The Structure of Selected Basic Acrobatic Jumps

Biomechanical research in gymnastics has often been haphazard. Very little of the research reveals attempts to develop biomechanical “critical features” of the movement (Bartlett, 2009) for the analyzed skills. The purpose of most research appears rather to have concentrated on what is considered contemporary good performances, and occasionally, to compare those with worse performances. However, identification of these critical features is probably the most important task that a qualitative or quantitative analyst needs to face.

One of the primary tasks of sports biomechanics is the quantitative analysis of movement techniques to facilitate the learning of motor skills. Quantitative biomechanical movement analysis primarily uses temporal characteristics of mechanical variables (Droszez et al., 2016). However, to effectively communicate with athletes, coaches often refer to the so-called ‘movement features’ (Schnabel, 1998). These features are primarily used in a qualitative analysis of sports activities. Movement features can also be useful in sports technique studies, because they have certain defined measures. In this paper, as a basis for consideration and analysis, an eight-element, hierarchical and modified classification of movement features recommended by Schnabel (1998) was employed. Three groups of movement features can be distinguished here: the parent feature (the general basic structure), complex features (the rhythm, and the motion coupling), and the elementary features (fluency i.e., fluidity, accuracy i.e., exactitude, constancy i.e., repeatability, the rate of the movement, amplitude i.e., extensiveness of motion). The most informative and useful feature in the evaluation of

the movement, is the structure of the movement (Schnabel, 1998). To the observer, only the spatiotemporal structure of the movement is available. Assessment of this parent feature and ”generalized category”, which is based only on visual observation, is relatively simple, but contains little information (Schnabel, 1998). Therefore, research on sport techniques should cover both the causes of motion (the internal and external-kinetic structure of the movement) as well as the external-kinematic structure that shows the effects of motion, i.e. all temporal characteristics of the movement (Krol and Golas, 2017).

Biomechanics is the basic sports science focusing on movement techniques. This is especially important in gymnastics where performance techniques of the best gymnasts become models for others to follow. A coach’s ability to direct technical and physical training for these specific skills is enhanced when thorough descriptions of the skills are available (’movement features’, in particular ’critical features’).

In scientific research on movement techniques of motor skills and in recording and/or the measurement of different features and variables of movement, various accessible biomechanical methods are utilized. In previous studies, we applied a set of methodological research tools to study the structure of movement in the flat bench press (Golas and Krol, 2014; Krol et al., 2010), and acrobatic jumps (Krol et al., 2014, 2016; Krol and Klyszcz-Morciniec, 2017). An understanding of both the internal (muscle activation) and external (kinematic and kinetic) structure of these sport activities was acquired by simultaneous application of several devices, such as: a force plate, electromyography, and cameras. In this study, our goals were the same for the following basic standing ‘acrobatic’ jumps: a tucked backward somersault (TS), a piked backward somersault (PS), a backward handspring (HS), and a countermovement jump (CMJ).

The structure of these acyclic movements generally consists of the initial, main and final phases. However, for the jumps with rotation (TS, PS, and HS), the names of the phases are specific: countermovement, take-off, flight and landing phases. The countermovement is a specific type of the initial phase aimed to create optimal conditions for the implementation of the main phase. This is achieved by the pre-stretching of muscles of both legs (e.g. triceps, quadriceps, gluteus maximus) and the trunk (e.g. erector spinae pars lumborum). The pre-stretching of the muscles increases its elastic energy, which is called, the stretch-shortening cycle – SSC (McNeal et al., 2007). The SSC, according to Bartlett (2009), is a common sequence of joint actions in which an eccentric (lengthening) muscle contraction, or pre-stretch, precedes a concentric (shortening) muscle contraction. Both the take-off and the flight are the main phases since the main task is performed then. The main task involves body rotation about the free transverse axis. The purpose of the take-off is to provide the projection velocity needed to lift the body and the angular momentum required to perform a rotary motion. The flight includes the tucking action (during the ascent of the body) and the opening out action (during the descent). The effectiveness of the flight phase is determined by the skillful use of the principle of the conservation of angular momentum. The first aim of the landing is to overcome both the momentum and the angular momentum of the body. The second aim is to protect the joints of the lower limbs from injury and to restore the standing position.

