**The Acute E**ff**ect of Match-Play on Hip Isometric Strength and Flexibility in Female Field Hockey Players**

#### **Violeta Sánchez-Migallón 1, Alvaro López-Samanes 1,\*, Pablo Terrón-Manrique 1, Esther Morencos 2, Vicente Fernández-Ruiz 1, Archit Navandar <sup>3</sup> and Victor Moreno-Pérez <sup>4</sup>**


Received: 22 June 2020; Accepted: 15 July 2020; Published: 17 July 2020

**Abstract:** The aim of this study was to determine the acute effect of simulated field hockey match-play on isometric knee flexion, adductor (ADD) and abductor (ABD) strength, adductor/abductor (ADD/ABD) strength ratio, countermovement jump height (CMJ), hip flexion and ankle dorsiflexion range of motion (ROM). Thirty competitive female field hockey players (23.0 ± 3.9 years old) participated in the study. Apart from the afore-mentioned variables, external (through GPS) and internal load (through RPE) were measured before (pre-match) and immediately after simulated hockey match-play (post-match) in both limbs. Isometric knee flexion strength (+7.0%, *p* = 0.047) and hip flexion ROM (+4.4%, *p* = 0.022) were higher post-match in the non-dominant limb, while CMJ values reduced (−11.33%, *p* = 0.008) when comparing from pre-match data. In addition, no differences were observed for isometric hip ADD, ABD, ADD/ABD strength ratio, passive hip flexion ROM and ankle dorsiflexion ROM test. A simulated field-hockey match produces an increment in hip isometric strength and hip flexion ROM values in the non-dominant limb and a decrease in jump height capacity. As a result, CMJ assessment should be considered post-match in order to identify players who would require further rest before returning to training.

**Keywords:** risk factors; performance; team sport; fatigue; groin; hamstring

#### **1. Introduction**

Field hockey is an intermittent sport where hockey players perform repeated actions such as changes of direction, dribbles, sprints, accelerations, decelerations and body impacts alternating high and moderate with low intensity efforts [1]. During an official field hockey match, consisting of four quarters of fifteen minutes each, hockey players cover around 6000–8000 m [2,3] primarily at low and medium intensities, with high-intensity efforts (>19 km/h<sup>−</sup>1) making up around 6% of the total playing time [2]. Previous studies have reported an average of 14 to 48 injuries per 1000 h attributed to the high physical demands of this sport [4,5]. Specifically, most of these field hockey injuries have been reported in the lower limbs, especially in the thigh and groin [6], with the hamstring strain injury being the most frequent muscle injury (32%) [4], followed by the groin injury (10%) [6]. Consequently, identification of the risk factors associated with groin and/or hamstring injury occurrence is essential.

In this sense, previous studies in different intermittent sports such as football [7], tennis [8] and ice hockey [9], have identified several modifiable intrinsic risk factors as causing an increased likelihood of developing groin and hamstring injuries. Among them, a weakness in the isometric adductor strength (ADD) [7,10] and lower adduction/abduction strength ratio (ADD/ABD) [9] have been associated with a higher risk of sustained groin injuries, while a lower hamstring strength [11–13], decreased range of hip flexion [14] and ankle dorsiflexion range of motion (ROM) have been associated with hamstring strain injuries [15]. However, some conflicting results have been found in literature regarding these factors [16–18]. Notably, most researches have investigated these risk factors before the commencement of the season or in off-season situations [19]. However, the ability to capture fluctuations in ROM and/or strength profile in-season, specifically in response to match-play, has not been studied [19].

Similar to other intermittent sports, field hockey players reported a higher incidence of injuries during matches compared to training [6] probably due to the higher intensity reported in matches versus training, and the appearance of fatigue [20,21]. It is well known that there is a decrement of lower limbs' power performance after match-play, and this measure is one of the potential factors in injury causation in intermittent sport [22]. Appearance of fatigue is known to reduce sports performance through reduced muscle strength, neuromuscular control and ROM [22]. In this line, the impact of match-play on ADD, ABD, hamstring strength and ROM during hip flexion and ankle dorsiflexion has been studied recently in several sports such as tennis, football and basketball [19,23–25]; however, to the best of the authors' knowledge, there is no information regarding this effect in field hockey. Therefore, the aim of this study was to examine the acute effect of hockey match-play on several risk factors such as isometric knee flexion and hip ADD and ABD strength, ADD/ABD strength ratio, passive hip flexion ROM and ankle dorsiflexion ROM and countermovement jump (CMJ) in elite female hockey players.

#### **2. Materials and Methods**

#### *2.1. Subjects*

Thirty highly competitive female hockey players (age, 23.0 ± 3.9 years; body mass, 60.0 ± 7.5 kg; height, 1.60 <sup>±</sup> 0.09 m; body mass index, 22.0 <sup>±</sup> 2.1 kg·m2; hours per week, 9.4 <sup>±</sup> 4.4; playing experience, 14.3 ± 4.9 years) volunteered to participate in this investigation. The players were recruited from two different professional teams. The inclusion criteria were: (a) To be healthy and able to complete a full game of field hockey; (b) to be uninjured and declared match-fit by the medical and coaching staff at the time of the experiment and not to have taken any type of medication to treat pain or musculoskeletal injuries at the time of the study; and (c) to have an absence of late onset muscle pain during the training session [26]. All players were informed of the tests they were to perform and signed the consent form. The experimental procedure of this study was conducted in accordance with the Declaration of Helsinki and the approval by the Ethics Committee of the University Francisco de Vitoria, number 45/2018.

#### *2.2. Experimental Protocol*

Following their arrival, female hockey players filled out a questionnaire which included personal information such as body mass, height, medical history, training frequency and experience (practice hours per week and playing experience in field hockey). Testing (i.e., ROM, isometric strength, and countermovement jump) was performed in the clinical area of each field hockey club. Testing was conducted by two sports physiotherapists: a senior physiotherapist with over nineteen years of experience and a junior physiotherapist with two years of experience, to ensure participants' positioning during measurements. Considering recommendations by Wollin, Thorborg and Pizzari [19], the testing order of the players and the selection of the leg tested were randomly chosen prior to the pre-match test. Pre-match testing was performed 60 min prior to match-play, and the post-match re-testing was performed immediately after the match. At the beginning of the pre-match testing, participants carried out a standardized warm-up that consisted of 5 min of jogging at 10 km·h−<sup>1</sup> and 5 min of

static stretches and joint mobility exercises [27]. Subsequently, participants played the simulated field hockey match according to the International Hockey Federation rules on a rectangular surface, 91.40 m long and 55.00 m wide. The external load of the simulated matches were estimated using a global positioning system (GPS) (Wimu ProTM, RealTrack Systems, Spain) placed in specific vests worn by the players, these devices operated at a sampling frequency of 10 Hz and its validity and reliability have been reported previously [28]. In addition, subjective internal load of the game was obtained using the modified RPE scale (i.e., 0–10 points) within 30 min of match termination [29]. The following variables were used to assess the external load during match-play, total distance covered per minute at different velocities ranges during a 60 min match as previously reported [3]. To reduce the interference of uncontrolled variables, all subjects were instructed to maintain their habitual lifestyle and normal dietary intake before and during the study, and refrain from caffeine ingestion 24 h before the experiment [30].

#### *2.3. Isometric Strength of Abductors (ABD) Adductors (ADD) and Knee Flexion*

Hip isometric ADD and ABD strength were measured according to the methodology previously reported [31] using a portable handheld dynamometer (Nicholas Manual Muscle Tester; Lafayette Indiana Instruments, Lafayette, IN, USA). Participants were placed in a supine position with their hips in a neutral position and told to stabilize themselves by holding onto the sides of the table. Examiner 1 applied a resistance on a fixed position (ABD: At 5 cm proximal to the lateral malleolus; ADD: At 5 cm proximal to the medial malleolus). The hockey players were instructed to exert a voluntary contraction for a maximum of 5 s against the dynamometer [31]. Two attempts were registered for each contraction of each limb and a 30 s rest period between attempts. Regarding the isometric knee flexion, the strength test was evaluated by placing the subject in the prone position, with 15 degrees of knee flexion and with their hips in a neutral position [32]. Examiner 1 placed the dynamometer on the distal portion of the sural triceps, three centimeters above the bimalleolar line. Examiner 2 clamped the subject's pelvis over the sacrum, to prevent elevation during the test. Examiner 1 requested the participant to flex their knee with the intention of bringing the heel of the foot to the gluteus. Similarly, two repetitions were recorded for each limb with a 30 s rest between attempts. Isometric hip ADD, ABD and knee flexion strength was expressed as the maximal hip and knee torque per kilogram of body weight (Nm·kg<sup>−</sup>1) using the external lever arm and body weight of each participant. The mean value out of two attempts was recorded and selected for further analysis.

#### *2.4. Ankle Dorsiflexion ROM*

Unilateral ankle dorsiflexion ROM was measured with LegMotion System (LegMotion, Check your Motion, Albacete, Spain). The testing was carried out following the methodology previously described by Calatayud et al. [33]; participants were in a standing position on the LegMotion System with the tested foot on the measurement platform and the contralateral foot out of the platform with the toes at the edge of it. Each player performed the test with their hands on the hips and the assigned foot in the middle of the longitudinal line behind the transversal line of the platform. From this position, subjects were instructed to flex the knee forwards, placing it in contact with the metal stick. When the subject was able to maintain heel and knee contact, the metal stick was progressively moved away from the knee, and the following achieved distance was recorded. Two attempts were allowed for each limb (i.e., left and right), with 15 s of passive recovery between trials. The mean value of the two attempts was selected for further analysis.

#### *2.5. Hip Flexion ROM*

Passive hip flexion ROM values with the knee extended were evaluated with the Straight Leg Elevation Test (SLET). Participants made two maximum passive attempts for the dominant and non-dominant leg, when the difference between one attempt and another was greater than 5%, a third attempt was made, selecting the mean value of the two attempts whose results were similar for further statistical analysis [34]. A unilevel inclinometer ISOMED (Portland, OR, USA) with a telescopic was used for the measurement. The test ended with one or more of the following criteria: (a) The examiner was unable to continue the joint movement evaluated due to the high resistance developed by the stretched muscle group; (b) The participant reported an important sensation of discomfort; or (c) The examiners noted compensations that could increase the ROM [35]. The inclinometer was placed approximately on the external malleolus and the distal arm was aligned parallel to an imaginary bisecting line of the extremity [35]. The mean value out of two attempts was recorded and selected for further analysis.

#### *2.6. Countermovement Jump (CMJ)*

Participants carried out three repetitions of CMJ using a contact mat jump system (Chronojump Boscosystem, Barcelona, Spain) with their arms on hips [36]. They were instructed to jump and land in the same place, with the body in an erect position during the jump until landing. Each participant performed two maximal CMJs interspersed with 45 s of passive recovery. In addition, the mean value out of two attempts was recorded and selected for further analysis.

#### *2.7. Statistical Analysis*

Data were calculated as means/standard deviation. The Shapiro–Wilk test was selected to assess the normal distribution. All study variables were compared using a *t* test (pre- vs. post-match). The statistical significance level was set at *p* < 0.05. Cohen's effect sizes were calculated and presented with their respective 95% confidence intervals (C.I.) based on the following criteria: Trivial effect (0–0.19), small effect (0.20–0.49), medium effect (0.50–0.79) and large effect (0.80 and greater) [37]. All the statistical analyses were completed using the SPSS software version 25 (SPSS Inc., Chicago, IL, USA).

