Introduction

During a soccer match, 83% of goals are preceded by an explosive action, often involving a straight-line sprint, a jump, or a change of direction (Faude et al. 2012). Given that 90% of sprints are shorter than 20 m during a soccer match (Vigne et al. 2010), the acceleration phase becomes crucial. Propulsion forces have a key role in sprint performance during the acceleration phase and are also significant for executing changes of direction (von Lieres and Wilkau et al. 2020).

The plyometric training is recognized for its ability to enhance muscle power necessary for explosive movements (Wilson and Flanagan 2008). However, the main focus of plyometric training protocols is primarily on exercises with a dominant vertical component, while the benefits on motor task involving a significant horizontal component remain contrasted. For instance, seven weeks of vertical plyometric training have improved countermovement jump performance, but did not result in any significant changes in 20 m sprint time (Ramirez-Campillo et al. 2014). Considering that 95% of soccer actions involves horizontal force production (Young 2006), the use of horizontal plyometric exercises could be particularly advantageous for enhancing sprint performances (Saez de Villarreal et al. 2012).

While some studies comparing the effects of vertical and horizontal orientations in plyometric training exercises have found no significant difference in the ability to change direction, as well as jump and sprint performances (Ramirez-Campillo et al. 2015), others studies have demonstrated specific benefits (Loturco et al. 2015; Dello Iacono et al. 2017). Horizontal plyometric training has been shown to enhance sprint performance and changes of direction, whereas vertical plyometric training has led to improvements in jump performances (Loturco et al. 2015; Dello Iacono et al. 2017). It is noteworthy that the specific influence of the relative orientation of horizontal force in the anterior and lateral direction during plyometric exercises on linear sprint performance and change of direction remains underexplored.

While sometimes prone to over-interpretation (Ettema 2024), force–velocity profile (FVP) allows to qualitatively estimate athletes’ behavior during a sprint (Samozino et al. 2022), particularly in younger soccer players (Zhang et al. 2021; Fernandez-Galvan et al. 2021). The FVP can also evaluate the plyometric training-induced effects on the changes in sprint performance and athletes’ behavior during a linear sprint (Barrera-Dominguez et al. 2023; Watkins et al. 2021). However, a comparison between horizontal and vertical plyometric training on sprint FVP has yet to be conducted. These findings highlight the importance of accurately assessing the specific benefits of horizontal and vertical plyometric training, particularly in young soccer players, with regard to sprint FVP.

The aim of this study was to evaluate and compare the effects of 8-week vertical and horizontal plyometric training on jump performances and sprint FVP in young soccer players. Through the assessment of sprint FVP, the primary hypothesis suggesting that horizontal plyometric training would be more effective than vertical training in improving sprint performance and the ability to change direction in highly skilled young soccer players was tested.

Materials and methods

Study population

Twenty-eight participants, soccer players at the highest regional level, were recruited between December 2020 and April 2021. They were separated in two trained groups performing vertical (VG | N = 14 | age: 14.5 ± 0.5 y [14—15], height: 170 ± 7 cm [153—176], body mass: 57 ± 9 kg [40 – 68]) and horizontal (HG | N = 14 | age: 14.5 ± 0.5 y [14—15], height: 167 ± 7 cm [153—175], body mass: 56 ± 9 kg [40—69]) plyometric training. A reproducibility group composed of young soccer players with similar anthropometric characteristics was also recruited (N = 14) to assess the measurements variability over a nine weeks period (Table 1). The number of participants was chosen based on a statistical power calculation (α = 0.05 and 1-β = 0.8) considering the variability of measurement observed in pilot experiments and reproducibility results. On that basis, a medium effect size (η2p = 0.06) after the training can be robustly detected. Our study was approved by the local ethical research committee and participants were fully informed about the nature and the aim of the study. They all gave their informed written consent to participate in this study conducted in conformity with the last version of the Declaration of Helsinki. Participants were included on the basis of the lack of affection after clinical interrogation, with no known musculoskeletal, articular, or cardiovascular dysfunction or abnormalities, chronic or central neurological disease and medication.