Both male and female artistic gymnasts must be able to properly perform basic motor skills such as backward somersaults and backward handsprings, as well as countermovement jumps. Proper performance of these gymnastic skills requires an appropriate level of fitness and coordination abilities. These skills are usually performed as parts of acrobatic tumbling sequences, in which the take-off plays an essential role. Most of the complicated tumbling combinations are performed backwards due to the advantageous anatomical conditions for the backward take-off (Knoll, 1996). Backward takeoffs are one of the most important and frequently used components of a floor exercise routine in artistic gymnastics. The backward take-off initiates the linear and rotational impulses essential for the optimization of take-off velocities by attaining a large amount of kinetic energy necessary to achieve a suitable magnitude of momentum and angular momentum (Geiblinger et al., 1995). In the countermovement jump, the kinetic energy is necessary only to achieve a large magnitude of linear momentum. The focus of our analysis included sports technical solutions for energy production during the take-off phase for the following standing ‘acrobatic’ jumps: a TS, a PS, a HS, and a CMJ.

At the end of the countermovement phase there is a decrease in the angle between segments of lower extremities. The leg muscles (triceps, quadriceps, gluteus maximus) are stretched and part of the kinetic energy is transformed into potential energy of elasticity of leg muscles. This form of mechanical energy is then, by contraction of leg muscles, transformed back into kinetic energy which is at its maximum during the takeoff phase (SSC).

As with all ballistic movements, a ballistic movement is initiated by muscle activity in one muscle group, continues during a ‘coasting’ period with no muscle activation, and terminates by deceleration caused by the antagonist muscle group or by passive tissue structures, such as ligaments (Bartlett, 2009). In gymnastics, the mechanics for initiating acrobatic jumps follows a specific movement pattern. According to George (2014) prior to the jump, a vigorous upward-throwing action of the arms occurs which sets the stage for the kinetic chain sequencing of the legs. Since it is more difficult to start a movement of a body than to keep a body in motion, a powerful extension at the hips provides the fundamental force. This initiation of hip-joint extension is the cue to begin immediate, forceful knee-joint extension, which in turn is the cue to plantarflexion of the ankle joints. The specific hip-knee-ankle sequencing order must occur to ensure the most effective momentum transfer (Bobbert and van Ingen Schenau, 1988; George, 2014; Pandy et al., 1990). In our study this initiation of hip-joint extension has also become the basis for the separation of the countermovement phase from the take-off phase. The end of the take-off phase occurs when the vertical component of ground reaction force is zero, while the end of the flight phase follows at the instant of touch-down.

The most comprehensive studies on backward take-offs on the floor have been conducted by Knoll and Krug (1989), Hwang et al. (1990), Bruggemann (1994), Geiblinger et al.(1995), and Hedbavny and Kalichova (2011). Unfortunately, there is still a lack of professional research analyzing the full gymnastic movement structures in detail, i.e., both the internal and external (kinetic and kinematic) structure of the movement. Taking into consideration the existing knowledge and its deficiencies, the first aim of this study was to investigate the internal and external structure of basic acrobatic jumps.The second aim was to investigate the relationships that characterize these structures, according to the type of the acrobatic jump (with or without rotation). Based on previous studies (Krol et al., 2014; Krol and Klyszcz-Morciniec, 2017), we hypothesized that: 1) differences in the electro-activity pattern of the same muscles during three defined phases exist between acrobatic jumps with (TS, PS, HS) and without rotation (CMJ), 2) in the take-off phase of the TS, PS and HS the average muscle activity level of the biceps femoris is significantly higher and of the rectus femoris significantly lower compared to the CMJ. Understanding these characteristics could facilitate appropriate technical and physical training for gymnasts performing these simple standing acrobatic jumps.

The execution assessment of selected acrobatic skills was performed by comparing the tucked somersault (TS) with three other acrobatic jumps (CMJ, PS and HS).

Bioelectrical activity of muscles – internal structure of standing acrobatic jumps. Temporarily normalized RMS EMG [% of MVC] linear envelopes during four acrobatic jumps for eight muscle are plotted in Figures 1-3.