#### **3. Results**

#### *3.1. Match-Play Workload*

The internal match-play workload was 6.83 ± 0.80 units (RPE). In addition, female hockey players covered a mean distance of 5456.50 ± 699.09 m across different velocity profiles (Table 1).


**Table 1.** Mean distances and % of total distance covered during the match at different velocity ranges.

Abbreviations: km·h−<sup>1</sup> = kilometers/hour; m = meters.

#### *3.2. Isometric Strength and Countermovement Jump*

No statistical differences were seen in relative isometric hip ABD strength in the dominant (+2.42%, *p* = 0.864, ES [C.I.] = 0.01 [−0.08, 0.11]) (Figure 1a) and non-dominant limb (−1.46%, *p* = 0.834, ES [C.I.] = −0.02 [−0.12, 0.07]) (Figure 1b); nor in the relative isometric hip ADD strength in the dominant(−2.10%, *p* = 0.399, ES [C.I.] = 0.10 [0.00, 0.19]) (Figure 1c) and non-dominant limb (+3.38%, *p* = 0.349, ES [C.I.] = 0.11 [0.02, 0.21]) (Figure 1d). In addition, no differences were obtained in ADD/ABD strength ratios in dominant (1.14 vs. 1.19, *p* = 0.220, ES [C.I.] = 0.28 [0.19–0.37]) and non-dominant limbs (0.98 vs. 0.96, *p* = 0.600, ES [C.I.] = 0.14 [0.05–0.24]) when comparing them pre and post-match. However, for isometric knee flexion strength, statistical differences were obtained in the non-dominant

limb (7.0%, *p* = 0.047; ES [C.I.] = 0.29 [0.20, 0.38]) (Figure 1f) but no differences were reported in the dominant limb (0.1%, *p* = 0.983; ES [C.I.] = 0.11 [0.02, 0.21]) (Figure 1f). Finally, neuromuscular fatigue was measured by a countermovement jump test after match-play (23.0 ± 4.9 vs. 20.5 ± 6.6 cm, *p* = 0.008, ES [C.I.] = 0.44 [0.34, 0.53]).

**Figure 1.** Hip and knee isometric hip abduction (ABD), adduction (ADD) and knee flexion strength values. (**a**) Relative isometric hip ABD strength in the dominant limb; (**b**) relative hip abductor strength in the non-dominant limb; (**c**) relative hip adductor strength in the dominant limb; (**d**) relative hip adductor strength in the non-dominant limb; (**e**) relative knee flexion strength in the dominant limb; (**f**) relative knee flexion strength in the non-dominant limb.

#### *3.3. Hip Flexion and Ankle ROM*

A significant increase was found when comparing the ROM in the hip flexion (straight leg elevation raise test) for the non-dominant limb (+4.38% *p* = 0.022). However, no differences were found in the dominant limb (+1.19%, *p* = 0.753) (Table 2). In addition, no differences were obtained for ankle dorsiflexion ROM values after field hockey match in the dominant limb (−3.77%; *p* = 0.316) and non-dominant limb (−2.34%; *p* = 0.362) (Table 2).



Abbreviations: DOM = dominant-side; NO-DOM = non-dominant side; ◦ = degrees; cm = centimeters; \* Significant differences compared to the PRE values at *p* < 0.05.

#### **4. Discussion**

The aim of this study was to determine the acute effect of hockey match-play on several risk factors such as isometric knee flexion, hip ADD and ABD strength, ADD/ABD strength ratio, passive hip flexion ROM, ankle dorsiflexion ROM and CMJ in competitive female hockey players. To the best of the authors' knowledge, this is the first study that analyzed the acute effects of hockey match-play on several risk factors in female athletes. The main results showed that hockey match-play acutely produced a decrease in CMJ performance, and an increase in isometric knee flexion strength and hip flexion ROM with knee extension in the non-dominant limb. However, no significant differences were found in isometric hip ADD, ABD strength and ADD/ABD strength ratio and ankle dorsiflexion ROM in both limbs.

The CMJ is one of the most important tests used to evaluate the lower-limb muscles fatigue [38,39]. The results in this study showed a significant reduction in levels of performance in CMJ (−11.33%) after match-play which was in agreement with previous studies conducted in other intermittent sports [40,41]. Recent research from Kim and Kipp [42] has shown that the gastrocnemius, soleus and vastus muscles have the largest contribution to vertical center of mass (COM) acceleration during the CMJ, and the soleus and gastrocnemius muscles function closest to their maximal capacities. If one were to look at the distances covered by the players (Table 1), one can observe that most of the distance had been covered at lower intensities. Here, the production of the horizontal force has been attributed to the muscles of the lower limb: namely tibialis anterior, gastrocnemius and soleus [2]. This indicates that a greater fatigue caused by distances ran at these intensities appears to have increased the neuromuscular fatigue associated with the CMJ, leading to a decrease in performance. In addition, albeit speculative, another possible reason for the decrease values in the CMJ test after match-play has been attributed to disruptions within the muscular fibers [41], increasing some markers of muscle damage (e.g., creatine-kinase, myoglobin) after a match in intermittent sports [41,43]. Moreover, previous literature described the inflammatory responses and fibrillar damage to the muscles and showed a decrease in metabolic indices in athletes after playing a match; however, this speculation requires further investigation.

Muscle strength in the lower limbs is essential to produce explosive actions in hockey (e.g., accelerations, changes of direction). The results obtained in the present study showed improvements in the isometric knee flexion strength (+7.0%) immediately post-match in the non-dominant limb. While no previous study has reported knee flexion strength values in field hockey players, these results differ from previous studies conducted in other intermittent sports, which revealed a lower knee flexion strength post-match-play [19,44]. The lack of agreement between studies could be related to the different match-play demands of the participants (e.g., total distance covered, duration of match, etc.). While in the current study hockey players reported an average of 5456.50 ± 699.09 m total distance covered and 21.59 <sup>±</sup> 23.66 m at high speeds over 21 km·h<sup>−</sup>1, previous studies in female soccer players showed they covered distances over 10,000 m during a match (90 min duration), of which at least 600 m were at high speed running intensities [45]. The current results suggest that the effect provoked by field-hockey match-play did not decrease the isometric knee flexion strength in the dominant limb, which has been related to a higher hamstring injury risk in previous studies [46,47]. The absence of changes in the isometric knee flexion strength can be attributed to the lower distances covered at high running speeds. Extensive research [47] has shown that the hamstring muscles are most active in this phase, when their function is to increase stride frequency and produce a greater horizontal force as the contact time reduces. Given the distances covered in the different zones, it can be assumed that the hamstring muscles are not fatigued as in sports such as soccer, where a greater distance is covered at these intensities.

A reduced ROM during the straight leg raise test and dorsiflexion of the ankle ROM has been linked to the risk of hamstring injury [14,15]. Present results have shown an increase in the straight leg raise ROM levels for the non-dominant limb (+5.81%) and no significant differences in the dominant limb. No differences were seen for the ankle dorsiflexion ROM either for both legs after match-play. To the best of the authors' knowledge, only Wollin et al. (2017) analyzed the straight leg raise ROM after match-play on intermittent sports and report no significant differences after match-play [19]. Concerning ankle dorsiflexion ROM, some previous studies conducted in football [24], basketball [25] and Australian-rules football [44] showed different results. While Charlton et al. [44] reported reductions in ankle dorsiflexion ROM immediately after an Australian-rules football match, Wollin et al. [19] showed non-significant decrements in football players, and finally Moreno-Pérez et al. [24,25] in

football and basketball observed increased ROM values post-match from pre-match in dominant and non-dominant limbs. While these sports involve the same multidirectional movements (e.g., accelerations, decelerations, changes of direction) during practice, they cannot be compared due to the differences in the characteristics of the sport; for example, the total duration is less in field hockey than other intermittent sports. Thus, increases in hip flexion ROM immediately post-match are likely due to the increase in tissue extensibility induced by temperature increment which leads to a reduction of the viscous resistance of muscle tissues and joints [48,49]. However, this is a speculation that needs to be elucidated in future studies.

As far as isometric hip ADD and ABD strength values and ADD/ABD strength ratio go, the results showed no significant differences between post-match from pre-match. While is difficult to establish comparisons, as no previous study has reported isometric hip ADD and ABD strength values in hockey players, the findings of the current study were similar to those of a previous study [50] conducted with 14 rugby players. However, the present results disagree with the results reported by a previous study in tennis players [23]. This lack of agreement between studies may be due to differences in physical demands and tactical aspects between the sports. In tennis, players are required to perform repetitive short high-intensity movements, which impose an elevated concentric and eccentric load on the ADD muscles, while hockey players perform multiple different movements at several intensities. One must also consider that hockey, like in rugby, permits rolling substitutions.

This study contains limitations that require acknowledgment. Firstly, as the current study was conducted with a specific sample of female field hockey players, conducted during a simulated match, the characteristics of the players do not permit a generalization of the results found. In addition, the selected measures in this study in response to hockey match-play were performed immediately post-match; future studies should evaluate these variables using several time-points to understand the recovery fatigue induces, for example registering data 48 h after the match. This study looked at the effect following a single simulated match; more data collected over a series of official matches, or even, over an entire season could provide more conclusive results about the effects of match-play on these potential risk factors.

#### **5. Conclusions**

A simulated field-hockey match increases hip isometric flexion strength and hip flexion ROM in the non-dominant limb and decreases jump height capacity. However, no differences were reported in isometric ADD and ABD strength and ADD/ABD strength ratio in the dominant and non-dominant limb, or in dominant hip flexion and ankle dorsiflexion ROM values between the pre- and post-match examinations. Finally, female hockey players who present a decrease in jump height capacity, may require additional rest between training and competitions.

**Author Contributions:** Conceptualization, V.S.-M., A.L.-S. and V.M.-P.; methodology, A.L.-S. and V.M.-P.; software, A.N.; validation, V.S.-M., A.L.-S., A.N. and V.M.-P.; formal analysis, V.S.-M., A.L.-S., A.N. and V.M.-P.; investigation, V.S.-M., A.L.-S., P.T.-M., E.M., V.F.-R., A.N. and V.M.-P.; resources, V.S.-M., A.L.-S., P.T.-M. and V.M.-P.; data curation, V.S.-M., A.L.-S., A.N. and V.M.-P.; writing-original draft preparation, V.S.-M., A.L.-S., A.N. and V.M.-P.; writing-review and editing, V.S.-M., A.L.-S., P.T.-M., E.M., V.F.-R., A.N. and V.M.-P.; visualization, A.L.-S. and V.M.-P.; supervision, A.L.-S. and V.M.-P.; project administration, A.L.-S. and V.M.-P.; funding acquisition, P.T.-M.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We would like to thank the participants for their uninterested participation in this project.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## **Jumping Side Volley in Soccer—A Biomechanical Preliminary Study on the Flying Kick and Its Coaching Know-How for Practitioners**

#### **Xiang Zhang 1,**†**, Gongbing Shan 1,2,3,\*,**†**, Feng Liu <sup>3</sup> and Yaguang Yu <sup>4</sup>**


Received: 4 May 2020; Accepted: 10 July 2020; Published: 12 July 2020

**Abstract:** The jumping side volley has created breathtaking moments and cherished memories for soccer fans. Regrettably, scientific studies on the skill cannot be found in the literature. Relying on the talent of athletes to improvise on the fly can hardly be considered a viable learning method. This study targeted to fill this gap by quantifying the factors of the jumping side volley and to contribute to the development of a coaching method for it. Using 3D motion capture (12 cameras, 200 Hz) and full-body biomechanical modeling, our study aimed to identify elements that govern the entrainment of skill execution. Given the rarity of players who have acquired this skill and the low success rate of the kick (even in professional games), we were able to achieve and review 23 successful trials from five college-level subjects and quantify them for the study. The results unveiled the following key elements: (1) the control of trunk rotation during jumping, (2) the angle between thighs upon take-off, (3) the whip-like control of the kicking leg while airborne, (4) timing between ball motion and limb coordination, and (5) damping mechanism during falling. An accurate kick can normally be achieved through repetitive training. This underlines the need for athletes to master a safe landing technique that minimizes risk of injury during practice. Therefore, training should begin with learning a safe falling technique.