Table 1 Reproducibility results obtained between two tests over a 9-week period for sprint performances and force–velocity parameters, and jump performances
Full size table

Experimental design

Pre-tests and post-tests were set up to assess the effects of an 8-week plyometric training program (i.e., vertical or horizontal), with 2 sessions per week in a progressive manner (Table 2). Participants were tested in a single session in several tests performed in a randomized order: (i) vertical and horizontal jump tests determining performances in squat jump (SJ), counter movement jump (CMJ), drop jump with a 30 cm drop height (DJ) and a horizontal five jumps (5-jumps); (ii) a force–velocity profile (FVP) assessed from a 30 m sprint (iii) a change of direction test on a short distance running with 4 direction changes. All of the trained participants performed the test session before (pre-test) and 1 week after (post-test) the end of the training period. The reproducibility of the measurements was assessed with two similar tests separated by a 9-week period.

Table 2 Plyometric trainings in vertical and horizontal groups.
Full size table

Training protocol

During the 8 week plyometric training, both trained groups (i.e., VG and HG) performed 2 sessions per week in a progressive manner for a total of 2000 vertical or horizontal contacts (Table 2). All the exercises were performed with the instruction to have the shortest ground contact time. The vertical plyometric training consisted of hurdle jumps at different heights (i.e., 15, 30 and 45 cm) and vertical CMJ. Vertical jumps were performed on two feet (i.e., hurdle jumps and CMJ) and on a single leg for unilateral hurdle jumps exercise (Table 2). To avoid an excessive horizontal component, the hurdles were spaced 30 cm apart for each exercise to promote vertical rather than horizontal displacement during the jumps. The horizontal plyometric training was composed of bouncing strides (i.e., anterior jump from one foot to the other) and lateral jumps (i.e., lateral jump from one foot to the other). Unilateral bouncing strides were performed on a single leg. The distance between jumps increased progressively (i.e., from 60 to 270 cm).

Jump performances

A warm-up was based on athletic routines including skipping, heel-buttocks, step-ups, proprioception movements and lunges. For all jump tests, hands were placed on hips.

Vertical SJ, CMJ and DJ performances were assessed from time flight measurements using an Optojump (Optojump Next, Microgate, Bolzano, Italy). Initial SJ position was knee flexed at 90° and was performed with no countermovement. During the vertical DJ (i.e., 30 cm high), both maximal vertical height and ground contact time (CT) were evaluated. A reactive strength index (RSI) was then calculated from the ratio between jump height and ground contact time (Haynes et al. 2019).

Horizontal SJ, CMJ and DJ performances were assessed from My Jump (v.6.1.2) application (Balsalobre-Fernandez et al. 2015). Horizontal jump videos were acquired at 29.98 frames/s. An additional 5-jumps test was also performed and measured with a decameter tape with the back of the foot as a reference. CT and RSI were also assessed from horizontal DJ.

Change of direction

The change of direction performance was evaluated from a specific test (Cazorla et al. 2008) (i.e., 20 m distance with four changes of direction) using photoelectric cells (Brower TC PhotoGate, Brower Timing Systems, Draper, United States of America).

Sprint force–velocity profile

A warm-up was performed including straight line sprints with increasing intensity, slalom workshops, and active dynamic stretching. Each test was performed three times by each participant and the best achievement was considered for analyses.

From a linear 30 m sprint, times at 5, 10, 15 and 30 m were quantified from My Sprint (v.2.0) application (Romero-Franco et al. 2017) and FVP was assessed for each participant, from a theoretical F-V relationship modeling of the 30 m linear sprint (Samozino et al. 2022). The maximum force (F0), maximum velocity (V0), maximum power (Pmax), maximal ratio of force (RFmax, maximal ratio of the step-averaged horizontal component of the ground-reaction force to the corresponding resultant force) and the decrease rate in RF with increasing speed (DRF, slope of the linear RF-velocity relationship) were characterized (Morin and Samozino 2016; Samozino et al. 2022).

