week and at least 48 hours between sessions. The free-weight back squat was the primary training exercise and was prescribed according to the allocated model. Secondary exercises were standardized across groups and consisted of the Romanian deadlift, leg press, and bench press, performed using the same planned sets, repetitions, and loading scheme. Posttests were conducted using the same procedures and testing order as the pretests. The overall study design is shown in Figure 1.

Technical framework of the trial. The image summarizes randomization, four prescription packages, intervention structure, outcome hierarchy, and the statistical interpretation boundary.

2.3 Training intervention

Participants in all groups completed 12 supervised training sessions over a 4-week period, with 3 sessions performed each week and at least 48 hours between sessions. A standardized warm-up was completed before each session. The free-weight back squat was the primary training exercise and was prescribed according to the allocated training model. The planned back squat prescriptions are presented in Table 2.

ParameterMIV (n=13)VLT (n=13)TVZ+VLT (n=13)PBT (n=13)Load basis% estimated 1RMLoad-velocity profileTarget velocity zone% estimated 1RMIntensity weeks 1-270-75% 1RM0.55-0.65 m·s-¹ (~75-80% 1RM)0.75-0.85 m·s-¹ (~60-70% 1RM)75% 1RMIntensity weeks 3-470-75% 1RM0.55-0.65 m·s-¹ (~75-80% 1RM)0.75-0.85 m·s-¹ (~60-70% 1RM)80% 1RMSets4444Repetitions weeks 1-26 fixed4-8; VLT governed4-8; VLT governed6 fixedRepetitions weeks 3-46 fixed4-8; VLT governed4-8; VLT governed5 fixedVelocity-loss criterionNone≥20% from reference repetition≥15% from first repetitionNoneVelocity feedbackYesYesYesNo

Planned back squat prescription by group.

The table describes planned prescription. Actual session-level completed repetitions, tonnage, realized velocity loss, and individual dose variability were not available; therefore, actual training-dose equivalence cannot be verified.

The MIV group performed the back squat with a fixed planned load of approximately 70% to 75% of estimated 1RM. Participants were instructed to complete every concentric repetition with maximal intended velocity, and this instruction was reinforced before and during each set. Sets were not terminated according to a velocity-loss rule. The VLT group used a load selected from a slower target mean concentric velocity range of approximately 0.55 to 0.65 m·s−1. During each set, repetitions were continued until the prescribed velocity-loss threshold was reached. This condition represented a velocity-based prescription model emphasizing fatigue regulation within a heavier, strength-oriented velocity zone. The TVZ+VLT group used a target first-repetition velocity zone of approximately 0.75 to 0.85 m·s−1 combined with velocity-loss set termination. This prescription was designed to regulate both load selection and within-set fatigue while maintaining a higher-velocity training stimulus. The PBT group followed a fixed percentage-based loading model without real-time velocity feedback. Participants completed the prescribed repetitions regardless of barbell velocity. This condition served as the traditional resistance-training comparison group.

Secondary exercises were standardized across groups and consisted of the Romanian deadlift, leg press, and bench press, performed using the same planned sets, repetitions, and loading scheme. Because session-level completed repetitions, total tonnage, mean concentric velocity across all repetitions, and realized velocity loss were not retained, the groups were considered matched at the level of planned training structure and secondary exercise programming rather than verified realized mechanical dose.

2.4 Procedures2.4.1 Countermovement jump

To conduct the CMJ test, participants were instructed to jump as high as possible. The procedure involved starting from a standing position, performing a downward countermovement to a self-selected depth, and then immediately jumping vertically in 1 continuous movement. Participants were instructed to keep their hands on their hips throughout the test to eliminate the influence of arm swing. All jumps were performed on a calibrated force platform, allowing direct measurement of vertical ground reaction forces. CMJ height, propulsion peak force, and peak power were calculated from the force–time record using established impulse–momentum procedures. Each participant completed 3 maximal CMJ trials, with approximately 30 seconds of rest between trials. The trial with the greatest jump height was retained for subsequent analysis (Linthorne, 2001; McMahon et al., 2018).

2.4.2 Sprint testing

Linear sprint performance was assessed over 0–10 m and 0–30 m using electronic timing gates. Before testing, participants completed sprint-specific warm-up efforts at submaximal intensities. Participants started each trial from a standing 2-point stance, with the front foot positioned behind the first timing gate to avoid premature triggering. Each participant completed 2 maximal sprint trials, separated by 3 minutes of passive recovery. Participants were strongly encouraged to sprint maximally through the final timing gate. The fastest 0–10 m and 0–30 m split times were retained for subsequent analysis.