**Keywords:** 3D motion capture; full-body biomechanical modeling; X-factor; hip flexibility; whip-like movement; dispersion of impact load during falling

#### **1. Introduction**

The great attraction of soccer for millions of fans may trace back to the basic idea of the game: the goal—an idea that never ceases to fascinate. Compared to many other sports, goals are relatively rare in soccer (on average <3 goals/game in the FIFA World Cup since the 1960s [1]). Because of the rarity, soccer goals are extremely exciting. The game can be thought of as an improvised drama, where emotional tension is built over long periods only to be fully released when a goal is achieved. In particular, the goals achieved by applying flying techniques such as the diving scorpion kick, bicycle kick, diving header, and jumping side volley are sources of rabid excitement. This uniqueness of soccer contributes to making the game the number one sport worldwide [2–5]. Among all the techniques, the jumping side volley is, no doubt, one of the infrequent scoring skills that fans invariably desire to see when attending games. Unfortunately, few players have performed this skill during national or international tournaments.

The jumping side volley is an acrobatic airborne technique (Figure 1). One can see its rarity from the European Championship 2016 (Euro 2016) where it was used for only one out of 108 goals [6]. However, this rare skill has created breathtaking moments and cherished memories for players and fans. A classic example is Wendell Lira's (Brazil) superb airborne side volley, which won the most prestigious FIFA Puskás Award 2015 [7]. In Euro 2016, Xherdan Shaqiri's (Swiss) jumping side volley was selected as the best goal of the tournament [8]. The novelty of the skill (just like the acrobatic bicycle kick) and its rarity are because these kicks are perceived as high-risk and low-return skills [9]. Regrettably, a scientific study on the skill cannot be found in the literature [4]. Relying on the talent of athletes to improvise on the fly can hardly be considered as a viable learning strategy. Therefore, in order to give efficient and effective information to practitioners, we launched a preliminary study [10] to give a scientific overview of the kick, to identify key features of the skill, and to formulate a scientific way of learning/training the technique. Specifically, this preliminary study had two objectives: (1) use of 3D motion capture technology to quantify the dominant factors contributing to the kick quality and (2) identify biomechanical elements that govern the entrainment of the jumping side volley in order to develop its coaching method.

**Figure 1.** Jumping side volley—a flying techniques for defeating goalies.

#### **2. Materials and Methods**

#### *2.1. How to Establish a Lab Test Condition for Mimicking Reality in an Application-Oriented Investigation?*

Since the study was an application-oriented investigation, the first challenge was how to gather realistic data. We went back to professional games to find a reasonable solution.

In reality, ball flying-in direction is arbitrary. Figure 2 supplies four successful examples selected from professional soccer games. Example 1 (Figure 2a) shows that the player used his head to set a flying-in ball vertically and then performed a jumping side volley. In examples 2 and 3 (Figure 2b,c), the player employed his chest and foot, respectively, to set a flying-in ball vertically for doing a jumping side volley. A vertically travelling ball can also be set by other players (Figure 2d, the goal keeper set a vertically travelling ball for a player). All four examples demonstrate that the vertically travelling ball is a frequent scenario for executing the jumping side volley. Therefore, the vertically travelling ball was used for each subject in our lab-based data collection.

**Figure 2.** Successful examples of the jumping side volley in professional games. The common feature of the examples is a vertically travelling ball. The ball condition can be created by a player using (**a**) his head; (**b**) his chest; (**c**) his foot; or (**d**) set by another player.

#### *2.2. Motion Capture and Biomechanical Modeling*

A 3D, twelve-camera VICON MX40 motion-capture system (VICON Motion Systems, Oxford Metrics Ltd., Oxford, England [11]) was used to measure the jumping side volley using 42 reflective, 9 mm markers on the body. The motion capture system tracked the markers at a rate of 200 frames/s. This capture rate has been widely applied in various analyses of complicated/elite sport-skills [12–16]. Figure 3 shows a 3D computer reconstruction of a single trial, including camera placement, capture volume, and a rendered stick figure. Markers were placed on subjects as follows: (1) four on the head, (2) trunk markers on the sternal end of the clavicle, xiphoid process of the sternum, C7 and T10 vertebrae, each scapula, left and right anterior superior iliac spine, posterior superior iliac spine, (3) upper-extremity markers on the right and left acromion, lateral side of upper arm, lateral epicondyle, lateral side of forearm, styloid processes of radius and ulna, and distal end of 3rd metacarpal bones, and (4) lower extremity markers on left and right lateral sides of thigh and shank, lateral tibial condyle, lateral malleolus, distal end of 5th metatarsal, calcaneus, and big toe. Please note that markers on the scapula, upper arms, forearms, thighs, and shanks, are referential, therefore, no accurate positions were required, i.e., as long as they are over the required bones, they will work. For reducing the risk

of injury, soft markers were applied for the tests. These compress (and decompress) easily, therefore, their influence on skill performance is negligible. Calibration residuals were determined in accordance with VICON's guidelines [11] and yielded positional data accurate within 1 mm.

**Figure 3.** 3D motion capture reconstruction showing the 12 camera placements and a wire frame mesh reproduction of a jumping side volley (left) and a 15-segment biomechanical model built from the 3D data collected.

Additionally, three reflex markers (made from 3M reflective paper) were glued to the ball in order to quantify the soccer release speed.

VICON software triangulated the positions of each marker and rendered them in three-dimensional computer space. The raw data collected was processed by a five-point smoothing filter. The five-point filter was a premier filter in the time domain that reduced random noise while retaining a sharp step response [16]. It is widely applied to reduce noise from possible vibrations of the markers during 3D motion capture of sports and arts performances [14,17,18]. The resultant data supplied primary information, such as marker position, position changes, velocities, and accelerations. From the marker-position data, anatomical landmarks were established that allowed modeling of the skeletal structure for each participant. Using basic physics, simple positional data were translated into skeletal movement. VICON software provide tools for building a 15-segment biomechanical model of the soccer kick (Figure 3) [19–22]. Model segments were identified as follows: head, upper trunk, lower trunk, upper arms, lower arms, hands, thighs, shanks, and feet. The model calculated segment lengths, joint angles, and ranges of motion (ROMs) for the joints [23,24]. In such biomechanical modeling, inertial characteristics of the body are estimated using anthropometric norms found through statistical studies [25,26]. The modeling enables researchers to postulate motor control patterns. After model calculations, descriptive statistics (i.e., average and standard deviation) of body kinematic data (i.e., joint angles, joint ROMs, and coordination timing of joints) and correlation analyses among the body kinematic data with ball release speed were performed using EXCEL 2016. The ball release speed is commonly used to judge the kick quality [3–5], as such, the correlation analyses aimed to find the key/dominant factors among the kinematic data that could govern the entrainment of the jumping side volley.

Motion capture technology permits considerable freedom of movement for participants without negatively influencing their motor skill control. Taking advantage of this, we placed no restrictions on subjects' movements within the capture volume in an effort to preserve their normal "style". Given the rarity of individuals who have acquired this skill, we used the "self-identification" method in the search for subjects at three universities (all have the VICON system) and recruited 14 male soccer players. The subjects were informed of the testing procedures, signed consent forms, and voluntarily participated in the data collection. The universities' human-subject committees scrutinized and approved the test as meeting the criteria of ethical conduct for research involving human subjects. Through pretests, we found only five players who could actually perform this skill in the required test conditions, but not at a 100% success-rate. Two of them were Canadian and the others were Chinese. The anthropometrical characteristics and experience in soccer training were as follows: body height 1.74 ± 0.04 m, body weight 70.4 ± 3.8 kg, age 22.0 ± 1.6 years, and training 15.8 ± 1.5 years. After warm-up, each subject was asked to perform the skill six times. In total, 23 successful trials (i.e., the ball was accurately and powerfully kicked) were captured.

#### **3. Results**

Figure 4 shows that two events—take-off and ball contact—divide the jumping side volley into three phases: (1) the jumping phase, (2) the airborne phase, and (3) the landing phase. Our data reveal the following characteristics of the 1st phase: Before the volley, the athlete's trunk and pelvis rotate away from the goal. During the takeoff, the non-kick-side (NKS) leg is raised, at the same time, the trunk reverses rotational directions and twists toward the goal. In order to increase the range of motion (ROM) of trunk rotation, both arms abduct to near horizontal (over 80◦, Table 1). Until the take-off, the ROM of the trunk twist (commonly known as the X-factor [27–29]) is about 40◦ (Table 1). At the end of the phase, the trunk-orientation approaches a more horizontal position (Figure 4). Correlation analyses confirm that the X-factor (α), angle between thighs (β) at take-off, and shoulder abduction during jumping are key/dominant factors, influencing the kick quality, i.e., the ball release speed (Figure 5a,b).

**Figure 4.** Phase identification based on 3D motion analysis data.

**Table 1.** Average and standard deviation of selected parameters, their confidence intervals (CI), coefficient of variation (CV), and their correlation with the ball release speed (r) (*p* < 0.05).


X-factor: the angle between shoulder line (upper trunk) and hip line (lower trunk). Flex/Ext: Flexion/Extension, KS: Kick Side. \*: the max value is the average of both shoulders.

Two notable characteristics of the 2nd phase (i.e., the airborne phase) are the flexed kick-side (KS) leg and the extended NKS leg (Figure 5c). Both legs form a scissor-movement, i.e., they move in opposite directions. The correlation analyses show that both the minimum angle and ROM of the KS knee are key factors affecting the kick quality (Table 1). From the timing perspective, the KS hip flexion starts when about 80% of the trunk twist toward the goal has finished. Similarly, the KS knee begins its extension after the KS hip finishes about 75% of its flexion (Table 1). Further, our data show that the explosive KS knee extension happens shortly before the ball contact, and is followed by an ankle flexion. The correlation analyses unveil that the sequential segment-coordination is also a key contributor to the kick quality.

**Figure 5.** Key/dominate factors influencing kick quality: (**a**) X-factor α (top view); (**b**) Take-off angle between thighs β (side view); (**c**) Minimum KS knee angle γ (top-front view).