Figure 1

Figure 1

Tucked backward somersault vs. countermovement jump. A comparison of EMG patterns in the set of muscles between the tucked backward somersault and the countermovement jump performed by the same group of gymnasts indicates large differences (Figures 1A and 1B;Table 2). Statistically significant differences (p < 0.05) in average muscle activation levels of five muscles were especially evident in the countermovement and flight phases. In four cases these were the same muscles. The mean muscle activation levels were somewhat higher for the TS than the CMJ (during the countermovement phase it equaled for MG 69.8 ±54.4% MVC and 30.9 ±25.5% MVC, respectively; for RF 7.2 ±8.7% MVC and 3.4 ±2.7% MVC, respectively; for BF 13.3 ±7.7% MVC and 7.3 ±4.8% MVC, respectively; for RA 10.4 ±14.3% MVC and 3.3 ±2.5% MVC, respectively; and in the flight phase for AT 76.1 ±17.5% MVC and 35.2 ±24.7% MVC, respectively; for RF 56.9 ±25.4% MVC and 25.2 ±24.2% MVC, respectively; for BF 40.0 ±12.8% MVC and 15.7 ±8.9% MVC, respectively; for RA 76.0 ±37.9% MVC and 19.6 ±11.0% MVC, respectively). In two cases, the opposite was true, which in Table 2 is indicated by a minus sign. In the countermovement phase the anterior deltoideus activity was slightly smaller (p < 0.05) for the TS than for the CMJ (12.6 ±9.7% MVC and 14.7 ±9.2% MVC, respectively).In the flight phasethe medial gastrocnemius activity was also slightly smaller (p < 0.05) for the TS jump than for the CMJ jump(53.3 ±13.3% MVC and 73.7 ±34.8% MVC, respectively).

In the take-off phase of the TS and CMJ the average muscle activation levels of the rectus femoris and biceps femoris are particularly interesting (Table 2). In the TS, the EMG activity of the rectus femoris was significantly smaller (p < 0.05) than in the CMJ (21.7 ±26.5% MVC and 63.4 ±34.9% MVC, respectively). However, the biceps femoris activity in the TS was significantly higher (p < 0.05) than in the CMJ (77.5 ±27.6% MVC and 38.9 ±23.6% MVC, respectively).

Tucked backward somersault vs. piked backward somersault. A comparison of EMG patterns in the set of muscles between the tucked and piked backward somersault performed by the same group of gymnasts found them to be very similar (Figures 2A and 2B;Table 2). The Wilcoxon test demonstrated that the average muscle activation levels for all eight muscles in three subsequent phases between the TS and the CMJ did not differ statistically (p > 0.05).

Figure 2

Figure 2

Figure 3

Figure 3

Tucked backward somersault vs. backward handspring. As could be expected, statistically significant differences (p < 0.05) in average muscle activation levels were found only in the flight phase between the TS and the HS (Table 2). This is true for almost all muscles. However, only in the rectus femoris and rectus abdominis the average muscle activation levels were slightly higher (p < 0.05) for the TS in comparison to the HS (for RF 56.9 ±25.4% MVC and 14.6 ±11.9% MVC, respectively; for RA 76.0 ±37.9% MVC and 15.7 ±6.4% MVC, respectively). In contrast, the activity of other muscles, the biceps femoris, gluteus maximus, erector spinae, and anterior deltoideus, was significantly smaller (p < 0.05) for the TS than for the HS (for BF 40.0 ±12.8% MVC and 61.4 ±18.8% MVC, respectively; for GM 16.0 ±5.4% MVC and 77.4 ±43.6% MVC, respectively; for ES 25.7 ±8.5% MVC and 109.1 ±39.0% MVC, respectively; for AD 27.8 ±20.6% MVC and 81.8 ±38.5% MVC, respectively).

External structure of standing acrobatic jumps

External structures of the TS, CMJ, PS, and HS are shown in Figures 4-9. The external structure of acrobatic jumps consists of the kinetic structure i.e., the averaged and temporarily normalized vertical ground reaction force curve (Figures 4-6) and of the kinematic structure i.e., the averaged and temporarily normalized angle-time curves for the relative angles of upper and lower leg joints (Figures 7-9). A comparison of all temporal characteristics and the variables between acrobatic jumps (Table 3) was performed for the same group of gymnasts.

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Table 3

Comparative statistics of kinetic and kinematic variables of standing ‘acrobatic’ jumps (tucked somersault - TS; countermovement jump - CMJ; piked somersault - PS; Handspring - HS) from a force plate and the infrared cameras. Variables of all ‘acrobatic’ jumps were compared for the same group of gymnasts.

Variables				Wilcoxon Signed-Ranks Test
	TS vs. CMJ			TS vs. PS			TS vs. HS
	Parameters from the force plate
	N	T	p	N	T	p	N	T	p
Vertical impulse	9	-0	0.0077*	7	0	0.0180*	8	0	0.0117*
Horizontal impulse	9	20	0.7671	7	3	0.0630	8	-0	0.0117*
Vertical velocity	9	-0	0.0077*	7	0	0.0180*	8	0	0.0117*
Vertical	9	-0	0.0077*	7	0	0.0180*	8	0	0.0117*
displacement
	Parameters the infrared cameras
	N	T	p	N	T	p	N	T	p
Vertical velocity	11	-0	0.0033*	7	0	0.0180*		No data
Vertical displacement	11	-0	0.0033*	7	0	0.0630		for technical reason
Take-off angle	10	-0	0.0051*	7	0	0.0180*	8	0	0.0180*
Moment of inertia		In the CMJ		7	-1	0.0280*	8	2	0.0251*
Angular velocity				7	-0	0.0180*	8	-0	0.0117*
Angular momentum		is only the		7	-0	0.0180*	8	-2	0.0251*
		translation motion

„-” The minus sign indicates that the value of this parameter in the second of the compared jumps was significantly greater;

N- number of valid sample pairs; * - significance at p < 0.05.