Statistical Analyses

The normality of the data distributions was checked using Shapiro–Wilk test. The reproducibility results were reported in Table 1 with coefficient of variation (CV) and minimal detectable change (MDC). A two-way repeated measures analysis of variance (group × time) was used to determine the effects of the chronic intervention on jump performances, FVP parameters and change of direction time in both vertical and horizontal groups. Tukey HSD test was used as post-hoc tests when appropriate. Mauchly’s sphericity test was performed and Greenhouse–Geisser correction was applied when necessary. A student t-test was also used to assess the change in performance in each group and to compare the relative changes in all the variables between the trained groups. When the normality of the data distribution was not verified (i.e., for anthropometric variables), a non-parametric Kruskal–Wallis or Friedmann test was performed. All the tests were performed using R-Commander software (2.6–0, GNU General Public License, Hamilton, Canada). Effect size was also assessed from partial eta squared (η2p) for the significant results retrieved from ANOVA and using Cohen’s d for post-hoc analysis and Student t-test comparison. Small, medium and large effect sizes were reported at the threshold values of 0.01, 0.06 and 0.14 for η2p and 0.2, 0.5 and 0.8 for Cohen’s d (Cohen 1969). The p value was fixed at 0.05 and the results were reported as mean ± standard deviation (SD) in tables and figures.

Results

Jump performance

Both VG and HG groups improved performance in SJ (VG: + 8.7% | HG: + 10.2%), CMJ (VG: + 5.0% | HG: + 5.9%) and DJ (VG: + 11.5% | HG: + 7.5%) with no difference between VG an HG (p > 0.05). No difference was observed for CT and RSI during the vertical DJ (p > 0.05) (Table 3).

Table 3 Vertical jump performances before (Pre-test) and after (Post-test) the 8-week plyometric training for the vertical (VG) and horizontal (HG) groups
Full size table

Horizontal performances were improved in CMJ (VG: + 5.6% | HG: + 8.6%) and DJ (VG: + 5.8%, p = 0.003, d = 0.576 | HG: + 9.8%, p = 0.001, d = 1.127). Horizontal SJ performance was significantly improved in HG (+ 8.6%, p = 0.001, d = 0.868) as well as 5-jumps length (+ 2.3%, p = 0.016, d = 0.690). CT and RSI determined during the horizontal DJ were unchanged after the training period for the two experimental groups (p > 0.05, Table 4). Relative changes after the training period were significantly higher in HG than in VG in horizontal SJ (p = 0.043, d = 0.670) and horizontal DJ (p = 0.045, d = 0.655) performances (Fig. 1).

Table 4 Horizontal jump performances before (Pre-test) and after (Post-test) the 8-week plyometric training for the vertical (VG) and horizontal (HG) groups
Full size table
Fig. 1
figure 1

Significant mean relative changes between pre- and post-tests for A the vertical group and B the horizontal group in horizontal jump performances (squat jump [SJ], countermovement jump [CMJ] and drop jump [DJ]), sprint performances (sprint time at 5 m, 10 m, 15 m and 30 m) and force–velocity profile parameters (including the maximal force [F0], the maximal power [Pmax], the maximal force ratio [RFmax] and the decrease rate in RF with increasing speed [DRF]). ns no significant changes in vertical group or horizontal group, a significant difference between relative changes assessed in VG and HG (p < 0.05)

Full size image

Change of direction performance

No significant change was found in the change of direction performance for both VG and HG (p > 0.05, Table 5).