The main issue in the 3rd phase (i.e., the landing phase) is how to dissipate the impact load produced by falling in order to avoid potential injuries. Our data reveal that the well-trained athletes apply multiple landings to share the impact loads for reducing the risk of injury. The 1st landing is the flexed arm–hand chain (like a spring) for the 1st damping. The 2nd landing is the hip landing, followed by body rolling, sharing the rest load among multiple contact points.

#### **4. Discussion**

Since this study focuses on coaching practice, the results should be coach-friendly. A coach-friendly study, in our vision, should use a well-established scientific method to supply explanations on (1) scientifically-identified motor-control sequencing; (2) dominant factors (determined by correlation analyses) contributing to the control of the motor skill; and (3) instructions that can be understood by coaches and applied in their intervention in practice. Therefore, based on numerous analyses of our 3D data and their correlation analyses, the following key factors have been selected for illustrating the secrets of the control of the acrobatic skill in order to establish a practical way to learn and to train the jumping side volley:


A focused communication would help practitioners to understand the complex motor control in a timely and efficient way.

Regarding the determinants influencing the kick quality, the following key/dominant factors were revealed: There are two key factors in phase 1: the twisting control of the upper body (α) and the instance angle between the two thighs at take-off (β). The larger α and β are, the more powerful is the kick. Actually, these two factors play a crucial role in laying a foundation for performing the whip-like control of the kick in phase 2. In conjunction with the key factors in phase 2, the effects of α and β are elaborated below.

One notable feature of phase 2 is the asymmetric control of the legs, i.e., multi-segment control of the kick leg vs. quasi-single-segment control of the non-kick leg. This asymmetric control results in a difference between the moments of inertia of the two legs. The difference is vital for forwardly accelerating the kick leg; as such, it influences the quality of the whip-like movement. In the airborne phase, the human body follows the Law of Conservation of Angular Momentum. That means, the angular momentum of the forward action (the KS leg) equals the angular momentum of the backward reaction (the NKS leg). The flexed KS leg leads to a smaller moment of inertia (I), resulting in a faster forward motion in comparison to the extended NKS leg (a larger I creates lower backward motion) (Figure 6). In conjunction with phase 1, we found that the whip-system of the jumping side volley consists of four segments: trunk, KS-thigh, KS-shank, and kick foot. The whip-like kick is actually initiated in phase 1, beginning with the trunk twisting toward the goal, followed by hip flexion, knee extension, and ankle flexion, showing a sequential flow of energy and momentum transfer. It is well known that increasing the ROM of each segment will enhance the effect of the whip-like movement [14,19,28–31]. Therefore, flexibility of hip and knees should be emphasized during training, and the training should also pay attention to segment coordination (i.e., the whip-like control). Of course, timing is the most crucial element for coordination training.

**Figure 6.** The asymmetric control of the legs during the airborne phase and their effects revealed by the Law of Conservation of Angular Momentum, i.e., the smaller the I, the faster the ω and vice versa.

For training of the jumping side volley, one should pay attention to the following three aspects: kick power, accuracy, and timing.

It is well known for practitioners that it is important to develop full-body whip-like control (from trunk to KS foot) for increasing kick power. For reaching this goal, jumping training should emphasize the upper-body twisting control, and the flexibility training should focus on the hip (one should always remember that the larger the β is, the more powerful is the kick). The airborne training should establish an asymmetrical control of legs as well as the ROM training of the KS-leg joints, i.e., KS leg performing a whip-like control and NKS leg remaining extended.

Repetitive training is a traditional way to increase the aspects of kick accuracy and timing. Those of us who are involved in the coaching practice know well that there are no short-cuts. The jumping side volley is considered high risk because of the inevitable fall to the ground, i.e., a realistic fear of injury to players/learners. We know that the feet and legs are naturally designed for absorbing landing impacts, but the arms and body are not. What does that mean? That means that bio-adaptation training for strengthening the arms and body against impact and safe-fall training to minimize injury risk during falling to the ground should first be considered in skill learning [32]. Hence, if we apply the repetitive-training approach in learning the jumping side volley, we should begin with phase 3, i.e., the learning and training of a safe falling technique, e.g., the multiple-landing technique revealed

by the current study. Mastering a safe landing technique is the foundation that can ensure the repetitive training for improving kick accuracy and timing as well as the kick power.

Every study has limitations. As application-oriented research, the current study has two potential weaknesses. The obvious one is the low subject number and level. Future studies, if researchers cannot gain data from professional-level players, could first train advanced players based on the results of this study. As such, more subjects could be "produced" for future studies. The other one is the ecological approach to motor learning. For initiating the study, we simplified the ball movement. In reality, due to dynamic changes in the playing environment, a vertical travelling ball is only one of many possibilities. Therefore, identifying the learner–environment relationship as the basis for learning design in sport intentionality plays an important role in training. It will help to capture perception and action as intertwined processes underpinning individual differences in movement behavior [33,34]. The application of a nonlinear pedagogy, particularly in open and complex skills such as a jumping side kick, could reinforce the authors' intentions to help coaches to develop soccer-skills with their players. It would be a practitioner-desired topic for future studies.

#### **5. Conclusions**

The current study indicates one possibility to entrain the jumping side volley in soccer. Based on the results, the element training should focus on increasing the flexibility of the hip, the efficiency of the whip-like movement of the proximal-to-distal acceleration of the kicking leg, and the damping mechanism during falling. For skill training, one should focus on timing, because accurate timing is vital for a successful kick. One traditional way for improving timing is through repetitive training, which means repetitive falls to the ground during learning/training. Without safe fall protection, the skill learning/training presents a high risk of injury. This underlines the need for athletes to learn a safe landing technique that minimizes the risk of injury during practice. Therefore, if one applies the traditional approach in practice, training should begin with phase 3: mastering a safe fall.

**Author Contributions:** G.S., X.Z., F.L., and Y.Y. conceived and designed the experiments; G.S. and X.Z. performed the experiments; G.S., X.Z., F.L., and Y.Y. analyzed and discussed the data; G.S. and X.Z. contributed /materials/analysis tools; F.L. and Y.Y. prepared figures; G.S. and X.Z. wrote the paper; all authors contributed to the revisions and proof reading of the article. All authors have read and agreed to the published version of the manuscript.

**Funding:** The research project was supported by the National Sciences and Engineering Research Council of Canada (NSERC), grant number: RGPIN-2014-03648.

**Conflicts of Interest:** The authors declare no conflict of interest. The founding sponsors had no role in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*

## **E**ff**ects of Plyometric Training on Surface Electromyographic Activity and Performance during Blocking Jumps in College Division I Men's Volleyball Athletes**

**Min-Hsien Wang 1, Ke-Chou Chen 2, Min-Hao Hung 3, Chi-Yao Chang 3, Chin-Shan Ho 3, Chun-Hao Chang <sup>3</sup> and Kuo-Chuan Lin 2,\***


Received: 29 April 2020; Accepted: 29 June 2020; Published: 30 June 2020

**Abstract:** In volleyball matches, there are three minute intervals between sets. Therefore, the improvement of the muscle output ratio is one of the most import foundational physical elements for the players. The purpose of this study was to investigate the effects of plyometric training on the changes in electrical signals in the lower limb muscles of male college volleyball players during continuous blocking and to examine the benefits of plyometric training on blocking agility and maximum vertical jump height. In this study, twenty elite male college volleyball players were recruited and divided into a plyometric training group (PTG) and a control group (CG). The wireless electromyography was used for data acquisition, and the electrodes were applied to the left and right rectus femoris, biceps femoris, tibialis anterior, and gastrocnemius. The median frequency was used as the measurement of the electromyographic signals during the jumping blocks. This study used covariate analysis methods, with previously measured results used as covariates to perform a two-way analysis of covariance for the independent samples. Based on the results of this study, after 6 weeks of training, the median frequency of the rectus femoris (2.13% to 4.75% improved) and that of the tibialis anterior muscles (4.14% to 7.71% improved) were significantly lower in the PTG than in the CG. Additionally, the blocking agility increased by 6.26% and the maximum vertical jump height increased by 3.33% in the PTG compared to the CG. The findings provide important insights on the neuromuscular status for volleyball players during continuous blocking jumps. Six weeks of appropriate plyometric training can facilitate the performance of volleyball players.

**Keywords:** muscular activity; blocking agility; maximum vertical jump height; median frequency

#### **1. Introduction**

Jumping ability is essential for performance in volleyball. Superior rebounding not only enables players to gain a competitive advantage on offense (increasing blocking height and attack angle) but also allows for a larger defensive range [1–3]. To increase the vertical jumping ability, weight-bearing jumping or plyometric jump-training methods can effectively improve leg muscles and explosive power, thereby improving overall strength and coordination in the legs. Vertical jumping is an important basic skill in many sports [4], but repetitive jumping is the primary cause of muscle fatigue. A study by Merletti and Parker [5] showed that the sites of neuromuscular fatigue can be divided into the

following three major categories: central fatigue, fatigue of the neuromuscular junction, and muscle fatigue. Power-generating nerve signal transmission can be divided into central and peripheral nerve levels; thus, fatigue can be classified as central fatigue or peripheral fatigue based on the different sites of fatigue. Decreased muscle strength during fatigue is accompanied by central and peripheral effects. The former is associated with a decrease in the number of motor units involved in the action or a reduction in the frequency at which motor units are evoked.

It is well known that fatigue caused by vertical jumping can alter muscular characteristics, reduce muscle effectiveness, and change the maximum joint torque [6,7]. However, superior vertical jumping ability requires an excellent coordination of movements, that is, the ability to control and adjust musculoskeletal characteristics [8]. Therefore, fatigue may be a factor that causes changes in the central nervous system (CNS) and alters the coordination of the limbs during vertical jumps. The effects of post-jump fatigue on motor coordination have been discussed in previous studies [9–11], and the results showed that continuous jumping can reduce knee and ankle strength. Additionally, in the case of decreased muscle strength, it is not possible to improve strength even if the range of motion of the lower limb joints is increased.

Previous studies [12] have indicated that the rectus femoris, vastus lateralis, vastus medialis, and biceps femoris are the primary muscles used during squat jumps. After 50 repeated squat jumps, the maximum voluntary contraction of the knee extensors decreases by 25 ± 11%, but there is no significant change in the knee flexors. In ve Dikey et al. [13], the correlation between maximum isokinetic strength, muscle activity, and jump height in 12 elite male volleyball players was investigated. The results of the study showed that hamstring/quadriceps ratios were greatest at an angular velocity of 240◦/s. Additionally, the degree of biceps femoris activity was greater than that of the vastus lateralis and vastus medialis regardless of the angular velocity, and the degree of activity of the biceps femoris and the rectus femoris was consistent. The degree of muscular activity (MVC%) of the knee extensors and knee flexors increases with increasing fatigue. Therefore, some studies have recommended the evaluation of vertical jumping based on actual sports conditions [14,15]. In this method, individuals performed intermittent fixed-height jumps, such as intermittent jumps to 95% of the maximum vertical jump height, until the target jump height cannot be reached for three consecutive attempts. The degree of fatigue is estimated from the resting interval after each jump and the number of vertical jumps [14].