2.4.3 Isometric mid-thigh pull

IMTP testing was used to assess maximal isometric force-production capacity and rate of force development. Participants adopted a standardized mid-thigh pull position with joint angles consistent with established recommendations (Comfort et al., 2018; McMahon et al., 2018). They were instructed to pull “as hard and as fast as possible” against the fixed bar and to sustain maximal effort for approximately 3 seconds. Peak force, relative peak force, and rate of force development over the 0–200 ms epoch from force onset were retained for analysis. Force onset was defined as the point at which vertical force exceeded 5 times the standard deviation of the quiet standing baseline, consistent with established methodological recommendations (Maffiuletti et al., 2016). The best trial, based on the primary force criterion, was used for subsequent analysis.

2.4.4 Load–velocity profiling and estimated 1RM

Participants completed an incremental loading protocol in the free-weight back squat to establish an individual load–velocity profile. Mean concentric velocity was recorded using a linear position transducer. Loads were progressively increased across trials, and participants were instructed to perform the concentric phase with maximal intended velocity while maintaining correct squat technique. The load–velocity relationship was fitted using linear regression, and estimated 1RM was derived from the load corresponding to the minimum velocity threshold for the back squat. The estimated 1RM was then used to guide the training load prescriptions for the relevant groups (González-Badillo and Sánchez-Medina, 2010; Comfort et al., 2015; Sánchez-Medina et al., 2017).

2.5 Statistical analysis

The data are presented as mean (SD). Normality was assessed using the Shapiro–Wilk test and visual inspection of quantile–quantile plots, and homogeneity of variance was assessed using Levene’s test. Posttest values were analyzed using analysis of covariance, with group as the fixed factor and the corresponding pretest value as the covariate. Holm correction was applied separately for the primary outcomes, including countermovement jump height, 0–30 m sprint time, and isometric mid-thigh pull rate of force development, and for the secondary outcomes, including countermovement jump peak force and peak power, 0–10 m sprint time, isometric mid-thigh pull peak force, relative isometric mid-thigh pull peak force, and dynamic strength index. Pairwise between-group differences in change scores were expressed as Hedges’ g with 95% confidence intervals. Partial eta-squared was calculated for analysis-of-covariance models, with values of ≥.06 and ≥.14 interpreted as moderate and large effects, respectively. All statistical analyses were performed using IBM SPSS Statistics software, version 29.0. The significance level was set at P <.05 after correction.

3 Results3.1 Participant flow and baseline comparability

All 52 randomized participants completed the intervention and were included in the final analysis. Each group contained 13 participants. No adverse events were recorded in the retained study documentation. All participants attended each of the 12 scheduled training sessions; no sessions were missed or required rescheduling. Squat technique was monitored by a qualified strength and conditioning coach throughout each session. No participants reported significant musculoskeletal discomfort or adverse responses attributable to the training intervention. Baseline values did not differ between groups for any analyzed performance outcome, and baseline demographic and training characteristics were similar across groups (Table 1). Planned training prescriptions are shown in Table 2. Descriptive pre- and post-intervention values and observed changes are summarized by outcome family in Tables 3 and 4. Selected pairwise change-score contrasts are reported in Table 5.

OutcomeUnitMIV change [95% CI]VLT change [95% CI]TVZ+VLT change [95% CI]PBT change [95% CI]Fpη²CMJ heightcm3.79 [2.54, 5.04]2.10 [1.04, 3.16]4.21 [3.15, 5.26]1.40 [0.44, 2.36]7.500.0010.3240-30 m sprints-0.14 [-0.20, -0.08]-0.08 [-0.13, -0.03]-0.15 [-0.20, -0.10]-0.05 [-0.11, 0.01]4.770.0110.233IMTP RFD 0-200 msN·s-¹580.01 [411.38, 748.64]498.02 [281.60, 714.43]612.00 [418.15, 805.85]411.98 [289.29, 534.67]1.430.2450.084

Primary outcome change scores and baseline-adjusted group effects.

Sprint-time changes are negative when performance improves. ANCOVA models used post-intervention outcome as the dependent variable, group as fixed factor, and baseline value as covariate. Holm correction was applied within the primary-outcome family.