Table 5 Performances in 30-m sprint and change of direction test before (Pre-test) and after (Post-test) the 8-week plyometric training for the vertical (VG) and horizontal (HG) groups
Full size table

30-m sprint force–velocity profile

Sprint times were decreased in both VG and HG after 5 m (VG: + 6.7% | HG: + 12.4%), 10 m (VG: + 5.6% | HG: + 8.2%), 15 m (VG: 4.7% | HG: + 9.1%) and 30 m (VG: + 4.8% | HG: + 6.0%) (Table 5). The relative decrease in sprint time was significantly higher in the HG than in VG at 5 m (p = 0.010, d = 0.942) and 15 m (p = 0.001, d = 0.717) (Fig. 1).

Concerning the FVP parameters, a significant interaction was found for RFmax (p = 0.037, η2p = 0.157). Both groups significantly improved F0 (VG: + 13.1%, p = 0.003, d = 1.139 | HG: + 26.5%, p = 0.0003, d = 1.526 | Fig. 2A), Pmax (VG: + 15.9%, p = 0.001, d = 1.304 | HG: + 27.0%, p = 0.001, d = 1.701 | Fig. 2C) and RFmax (VG: + 7.2%, p = 0.001, d = 1.297 | HG: + 13.5%, p = 0.001, d = 2.142 | Fig. 2D). DRF was significantly improved in HG (+ 28.6%, p = 0.003, d = 1.168 | Fig. 2E). No significant change was found for V0 neither in VG nor in HG (p > 0.05, Fig. 2B). A significant difference was found in relative change between VG and HG for Pmax (p = 0.028, d = 0.754), DRF (p = 0.037, d = 0.701) and RFmax (p = 0.022, d = 0.796).

Fig. 2
figure 2

Parameters of the force–velocity profile including A the theoretical maximal force (F0), B the theoretical maximal velocity (V0), C the theoretical maximal power (Pmax), D the maximal force ratio (RFmax) and, E the decrease rate in RF with increasing speed (DRF) before (Pre-test) and after (Post-test) the plyometric training period for the trained vertical group (VG) and horizontal group (HG). asignificantly different from Pre-test

Full size image

Discussion

The aim of this study was to compare the effects of 8-week horizontal and vertical plyometric training programs on explosive qualities in young soccer players. Both types of plyometric training, regardless of orientation, led to improvements in sprint and jump performances. However, relative gains in horizontal performances (i.e., jumps, sprint times and FVP parameters) were significantly higher in the horizontal plyometric group than in the vertical one while improvements in vertical performances were similar between the two groups.

The jump performances (i.e., vertical and horizontal) and sprint results evaluated in the young soccer players included in the present study fall within the range as those reported in previous studies involving young athletes (Keller et al. 2020; Ramirez-Campillo et al. 2015). The values of the FVP parameters reported in young soccer players of the present study are also comparable to those documented in earlier investigations (Fernandez-Galvan et al. 2021; Zhang et al. 2021). The observed improvements in vertical jumps (e.g., SJ, DJ) and sprint performances after 8-weeks of plyometric training assessed are in accordance with findings from previous studies involving plyometric training durations ranging from 6 to 10 weeks (Spurrs et al. 2003; Fouré et al. 2009; Chelly et al. 2014; Moran et al. 2021).

Only a few studies have conducted comparisons between the two orientations of plyometric training (Moran et al. 2021). Previous research examining younger soccer players who underwent 6 week of vertical or horizontal plyometric training found trivial or small differences in vertical and horizontal jump and sprint performances (Ramirez-Campillo et al. 2015). In elite handball players, a horizontal DJ protocol was shown to be more effective than a vertical DJ protocol, as evidenced by improved horizontal performance in 10 m sprint time and change of direction test (Dello Iacono et al. 2017). In the present study, both trained groups exhibited improvements in vertical SJ, vertical CMJ, and both vertical and horizontal DJ performances. However, only the HG demonstrated enhancements in horizontal SJ performances and improvements in horizontal DJ were larger in HG than in VG. Furthermore, a significantly greater relative decrease in 5 m and 15 m sprint times was observed in HG. Consequently, horizontal-oriented plyometric training resulted in improved performances in horizontal tests (i.e., sprint and horizontal jumps). However, no improvement in sprint times involving changes of direction was observed for either the VG or the HG. This lack of improvement may stem from the comparatively low emphasis on lateral movements during horizontal training (i.e., only 28% of the 2000 jumps) as well as the absence of lateral displacement in the vertical plyometric exercises.