Plyometric training is a typical high-intensity exercise, and the generated peripheral fatigue can be significantly improved after plyometric training [16]. A main finding is that the maximum voluntary contraction and degree of activity generated by muscles are significantly improved after plyometric training [17]. Plyometric training can effectively enhance muscle activity and improve the muscle output ratio. This effect not only significantly increases the maximum jump height but also enhances the coordination of the lower limb muscles [18]. Plyometric training is a jumping training method that uses the physiological phenomenon of a stretch shortening cycle (SSC) to produce stronger contractions during the centripetal contraction phase [19]. Because volleyball players must perform movements such as repeatedly jumping, sprinting and changing direction, this training method is appropriate for targeting their physical training needs and is widely applicable [20,21].

According to previous studies, plyometric training has many positive effects on volleyball players. However, most of the prior studies focused on improvements in general physical abilities due to plyometric training. In this study, sport-specific tests were evaluated with game-like situations simulating volleyball blocking jumps. The tests required repetitions of blocking jumps, during which the fatigue state of the lower limb muscles was recorded. The measurement methods used in this study were modified from the study by Sheppard et al. [1], which used repetitive jumping blocks in competition-like conditions. Surface electromyography with electrodes attached to the lower limb muscle groups were used to elucidate the mechanisms and conditions caused by muscle fatigue. Plyometric training was used to strengthen muscle groups prone to fatigue so as to reduce the rate at which fatigue develops and enhance athletic performance. The initial hypothesis of this study was that plyometric training can induce positive changes in muscle activities in male college volleyball players.

#### **2. Methods**

#### *2.1. Participants*

This study was designed to be similar to a blocking action situation in actual competition. Players were divided into a plyometric training group (PTG) and a control group (CG) during the training period. Before entering the training cycle, the players in both groups underwent a pre-test. The PTG underwent six weeks of training after the pre-test, and the CG maintained the original training program. The two groups underwent a post-test after six weeks. In this study, 20 elite male college volleyball players were used as the study participants, and they were equally divided into the PTG (n = 10; mean age = 21.5 ± 1.2 years; mean height = 186.5 ± 5.1 cm; mean weight = 78.1 ± 4.7 kg) and the CG (n = 10; mean age = 22.1 ± 1.5 years; mean height = 176.5 ± 4.4 cm; mean weight = 77.4 ± 5.2 kg). All the participants in the study had participated in professional volleyball training for over five years and were registered in the men's Division I of the Republic of China University Volleyball League. The subjects were free of major musculoskeletal system disorders within the year preceding the study. This study conducted experiments using competition-like conditions during testing. Although slightly different from the actual competition conditions, verbal cues were used in the experiment, and the participants were required to make their best effort. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Fu Jen Catholic University Institutional Review Board (C103117).

#### *2.2. Data Collection*

In this study, 10 Vicon T40 motion capture systems (Vicon MX-Giganet, Oxford Metrics Ltd., UK) and Nexus software (Version 4.0.2.; Vicon Motion System Ltd, UK) were used with a frame rate of 250 Hz. One marker was attached on the seventh cervical vertebra (C7) to assess the jump height. In this study, wireless electromyography (BTS Free EMG; BTS Bioengineering Corp., USA) was used by attaching electrodes to muscle groups in the left and right legs, including the rectus femoris (RFM), biceps femoris (BFM), tibialis anterior (TAM), and gastrocnemius (GM) [22]. The sampling frequency was set to 1000 Hz. According to previous research [22], five muscles, each with their own specific anatomical function, have the major influences on jumping movements, as follows: RFM (hip joint flexion, knee joint extension), BFM (hip joint extension, knee joint flexion), vastus lateralis muscle (knee joint extension), GM (knee joint flexion, ankle plantar flexion), and TAM (ankle dorsiflexion, inversion). In this study, the athletes were required to perform blocking jumps in two directions for a total of 14 times. We aimed to observe the muscles that dominated in at least two functions during the blocking jumps. Therefore, the EMG measurements of the RFM, BFM, TAM, and GM were recorded. The blocking agility system (1000Hz sampling frequency) was used to record the total response time [23].

#### *2.3. Experimental Design*

Actual testing was performed as modified in the study by Sheppard et al. [1]. The blocking target was placed three meters away from the starting position of the movement, and the blocking height was set to 90% of the subject's individual jump height. The subjects performed 14 moving blocks. A total of three groups of tests (T1, T2, T3) were performed, with a three-minute rest between the groups. To give the subjects a fixed load, the blocking agility (BA) test from a study by Ho et al. [23] was used, with an eight-second interval between the blocks. All the data in this study were collected simultaneously using the Vicon server. The site layout is shown in Figure 1.

**Figure 1.** Diagram and explanation of the blocking agility test for volleyball. The athlete was required to perform a total of 14 blocks from two directions. The blocking agility test system and wireless electromyography (BTS Free EMG) were set in the front of the athlete.

#### *2.4. Procedures*

Before the start of the experiment, the participants were collectively informed about the purpose of the study and the experimental process, and any issues needing attention were addressed. Additionally, their rights during the study were disclosed, after which the subjects gave written informed consent. One week before testing, all the participants were provided with information about the experiment and administered two practice tests. During the course of the experiment, the subjects were instructed to not change their daily training program or training volume. The PTG underwent six weeks of plyometric training during the study. The primary purpose of their regimen was to train the muscle groups in the legs. The training program for this study was formulated as described by Makaruk and Sacewicz [18] (Table 1). During the six-week plyometric training program, the subjects participated in training sessions twice a week. The intensity and training volume of the regimen was based on the number of groups and repetitions proposed by Piper and Erdmann [24]. Two weeks of medium-intensity training and four weeks of high-intensity training were implemented in the program to improve the transmission of the CNS signals in the subjects and avoid excessive load or fatigue.


#### **Table 1.** Plyometric training program (PTG).

#### *2.5. Dependent Measures*

The parameters investigated in this study were the myoelectric signals, blocking agility (BA), and the maximum vertical jump height. The collection and processing of each parameter are described as follows:

1. Collection and processing of the maximum vertical jump height

The participants were asked to stand on a force plate to perform a countermovement jump, which was measured using a three-dimensional motion capture system. The maximum vertical jump height test was performed three times. A program was written in Matlab software (Version R2008a; The MathWorks Inc., USA) to calculate the height of the highest point after the player takes off.

2. Blocking agility (BA) test

The subjects were required to warm up adequately for 10–15 min, with special attention focused on extending the joint ligaments in the legs. The subjects could complete the warm-up when they felt comfortable and ready. In the test, the blocking agility system was used to produce visual stimuli and record the total response time during blocking. The examiner gave a voiced signal to the athlete and triggered the blocking agility test system at the same time. The light cue was activated as stimulation 8 s after the system was triggered. During the BA test, the subjects were instructed to land on both feet. The actions were performed seven times in each direction (left and right) for a total of 14 blocks, and three repeated sets of records were analyzed. The subjects were directed to stand in the preparation area, which was three meters away from the blocking point, and to wait for a light to turn on. When the light turned on, the subject completed the blocking action as quickly as possible by touching the target.

3. Collection and processing of the muscle median frequency

Free EMG was used to wirelessly transmit data to a computer, and the collected EMG data were stored and Matlab 7.0.1 was used to read and write a program. The EMG signals were band-pass filtered (10–500 Hz) with a fourth order Butterworth filter. The median frequency (MDF) was derived from the EMG data; MDF is the frequency value that divides the power spectrum into two equal regions and is considered a reliable method for assessing muscle fatigue during exercise [25]. Studies [26,27] have shown that when muscles become fatigued, high-frequency motor units are evoked less frequently, and low-frequency motor units are evoked more frequently and recruited in larger numbers. This increases the slow contraction effects and slows muscle fiber conduction velocity, such that the MDF of the power spectrum is biased towards low frequencies (Figure 2); this characteristic can be used to assess muscle fatigue.

(b)

**Figure 2.** The signals of surface electromyographic activity and the results of median frequency (MDF) on the lower extremity muscles. (**a**) The EMG signals for the rectus femoris (RFM), biceps femoris (BFM), tibialis anterior (TAM), and the gastrocnemius (GM) during the blocking jumps. The time series of the EMGs were captured and analyzed between the subject contact "\*" with the force-plate and when it leaves "o". (**b**)The median frequency of the signal spectrum.

#### *2.6. Statistical Analysis*

The Statistical Package for the Social Sciences (SPSS) 20.0 software (version 20.0; SPSS Inc., Chicago, IL, USA) was used for the statistics and data analysis. First, the reliability of the measured data was tested using the intra-class correlation coefficient (ICC). This study primarily investigated the effects of plyometric training and continuous blocking on the MDFs of the RFM, TAM, GM, and BFM of the study participants. Descriptive statistics methods were used to describe the median frequency of the participants tested before and after the blocking rounds. As some interfering factors in actual

experimental situations can affect experimental results, a two-way analysis of covariance (ANCOVA) with pre-test performance as the covariate was used to correct for sources of error and increase the accuracy. If the effect of the two factors reached the level of significance, a simple effects test (Bonferroni post-hoc test) was performed. If the interaction did not reach the level of significance, a main effects test was performed. The significance level was set at alpha ≤ 0.05.

#### **3. Results**

The pre- and post-test ICCs for the MDF (RFM, TAM, GM, and BFM) and jump performances are shown in Table 2. All the variables showed moderate to excellent results. ANCOVA, which was used in this study, requires that the variables in the data are homogeneous. The results of Levene's test showed that all the variables were homogeneous (p > 0.05) and that there were no interactions between the covariate and the independent variables that met the basic assumptions for ANCOVA.

**Table 2.** The pre- and post-test intra-class correlation coefficient (ICC) for the MDF (RFM, TAM, GM, and BFM) and jump performances (blocking agility (BA) and maximum vertical jump height).


#### *3.1. Muscle Activities of Blocking Jumps*

The results of pre- and post-testing in the PTG and CG (Table 3) show that as the number of rounds increased, the MDF of both the RFM and TAM decreased, but the MDFs of the GM and BFM did not. The frequency was highest in the BFM followed by the TAM; the frequencies of the RFM and GM were similar. The differences in the RFM reached the level of significance with respect to the group and group × round but not with respect to the round. The comparison of the marginal means showed that the PTG > the CG. The differences in the TAM reached the level of significance with respect to the group and group × round but not with respect to the round. The comparison of the marginal means showed that the PTG > the CG. The differences in the GM did not reach the level of significance with respect to the group, round, or group × round. The differences in the BFM did not reach the level of significance with respect to the group, round, or group × round. As the interaction between the RFM and the TAM reached the level of significance with respect to the group × round, a simple main effects test was performed. The results for the RFM in the CG showed that the MDF decreased by 11.55% and 20.61% in the second and third rounds of the pre-test, respectively, and by 11.04% and 21.52% in the second and third rounds of the post-test, respectively, exhibiting similar declines. In the PTG, the MDF decreased by 13.49% and 22.42% in the second and third rounds of the pre-test, respectively, and by 11.36% and 17.67% in the second and third rounds of the post-test, respectively; the decrease in the MDF in the PTG was significantly slower than that in the CG. The results of the MDF for the TAM in the CG showed that the MDF decreased by 23.96% and 31.79% in the second and third rounds of the pre-test, respectively, and by 22.20% and 30.99% in the second and third rounds of the post-test, respectively; the degrees of decline in the MDF were nearly the same in both groups. In the PTG, the MDF decreased by 25.08% and 35.02% in the second and third rounds of the pre-test, respectively, and by 20.94% and 27.31% in the second and third rounds of the post-test, respectively.