OutcomeUnitMIV change [95% CI]VLT change [95% CI]TVZ+VLT change [95% CI]PBT change [95% CI]Fpη²CMJ peak forceN196.04 [112.02, 280.05]125.32 [51.58, 199.07]185.70 [99.01, 272.39]96.45 [-7.20, 200.09]1.750.5080.101CMJ peak powerW312.00 [204.65, 419.35]197.99 [93.72, 302.26]341.00 [250.00, 432.00]142.00 [59.89, 224.11]4.620.0330.2280-10 m sprints-0.08 [-0.12, -0.04]-0.05 [-0.10, -0.00]-0.09 [-0.11, -0.07]-0.03 [-0.07, 0.01]3.640.0770.188IMTP peak forceN188.98 [108.65, 269.31]256.00 [145.61, 366.39]197.98 [94.26, 301.71]288.99 [171.59, 406.39]0.940.8570.057Relative IMTP peak forceN·kg-¹2.49 [1.08, 3.90]3.38 [1.51, 5.26]2.61 [0.65, 4.56]3.89 [2.37, 5.42]0.850.8570.052Dynamic Strength IndexAU0.022 [0.008, 0.035]-0.020 [-0.037, -0.003]0.018 [-0.001, 0.038]-0.038 [-0.058, -0.018]14.49<0.0010.481

Secondary outcome change scores and baseline-adjusted group effects.

Holm correction was applied within the secondary-outcome family. DSI is interpreted as an exploratory derived index rather than as a standalone diagnostic tool.

OutcomeComparisonMean difference95% CIHedges gpCMJ heightMIV vs PBT2.390.90 to 3.891.260.012CMJ heightTVZ+VLT vs PBT2.811.46 to 4.161.630.001CMJ heightTVZ+VLT vs VLT2.110.69 to 3.521.170.016CMJ heightMIV vs TVZ+VLT-0.42-1.96 to 1.13-0.210.5850-30 m sprintMIV vs PBT-0.09-0.17 to -0.01-0.900.1080-30 m sprintTVZ+VLT vs PBT-0.10-0.17 to -0.03-1.110.0370-30 m sprintTVZ+VLT vs VLT-0.07-0.13 to -0.00-0.840.1130-30 m sprintMIV vs TVZ+VLT0.01-0.06 to 0.080.100.797CMJ peak powerMIV vs PBT170.0041.97 to 298.031.040.046CMJ peak powerTVZ+VLT vs PBT199.0082.89 to 315.111.340.008CMJ peak powerTVZ+VLT vs VLT143.0111.91 to 274.100.860.101CMJ peak powerMIV vs TVZ+VLT-29.00-162.31 to 104.31-0.170.657Dynamic Strength IndexMIV vs PBT0.0600.037 to 0.0832.04<0.001Dynamic Strength IndexTVZ+VLT vs PBT0.0570.030 to 0.0841.67<0.001Dynamic Strength IndexTVZ+VLT vs VLT0.0380.014 to 0.0631.210.008Dynamic Strength IndexMIV vs TVZ+VLT0.003-0.020 to 0.0260.100.785IMTP RFD 0-200 msMIV vs PBT168.02-29.52 to 365.560.670.368IMTP RFD 0-200 msTVZ+VLT vs PBT200.02-17.30 to 417.330.720.348IMTP RFD 0-200 msTVZ+VLT vs VLT113.98-161.23 to 389.200.321.000IMTP RFD 0-200 msMIV vs TVZ+VLT-31.99-275.37 to 211.38-0.101.000

Selected pairwise comparisons of change scores.

For sprint time, negative mean differences indicate greater improvement for the first-listed group. Hedges g uses the pooled SD of change scores; interpretation should consider outcome direction.

The results are presented according to the revised hierarchy of outcomes. This is important because several variables changed in the expected direction, but not all retained corrected between-group evidence. The narrative therefore distinguishes between corrected group effects, descriptive within-group improvements, and exploratory patterns. Individual change-score distributions are displayed in Figure 2, and standardized pairwise contrasts are displayed in Figure 3.

Individual change-score distributions across groups for (A) CMJ height, (B) 0–30 m sprint time, (C) IMTP RFD, (D) CMJ peak power, and (E) Dynamic Strength Index. Boxplots show median and interquartile range; open circles represent individual participants; diamonds represent group mean with 95% Cl. For 0-30 m sprint time, more negative values indicate faster performance.

Selected standardized between-group change-score contrasts for key outcomes. Points represent Hedges g and horizontal lines represent approximate 95% confidence intervals derived from the corresponding change-score contrasts. Interpretation should consider the direction of each outcome, especially sprint time.

3.2 Primary outcomes

For CMJ height, the baseline-adjusted group effect remained significant afte