Concerning FVP parameters assessed during a linear sprint, a single study has examined the short-term and combined effects of horizontal and vertical plyometric training protocols (Watkins et al. 2021). However, it remains challenging to isolate the potential impact of orientation from that study, given its comparatively low volume (i.e., 40–60 ground contacts per session) and short duration (i.e., 3 weeks) for each training intervention. In the present study, a significant effect of both vertical and horizontal training was observed for F0, Pmax and RFmax. Higher relative changes in RFmax and Pmax were found for HG compared to VG. Moreover, a significant decrease in DRF was only observed in HG. Changes in functional performances (i.e., jump and sprint) following plyometric training can be attributed to increased force capacities, primarily resulting from nervous adaptations such as a decreased inhibition of the myotatic reflex (Avela et al. 1999) an improved activation of agonist muscles and intermuscular coordination (Markovic and Mikulic 2010) as well as structural and mechanical adaptations (Spurrs et al. 2003; Fouré et al. 2009, 2010, 2011; Burgess et al. 2007; Wu et al. 2010) contributing to improve force production and transmission (Fouré et al. 2010; Wu et al. 2010; Burgess et al. 2007). The observed force gains may have contributed to improvements in FVP parameters, particularly in F0, Pmax, and RFmax. The improvements in horizontal jumps, linear sprint time, and DRF change during the 30 m sprint observed in the HG can be attributed, in this context of repeated measurements in the same individuals without changes in body weight after the training period, to a more efficient application of ground force (Samozino et al. 2022). A recent meta-analysis demonstrated that horizontal plyometric training yields superior performance on horizontally-oriented motor tasks compared to vertical plyometric training (Moran et al. 2021). The results obtained in the present study support the conclusions of the aforementioned analysis and highlight the significant influence of the horizontal plyometric training on the FVP determined during sprints. Regarding potential practical application, both horizontal and vertical plyometric training can be utilized to enhance both vertical and horizontal performances in jumps and sprints among young soccer player. However, it is noteworthy that the relative changes in horizontal jump and sprint performances, as well as force–velocity parameters, were more pronounced in the group undergoing horizontal plyometric training compared to the vertical plyometric training group.

It should be pointed out that comparing the training load between horizontally- and vertically-oriented plyometric exercises in the current study is challenging without assessing force and velocity during these exercises. Consequently, potential differences in motor task force and velocity characteristics may have significantly influence the results. Furthermore, data regarding muscular activity during both vertical and horizontal plyometric exercises, as well as testing tasks, would have been highly valuable for comparing coordination differences between the two orientations of the plyometric training (Jacobs and van Ingen Schenau 1992).

Lastly, it has been noted that the development of physical qualities is correlated with biological maturation in young soccer players (Mendez-Villanueva et al. 2011). Hence, the training-induced adaptations observed in the present study may have been more readily achieved in the population of young soccer players compared to a more general population.

Limitations

Regarding the methodological limitations related to the present study, the inclusion of FVP assessment during vertical jump could have been highly valuable in determining the potential significant and specific influence of vertical plyometric training on vertically-oriented motor task, despite no significant relative gains for the VG regarding vertical jump performances in comparison to the HG. Additionally, no electromyographic data collection during jumping and agility tests have been undertaken. This kind of measurements could provide insight into the muscular synergies involved in executing precise movement (Hug et al. 2010), as well as the impact of flexibility quality in motor coordination (Rodriguez Fernandez et al. 2016).

Perspective

Further studies are necessary to evaluate the potential specific effects of longer-duration horizontal and vertical training protocols, particularly regarding musculoskeletal structural adaptations. Additionally, the comparison of directions (i.e., lateral and anterior) of horizontal plyometric exercises needs to be further investigated.