**Table 3.** Values are the mean ± standard deviation. Data are reported for the MDF of the RFM, BFM, TAM and the GM during the blocking jumps. The unit of MDF

PTG: plyometric training group; CG: control group. The table indicates significant change (post–pre) when using the pre-test score as a covariate: \* *p* < 0.05; \*\* *p* < 0.01. RFM:femoris muscle; BFM: biceps femoris muscle; TAM: tibialis anterior muscle; GM: gastrocnemius muscle.

 rectus

#### *3.2. Jump Performances*

Table 4 shows the BA of the two groups measured before and after training. The average BA test score over the three rounds in the PTG and CG was 2.08 ± 0.14 s and 2.09 ± 0.14 s in the pre-test, respectively, and 1.95 ± 0.11 s and 2.08 ± 0.17 s in the post-test, respectively. This shows that the magnitude of decrease in the post-test was larger in the PTG than in the CG. The ANCOVA results showed that the BA test reached the level of significance with respect to the group but not with respect to the round or group × round (p > 0.05). Since there were only two levels of group factors, comparing the marginal means showed that the PTG > the CG; that is, the BA of the PTG was significantly higher than that of the CG. However, the results of the descriptive statistics showed that in the CG, the BA increased by 0.48% and 1.93% in the second and third rounds of the pre-test, respectively, and by 0.97% and 1.46% in the second and third rounds of the post-test, respectively, exhibiting similar increases. In the PTG, the BA increased by 1.96% and 3.43% in the second and third rounds of the pre-test, respectively, and by 1.04% and 1.55% in the second and third rounds of the post-test, respectively.

**Table 4.** Values are the mean ± standard deviation. Data are reported for the blocking agility (BA) test. The unit of BA is second (s).


PTG: plyometric training group; CG: control group. The table indicates significant change (post–pre) when using the pre-test score as a covariate: \*\* *p* < 0.01.

Table 5 shows the maximum vertical jump height of the two groups measured before and after training. The average jump height in the PTG and CG were 67.04 ± 3.83 cm and 66.86 ± 4.06 cm in the pre-test, respectively, and 69.27 ± 3.87 cm and 66.79 ± 3.76 cm in the post-test, respectively. This shows that the PTG had a large magnitude of improvement after six weeks of plyometric training. There were significant differences between the groups. Comparing the marginal means showed that the PTG > the CG, that is, the plyometric training could effectively improve the maximum vertical jumping ability of the study participants.

**Table 5.** Values are the mean ± standard deviation. Data are reported for the maximum vertical jump height. Units in centimeters (cm).


PTG: plyometric training group; CG: control group. The table indicates significant change (post–pre) with the pre-test score as a covariate: \*\* *p* < 0.01.

#### **4. Discussion**

Based on the results of this study, the decrease in the MDF of the RFM and TAM was significantly lesser in the PTG after six weeks of training compared to that of the CG, and there was significant improvement in both the BA performance and maximum vertical jump height in the PTG.

From the results, the MDF increased in both the RFM and TAM as the number of rounds increased. In other words, under the effects of the exercise load in this study, the contraction ability of the muscles was reduced, and the characteristics of the power spectrum were altered. Among the MDF data for the four muscles, the TAM showed the greatest improvement. However, the MDF of these four muscles still decreased as the number of rounds in the blocking test increased. The amount of muscle recruitment in various parts of the legs when jumping was studied [22], and the results showed that the principal muscle groups recruited when performing squat jumps are the TAM > RFM > GM >

BFM; that is, the TAM and RFM are the principal muscle groups used when jumping. In the results of the continuous jump blocking test conducted on the study participants in the present study, the decrease in MDF in the legs improved in the PTG after a six-week interventional training; this result is consistent with that of previous studies [16]. The report [16] indicated that the plyometric training (PT) increased central fatigue significantly by about 15–20%, but significantly decreased peripheral (muscle) fatigue during the 2-min MVC by about 10% in the quadriceps femoris. In this study, the MDF of the RMF and TAM of the PTG and CG decreased by similar percentages in the pre-test. However, the MDF of the PTG was significantly higher than that of the CG in the second round (by 2.13% and 4.14% for the PTG) and the third round (by 4.75% and 7.71% for the PTG). In other words, plyometric training can delay the decrease in MDF, but it cannot alter the development of fatigue.

The ANCOVA results show that the BA of the PTG improved significantly after six weeks of training. Although not reflected as statistically significant, the BA increased as the number of rounds increased. These results show that the change occurred more gradually in the PTG. With respect to the average percentage of improvement, the PTG improved by 6.26% after six weeks of interventional training, whereas the CG improved by only 0.48%. In volleyball, blocking is not only a defensive skill but also a skill essential for scoring. In team sports, the defensive ability is considered to be a display of individual agility [28,29]. Agility is an important index in team sports and therefore must be improved through functional training. Improving the ability of the neuromuscular system to adapt and control may be factors that improve BA. Plyometric training can stimulate the CNS signal transmission, which can improve the stretch-shortening cycle ability in the leg muscles. Through the training process, exercise patterns involving the ability to change directions in a short period of time can produce the effect of muscle stretching–contraction–circulation, yielding the benefits of reflexive muscle strength and stretch reflex characteristics and improving the instant reaction ability.

The ANCOVA results showed that the standing jump height in the PTG after six weeks of training was significantly higher than that of the CG, increasing by 2.23 cm (3.33%) in the PTG and slightly decreasing by 0.07 cm (–0.10%) in the CG. Although the improvement in the PTG was only approximately 2 cm, a difference in height of 2 cm is a sign of approaching the ability to break through the maximum limits of the body in players performing at high levels. The results of the previous studies have shown that undergoing plyometric training two to three times a week can effectively improve the maximum output power and the rebounding ability of muscles [30], and the results of the present study are consistent with those of previous studies.

The results from this study show that repeated jumping causes fatigue-related declines in the RFM and TAM. In particular, blocking jumps require fast SSC muscle activity. Because of the lower leg muscle activity in the braking phase, the muscle stiffness decreases and simultaneously diminishes the efficacy of SSC actions [31,32]. To achieve the maximum rebound in muscle elastic strain energy during the push-off phase, the muscle must maintain a high stiffness during the braking phase [33]. Therefore, the EMG activities change during the jump, depending on the stiffness of the muscle. In this study, the power spectrum of MDFs (RFM and TAM) was biased towards low frequencies. The blocking agility test in this experiment comprised 14 blocking jumps, which required the athletes to perform repeated SSC movements, which in turn led to declines in muscle activities. In this research, a six-week plyometric training period was applied to provide additional loads on the athletes and thereby stimulate adaptation. At the same time, the training program included sufficient rest time, which is important for the recovery of muscle functions, thereby delaying the occurrence of muscle fatigue and improving the CNS function after the training period. In addition, enhancements in the acquired abilities through plyometric training primarily result from the stimulation of the muscular and nervous systems, and the process of adaptation.

The athletes recruited in this study were sufficiently proficient and physically fit to fulfill the requirements of volleyball matches. While we found effective improvements in these volleyball players after plyometric training, we produced no conclusive results regarding the settings that should be used for optimal effects. Further studies should help determine the appropriate intensity of plyometric training. The results of this study may not be generalizable to populations differing from the skill level, sex and age. In this study, a potential weakness of the data collection method was that the BA test program included a fixed exercise intensity. However, the exercise intensity in real games is variable; therefore, the time intervals could be varied in future studies.

#### **5. Conclusions**

Based on the results of this study, six weeks of appropriate plyometric training can delay the decrease in MDF in the RFM and TAM, improve vertical jump height, and significantly shorten BA in volleyball players when movement and jumping are combined. Increasing the muscle strength and coordination of movements in the legs facilitates rapid and complete movements during a moving block; that is, the total power output during the nerve conduction and muscle contraction processes is improved. For volleyball athletes, jumping high and fast and maintaining a high level of performance on the court is the goal of training. Based on fundamental training principles (e.g., variation, periodicity, individualization), recommendations to coaches include the use of different jumping loads to stimulate adaptation in the athletes. Neuromuscular fatigue during continuous jumping can be evaluated using the MDF after EMG spectrum analysis. Such markers can therefore be used to assess the fatigue resistance of athletes in competition-like situations.

**Author Contributions:** Conceptualization, M.-H.W. and K.-C.C.; methodology, M.-H.W.; software, C.-S.H.; validation, M.-H.H., C.-Y.C. and K.-C.L.; formal analysis, M.-H.H.; investigation, M.-H.W.; resources, K.-C.C.; data curation, C.-H.C.; writing—original draft preparation, C.-Y.C.; writing—review and editing, K.-C.L.; visualization, K.-C.L.; supervision, M.-H.W.; project administration, C.-S.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*

## **Can We Rely on Flight Time to Measure Jumping Performance or Neuromuscular Fatigue-Overload in Professional Female Soccer Players?**

**Estrella Armada-Cortés 1,2, Javier Peláez Barrajón 1, José Antonio Benítez-Muñoz 3, Enrique Navarro <sup>1</sup> and Alejandro F. San Juan 1,\***


Received: 27 May 2020; Accepted: 24 June 2020; Published: 27 June 2020

**Featured Application: It is recommended that high-level sportswomen and men should be assessed with the force platform through the take-o**ff **velocity method in a vertical jump as gold standard technology to ensure correct performance and**/**or fatigue-overload control during the sport season.**

**Abstract:** The main purpose of this study was to compare the validity of the take-off velocity method (TOV) measured with a force platform (FP) (gold standard) versus the flight time method (FT) in a vertical jump to measure jumping performance or neuromuscular fatigue-overload in professional female football players. For this purpose, we used a FP and a validated smartphone application (APP). A total of eight healthy professional female football players (aged 27.25 ± 6.48 years) participated in this study. All performed three valid trials of a countermovement jump and squat jump and were measured at the same time with the APP and the FP. The results show that there is a lack of validity and reliability between jump height (JH) calculated through the TOV method with the FP and the FT method with the FP (r = 0.028, *p* > 0.84, intraclass correlation coefficient (ICC) = −0.026) and between the JH measured with the FP through the TOV method and the APP with the FT method (r = 0.116, *p* > 0.43, ICC = −0.094 (−0.314–0.157)). A significant difference between the JH measured through the TOV with the FP versus the APP (*p* < 0.05), and a trend between the JH obtained with the FP through the TOV and the FT (*p* = 0.052) is also shown. Finally, the JH with the FP through the FT and the APP did not differ (*p* > 0.05). The eta-squared of the one-way ANOVA was η2 = 0.085. It seems that only the TOV measured with a FP could guarantee the accuracy of the jump test in SJ+CMJ and SJ, so it is recommended that high-level sportswomen and men should be assessed with the FP through TOV as gold standard technology to ensure correct performance and/or fatigue-overload control during the sport season.

**Keywords:** vertical jump; flight time; take-off velocity; fatigue; muscle overload; performance; force platform; smartphone application; APP; female soccer

#### **1. Introduction**

The vertical jump (VJ) is one of the most widely used performance tests. Its popularity is due both to its simplicity and effectiveness [1], and to the similarity between its movement patterns and sports, explosiveness, speed, and intensity [2,3]. Moreover, it is a useful tool to assess physical fitness in a healthy population (i.e., children, adults, and elderly people) [4–6], and to control an adequate musculoskeletal injury recovery [7]. There are different jump protocols; two of the most frequently used are: (1) The squat jump (SJ), which provides a basic leg strength parameter and is described by some authors as pure concentric [3]; (2) the counter movement jump (CMJ) adds an elastic-reactive component of the subject. It is estimated that the CMJ value can be up to 25% higher than the SJ in athletes [8]. Their performance has been used to: (1) monitor the positive effects of strength, plyometric, resistance, and speed training; and (2) control mechanical and neuromuscular fatigue status in individual and team sports [9]. Several researchers have found that CMJ performance is an interesting objective marker of fatigue and overcompensation for athlete performance [9], being one of the factors related to the high incidence of injuries (e.g., muscle overload) in the lower limb muscles [10,11]. Thus, a relationship has been observed between height loss in the CMJ and metabolic markers such as lactate or ammonium in the sprint [12], and after a Wingate test [13]. This suggests that through decreases in the CMJ's mechanical variables such as jump height (JH), it should be possible to estimate the metabolic stress, neuromuscular fatigue, and overload of the subject [5,12,14].

Specific performance factors for both the CMJ and SJ include a number of different kinematic and kinetic variables. Some of these variables are more sensitive than others to determine an athlete's neuromuscular status and may depends of the age and level of the subjects: JH, power, velocity and force (peak, relative peak, and mean of the last three), rate of force development (RFD), and mean impulse [9]. In addition, it has been proven that it is better to average the number of jumps executed than to choose the one that obtained the maximum height [9].

To obtain the height value of the jump we can use two variables: the take-off velocity (TOV) or the flight time (FT) [15,16]:

1. The TOV is the vertical velocity of the center of mass at takeoff. The height value of the jump will be the result of velocity at takeoff squared divided by twice the gravity:

$$\left(h = \frac{V^2}{2g}\right) \tag{1}$$

2. The FT is the time period between takeoff and land contact (i.e., the subject is on the air, without land contact). The height value of the jump will be the result of the gravity value multiplied by FT squared, divided by eight:

$$\left(h = \frac{gt^2}{8}\right) \tag{2}$$

There are many tools used to evaluate the performance of the jump (e.g., force platforms (FP), contact mats, video analysis through a smartphone applications (APPs), accelerometers, infrared systems, *Vertec* Vertical Jump Measuring Device). Most devices use the FT to calculate the JH [17]. However, to measure it in an accurate manner, the height of the center of mass at takeoff and landing has to be the same. Any difference between the athlete's takeoff and landing center of mass position (e.g., different joint position of ankle, knee, or hip) could increase or decrease the FT, and then change the result of the JH estimation [15,17]. Considering these technical implications, TOV has been suggested as the more suitable method to evaluate the JH [18]. If the TOV can be measured easily by a FP, then it is considered the gold standard for measuring the height of the VJ [19,20]. A FP can measure and use either the FT or the TOV methodology, the latter being the most accurate method for determining the height of the VJ [19].

The need to quantitatively evaluate the performance of the athlete's JH has promoted the emergence of APPs for this purpose. Some authors have reported that these APPs have a high reliability and accuracy in their measures [21,22], while others qualify their results as equivocal, so it is not yet clear what is the ideal methodology to use [23].

Although the VJ and its performance can be extrapolated to any sport level, the greater the level the subjects have, the more correlation exists in its variables, so it is in high performance sport where it has more utility [24]. Moreover, the accuracy of the results is essential from a competitive point of view in high-level sports, for example in elite-cyclists improvements of around 0.6% are sufficient to make a difference [25].

Therefore, the main purpose of this study was to compare the validity of the TOV method using a FP versus the FT method using a FP and a smartphone application to measure jumping performance or neuromuscular fatigue-overload in professional female football players.

#### **2. Materials and Methods**

#### *2.1. Participant Selection: Inclusion and Exclusion Criteria*

A total of eight healthy professional female football players of the First and Second Spanish Football Divisions (Liga Iberdrola and Reto Iberbrola, respectively) (aged 27.25 ± 6.48 years; body mass 56.73 ± 4.86 kg; height 1.61 ± 0.06 m. Values expressed as mean ± standard deviation) participated in this study.

Inclusion criteria were: (1) playing in a female professional soccer league; (2) not having suffered a musculoskeletal injury one year prior to the date of the protocol (i.e., checked through a previous exclusion questionnaire); (3) not presenting any cardiovascular, musculoskeletal, and/or neurological disease, nor previous ones that could affect participation in the study. Exclusion criteria were: (1) aged younger than 18 years; (2) having consumed any narcotic and/or psychotropic agents or drugs during the test. We have selected this specific study population to homogenize the level of the sample and to deepen the knowledge of women's professional soccer.

At the study outset, participants were informed of the study protocol, schedule, and nature of the exercises and tests to be performed before signing an informed consent form. The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of the Technical University of Madrid (Madrid, Spain).

#### *2.2. Experimental Design*

This study consisted of a single evaluation session (see Figure 1) in a laboratory in the same time frame to avoid the detrimental performance effects associated with circadian rhythm [26]. Subjects were required to avoid physically demanding activity in the 24 h prior to the session.

Three valid trials of a CMJ and SJ were conducted with each subject, with 60 s recovery between trials, and three minutes between each jump test (see Figure 1). All jumps were measured at the same time with the mobile application and on the FP.

The sessions began with a 10-min general warm up consisting of continuous running, specific running, joint mobility, and ballistic stretching exercises, followed by a specific pre-test warm-up where participants performed five SJs and CMJs with 30 s between each jump. After three minutes of rest, the participants started the jumping test.

**Figure 1.** Experimental design. CMJ = countermovement jump test; SJ = squat jump.

#### *2.3. Vertical Jump Test*

The VJs included in the present study were the CMJ and SJ, which have been previously demonstrated to be reliable measures (intraclass correlation coefficient (ICC) = 0.97 and 0.96, respectively) [27]. All participants were completely familiarized with both techniques (i.e., they realized these tests during the soccer season). Moreover, they practice before the test during the specific warm-up (see warm-up description above). Participants were always instructed to jump as high as possible, keeping their hands on their hips and that their legs should remain straight during flight, making contact with the ground with the tips of their feet and knees extended [28]. In both modalities, the aim was to reach the maximum height by means of the jump; nevertheless, each modality of jump consists of different characteristics [29].

The CMJ (Figure 2a) initial position consists of a static standing position with hands on the hips. From this position, a continuous and fast triple hip, knee, and ankle flexion movement is executed until reaching ≈ 90◦ of knee flexion, followed by the triple extension of the same joints in a fluid, fast, and continuous way [29]. In this type of VJ, there is a stretching-shortening cycle (SSC), which takes place during the consecutive eccentric, isometric, and concentric phases [29].

**Figure 2.** (**a**) Countermovement Jump; (**b**) Squat Jump.

The SJ (Figure 2b) only presents the concentric component. The initial position consists of maintaining the hips and knees flexed at ≈ 90◦ for approximately four seconds to avoid countermovement and the elastic component. From this position, the concentric phase of the jump with an explosive extension of the lower limb joints is performed. All participants were asked to jump as high as possible and without performing a SSC [29]. The SJ forces before takeoff were checked on the FP to ensure that the participant did not perform countermovement.

For both jumps, participants were asked to take off and land at the same place to avoid lateral or horizontal displacement. Only successful trials were considered. Participants were asked to repeat the trial if a jump was incorrectly performed.

#### *2.4. Instruments*

#### 2.4.1. Force Platform

Two 600 mm × 400 mm piezoelectric platforms (Type 9286AA; Kistler Instruments AG, Winterthur, Switzerland) mounted together on the floor according to the manufacturer's instructions, were used in this experiment. The force sensors of these platforms were constituted by piezoelectric transducers 5 kN, total max 20 kN. Data were recorded at a sampling frequency of 1000 Hz. The FPs were connected to a portable computer with specific software, the Measurement, Analysis and Reporting Software "Kistler MARS" (Kistler Instruments AG, Winterthur, Switzerland).

#### 2.4.2. Smartphone Application

The validated smartphone application used to measure the performance of the jump was My Jump2 (APP) [30] installed on a mobile phone (iPhone 6s Apple, Cupertino, CA, USA) at a sampling rate of 240 Hz. My Jump was designed for analyzing VJ measuring the time (in ms) between two frames selected by the user and subsequently to calculate the JH using the equation based on FT. The position of the camera was always at ~1.5 m from the middle line of the subject so that it was constantly aligned with the reference joint points.

After the recording, one researcher manually selected the frames in which the subject performed exactly the moments of take-off and landing. A second person was consulted in case of doubt when analyzing the jumps with the APP. The JH was calculated based on the FT between the selected take-off and landing frame, as previously described [30].

#### *2.5. Statistical Analysis*

The normality and homogeneity of the data were analyzed using the Shapiro–Wilk and Levene tests, respectively. The normality and homogeneity of the dependent variables was confirmed (*p* > 0.05); all of the data are provided as their means and standard deviations.

Various statistical analyses were used to prove the APP validity and reliability in comparison with the FP using FT and TOV to calculate JH. The Pearson correlation coefficient (r) was used to calculate the concurrent validity between the APP and the FP using different measurement methods.

To expound the magnitude of the relationship between height measurements by the APP and the FP, Cohen's convention was used. To measure reliability between the APP and the FP, the coefficient of variation (CV), an absolute agreement Intraclass Correlation Coefficient (ICC), and Cronbach's Alpha were used. The CV was used to observe the uniformity of the values with respect to the mean. A Bland–Altman plot was created to graphically represent the agreement between the APP's measured heights and FP FT measured heights.

The mean differences between the measurements for each jump (SJ and CMJ) and for both jumps together (CMJ + SJ) were calculated through a one-way ANOVA, and Tukey's post-hoc test was used to analyze pairwise comparisons between means. The standard error of estimate (SEE) was used to show the typical error in measurement. Significance was set at *p* < 0.05. All statistical analyses were performed using the software package SPSS® version 25 (IBM Co., USA).

#### **3. Results**

The coefficients of variation values were very small for the CMJs with the FP through the TOV, the FP through the FT and the APP (2.82%, 3.56%, and 3.55%, respectively), and for the SJ (12.44%, 5.24%, and 4.88%, respectively).

The Pearson's correlation showed a poor relationship and a small reliability between the JH measured through the FT and the TOV, both measured with the FP (r = 0.028, *p* > 0.84, ICC = −0.026, 95% CI = −0.276–0.24) (Figure 3), and between the JH measured with the FP through the TOV and the APP (r = −0.116, *p* > 0.43, ICC = −0.094, 95% CI = −0.314–0.157) (Figure 4).

**Figure 3.** Pearson's correlation between jump heights measured by the FP using take-off velocity and the FP using flight time; FP, force platform; graph made by the authors.

**Figure 4.** Pearson's correlation between jump heights measured by the FP using take-off velocity and the APP using flight time; FP, force platform; APP, My Jump App; graph made by the authors.

On the other hand, a high and significant relationship between the JH measured with the FP through the FT and the APP (r = 0.872, *p* < 0.01) was found (Figure 5). Moreover, there was a very high agreement between these two variables as revealed by the ICC = 0.843 (95% CI = 0.681–0.919) and Bland–Altman plots (Figure 6).

**Figure 5.** Pearson's correlation between jump heights measured by the FP using flight time and the APP using flight time; FP, force platform; APP, My Jump App; graph made by the authors.

**Figure 6.** Bland–Altman plot between the FP using flight time and the APP using flight time to measure jump height. The central line represents the systematic bias between instruments, while the upper and the lower lines represent ± 1.96 SD; FP, force platform; APP, My Jump App; graph made by the authors.

The one-way ANOVA shows a significant difference between the JH measured with the FP through the TOV versus the APP (*p* = 0.002), and a trend versus the FP through the FT (*p* = 0.052). Finally, the JH analyzed with the FP through the FT and the APP did not differ (*p* = 0.457) (Table 1). The eta squared of the one-way ANOVA was η2 = 0.085.

Specifically, in SJ, the one-way ANOVA shows a significant difference in JH measured with the FP through the TOV versus the FP and the APP both through the FT (*p* < 0.01). Finally, the JH analyzed with the FP through FT and the APP did not differ (*p* = 0.826) (Table 2). The eta squared of the one-way ANOVA used to compare SJ was η2 = 0.374.


**Table 1.** Values of the different methods for measuring jump height combining CMJ and SJ.

Data are presented as mean ± standard deviation. FP–TOV: Force Platform–Take-Off Velocity; FP–FT: Force Platform– Flight Time; APP: My Jump Application (flight time). \* Significant difference.



Data are presented as mean ± standard deviation. FP–TOV: Force Platform–Take-Off Velocity; FP–FT: Force Platform– Flight Time; APP: My Jump Application (flight time); SJ: squat jump. \* Significant difference.

Lastly, in CMJ, the one-way ANOVA did not show significant differences between JH measured by the different methods (*p* = 0.07, η2 = 0.073) (Table 3).

**Table 3.** Values of the different methods for measuring CMJ jump height.


Data are presented as mean ± standard deviation. FP–TOV: Force Platform–Take-Off Velocity; FP–FT: Force Platform– Flight Time; APP: My Jump Application (flight time); CMJ; countermovement jump.

#### **4. Discussion**

The main purpose of this study was to compare the validity of the TOV method using a FP versus the FT method using a FP and a smartphone application to measure jumping performance or neuromuscular fatigue-overload in professional female soccer players.

The major findings were the lack of reliability and validity between JH measured with the FP through the TOV and JH measured through the FT either by the FP (r = 0.028, *p* > 0.84, ICC = –0.026 (–0.276–0.24)) or the APP (r = 0.116, *p* > 0.43, ICC = −0.094 (−0.314–0.157)). To our knowledge, this is the first study that has researched the JH of SJ and CMJ together (SJ + CMJ) with these technologies: FP trough the TOV and trough FT, and APP through FT. Two previous studies [20,31] have analyzed the SJ, CMJ and CMJ with arms (CMJA), together comparing JH with FP vs. contact mat (both through FT method) [20], and with FP (TOV and FT methods) vs. contact mat (FT method) [31]. This latter study [31] concluded that the TOV must be used instead of the FT as a more valid and accurate way to calculate the VJ performance. It provides a more sensitive approach to scientifically establishing jump determinants, greater sensitivity in analyzing changes in performance factors after training (i.e., overcompensation, overload), and greater accuracy in monitoring performance within the session (i.e., fatigue) [18,19]. Further, the mean comparison shows a significant difference between the JH measured with the FP through the TOV versus the APP (*p* = 0.002), and a trend versus the FP through the FT (*p* = 0.052). To our knowledge, there is only one previous study that researched the differences between the FP through the TOV versus a smartphone application (FT method) [19], and it showed near perfect correlations (ICC = 0.996; *p* < 0.001). These differences with our results may be explain by some reasons: 1) In our study the subjects performed three different CMJs and SJs, while in the previous study the subjects performed five CMJs [19]; (2) The distance between the researcher and the subject was of 1 m in the previous study [19], while in our study it was of 1.5 m; (3) In addition, rest

between jumps was not specified in the previous study [19], so it is not possible to know if fatigue could have affected their results through the FT method (i.e., Different joint position between take-off and landing), whilst our subjects rested 60 s between trials and three minutes between each jump test. Despite these methodological differences, their results comparing the FP through the TOV versus a smartphone application (FT method) regarding only CMJ did not differ with our CMJ results. Based on the present study and the literature [19,20], it seems that the flight time method, independent of the technology used (e.g., FP, contact mat, smartphone application), is an imprecise means of assessment.

On the other hand, there is a high correlation and good reliability between the JH SJ+CMJ measured through the FT with the FP and the APP (r = 0.872, *p* < 0.01, ICC = 0.843, CV = 4.88%). These results are in accordance with other studies that compare the SJ+CMJ+CMJA height using a FP and a contact mat (both through FT) (r = 0.99, *p* < 0.001, ICC = 0.985, CV% = 2.7) [21]. Additionally, the mean comparison of the present study shows that the JH analyzed with the FP through the FT and the APP did not differ (*p* = 0.457). Therefore, the APP and the contact mat seem to equal the accuracy of the FP with the FT method. However, it is important to highlight the fact that some studies compare the My Jump APP based on FT, with other instruments also based on FT (i.e., contact mat or FP through the FT) [21,30,31], instead of comparing it with the FP through the TOV (gold standard) [16]. Therefore, all these technologies that use the FT method are not as precise as the TOV method. Bearing in mind that the most accurate and considered the gold standard method to measure the height of the jump is the FP based on the TOV [19,20], these are the results that should be trusted.

Analyzing specifically the different jumps, it was found in SJ significant differences between the JH measured with the FP through the TOV and the FP and the APP through the FT (*p* < 0.01 in both cases). However, we found no significant difference comparing the JH measured with the FP and the APP (both through the FT) (*p* = 0.826). Further, CMJ data showed no significant differences among the three methods (i.e., FP through the TOV versus the FP and the APP through the FT) (*p* > 0.05 in both cases). Our results are in line with the only one previous study that researched the differences between the FP through the TOV versus a smartphone application (FT method) [19], and it showed near perfect correlations (ICC = 0.996; *p* < 0.001). Results of the reliability and validity obtained in the present study comparing the CMJ height through the FT with the FP and the APP (r = 0.872, *p* < 0.01, ICC = 0.843 and CV = 4.88%) are in accordance with the first validation study of My Jump App [30] that compare the CMJ height using a FP through FT (r = 0.995, *p* < 0.001, ICC = 0.997, CV% = 3.5), and other study that compare the CMJ height using a contact mat and the My Jump App (both through FT) (r = 0.999, *p* < 0.001, ICC = 0.948, CV% = 10.096) [21]. Despite the differences between the sampling frequency of the devices (i.e., FP ≈1000 Hz and the smartphone camera ≈ 240 Hz), the height values obtained were very similar between the two methods, as reflected by the ICC.

The literature that evaluates the smartphone applications in this area has analyzed the CMJ as the only vertical jump option discarding the SJ [19,21]. However, with this study it has been demonstrated that the results of the CMJ cannot be extrapolated to the SJ. The height of the SJ and CMJ is defended by several authors as a method to detect fatigue and individualize load prescriptions, so the reduction of the height of the VJ can be interpreted as evidence of the deterioration of neuromuscular function (i.e., fatigue, neuromuscular overload) [12]. Its measurement could provide a relatively simple and accessible tool for quantifying the degree of neuromuscular, mechanical, and metabolic fatigue [5,12,32]. This fatigue can remain for up to 48–72 h after a high-intensity session [33]. It could therefore provide useful feedback to coaches to determine the impact of training loads on player recovery and performance [12]. The neuromuscular factor is considered the most influential factor in training and is also an intrinsic risk factor of an athlete's injury [34]. The lack of strength and poor coordination capacity are part of it [35]. Several studies have observed that after a fatiguing exercise bout and during the landing phase of different jump types, there is a change in the neuromuscular control strategies used by these subjects [36,37]. These motor control alterations are associated with different injury risk factors, such as reduced knee and hip flexion, increased knee valgus, increased ground reaction force, and greater stabilization time [37]. In high-level sports, measurement accuracy is critical in the assessment of performance and in the control and assimilation of training loads to avoid muscle–tendon overloads [25]. In addition, some authors recommend the evaluation of jumping performance, especially at high and elite levels [24]. In this sense and referring to the exclusive sample of this study, as they are professional female soccer players, these small variations described above should be considered and limited as far as possible. In our study, the differences in JH between the FP through the TOV versus through the FT was 5.23% (1.84 cm) and the differences between the FP through the TOV versus the APP was 7.90% (2.78 cm). The only previous study that researched the differences between the FP (TOV method) versus a smartphone application (FT method) [19], found a small overestimated JH obtained from the APP of 0.78% compared to the FP based on TOV [19]. This small difference (0.78%) is more than ten times lower than the differences reported by us (7.90%), more than six times the difference between FP–TOV vs. FT methods (5.23%), and more than three times lower than the differences between the FP (FT method) and the APP (2.8%). This latter difference reported in the present study between the FP (FT method) and the APP agree with other studies comparing the My Jump application and a FP (FT method) [30], and an infrared contact mat and a FP (both through the FT method) [19]. More research regarding the issue is needed to clarify if smartphone applications may replace a FP (TOV method) in professional environments where accuracy is a key factor (i.e., high-level sport, medical clinic).

Furthermore, we observed that the difference of JH between the FP and the APP both through the FT was smaller, only 2.8% (0.94 cm). These latter results are in accordance with previous studies that observed differences in JH between an infrared contact mat and a FP, both through the FT of ≈ 2.5% (1.06 cm) [19], and between the My Jump application and a FP both through the FT (1.1 ± 0.5 cm) [30]. These concordances are produced because they use the FT method (i.e., less accurate) to compare different technologies. When the FT method is used with any technology, even the FP, there are many possibilities to produce measurement errors because the takeoff and the landing positions are different (e.g., different degree of lower limbs joint/s flexion) [38]. Then, it seems recommendable that high-level sportswomen and men should be assessed with this gold standard technology and method (i.e., FP and TOV method) to ensure correct performance and/or overload control during the sport season.

#### **5. Conclusions**

The results of the present study showed that in female professional soccer players, there is a lack of validity and reliability between JH in SJ + CMJ and SJ calculated with the TOV method versus the FT method. It seems that only the TOV measured with a FP could guarantee the accuracy of the jump test.

In order to be able to generalize our conclusions to different athlete populations, future research is needed, aiming at comparing the validity and reliability between JH of different kind of jumps calculated with the TOV method versus the FT method on high-level and recreationally trained athletes from different sport modalities and both sexes.

**Author Contributions:** Conceptualization, E.A.-C., J.A.B.-M. and A.F.S.J.; methodology, E.A.-C., J.P.B., J.A.B.-M., E.N. and A.F.S.J.; validation, E.A.-C. and A.F.S.J.; formal analysis, E.A.-C., J.P.B. and A.F.S.J.; investigation, E.A.-C., J.P.B., J.A.B.-M. and A.F.S.J.; resources, E.A.-C., E.N. and A.F.S.J.; supervision, E.N. and A.F.S.J.; project administration, E.N. and A.F.S.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Applied Sciences* Editorial Office E-mail: applsci@mdpi.com www.mdpi.com/journal/applsci

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-2609-6