This study presents the phase structure of the movement and is focused on the variables that can primarily affect the take-off by comparing four different acrobatic jumps. A large number of studies have focused on the biomechanical and neurophysiological variables involved in vertical jumps (Jacobs et al., 1996; Pereira et al., 2008; van Soest et al., 1993; Wikstrom et al., 2008; van Zandwijk et al., 2000) and their influence on performance. However, significantly fewer studies have been done on acrobatic jumps (McNeal et al., 2007; Mkaouer et al., 2012, 2014; Okubo, 2012).

Bearing in mind that biomechanics is concentrated on forces that act on a human body and the effects of these forces, the first thing to be considered is the muscle action. Skeletal muscles are the primary initiator of movement and are a real biological system designed to produce mechanical force and induce movement. Knowledge of the characteristics of muscle activation during the subsequent phases of the movement may enhance our understanding of how gymnasts perform these basic acrobatic jumps and may assist coaches and physical educators in prescribing appropriate training drills for particular jumps.

Studying the muscle activity of eight muscles in the three successive phases (the countermovement, take-off, and flight phase) of four standing ‘acrobatic jumps’(TS, PS, HS and CMJ) it can be stated that in most cases, the highest average muscle activation levels of these muscles were found during the take-off phase (Figures 1-3). The exceptions were the rectus femoris and the rectus abdominis, which in the tucked and piked backward somersault had the highest values in the flight phase. The difference of the rectus abdominis activity in the flight phase, between the TS and the CMJ was 56.4% MVC, which, with standard deviations, ±37.9 and ±11.0% MVC, respectively, was significant (p = 0.0033; Table. 2). A similar regularity was also found for the rectus femoris activity (p = 0.0164). The same phenomenon in the flight phase was observed in the rectus abdominis and the rectus femoris by Okubo (2012), who found that greater muscular activity was present during a tucked backward somersault than in a vertical jump. This is understandable due to rapid flexing in the hip joints during the flight phase of somersaults.

Jacobs et al. (1996) proposed that biarticular muscles were very important for jumping. These authors found that the rectus femoris transferred power from the hip to the knee. In contrast the kinematic analysis during vertical jumps showed that hamstrings were involved in the power transfer from knees to the hip. The result of the simultaneous contraction of these antagonist muscles is the net transfer of force from the hip to the knees. Taking this into consideration, the current findings are especially interesting. In the take-off phase of acrobatic jumps with rotation (TS, PS and HS) the average muscle activation levels of biarticular biceps femoris were significantly higher (p < 0.05) and biarticular rectus femoris significantly lower (p < 0.05) than in the countermovement jump. This means that these muscles have different roles. In the TS, PS and HS the role of the rectus femoris muscle changes from a prime mover to a supportive-prime mover (a stabilizing muscle). Conversely in the CMJ, the rectus femoris acts as a prime mover and the biceps femoris acts as a stabilizing muscle. Therefore, electromyography is a dynamic type of assessment for neuromuscular control, which quantifies preparatory and reflexive muscle amplitudes around a particular joint, which is an important indicator in determining joint stability during functional tasks (Ebig et al., 1997; McKinley and Pedotti, 1992; Pereira et al., 2008; Wikstrom et al., 2008). However, an equally important indicator of muscle activation is the possibility of movement in the joint.

As expected, there were no statistically significant differences between the tucked and piked backward somersault in muscle activation levels of all eight muscles during the three successive phases.

There were statistically significant differences (p < 0.05) between the TS and HS in mean muscle activation levels of almost all muscles in the flight phase. In one muscle (medial gastrocnemius) a statistical tendency (p = 0.0687) was found. The high muscle activation levels of almost all muscles are probably associated with the stabilization of joints during the flight phase in the HS.

Changes in the myoelectrical activity of the prime movers may be associated with the take-off angle in particular jumps (Table 4). Decreasing the take-off angle in the TS, PS and HS, respectively, in comparison to the CMJ (p < 0.05) corresponded to greater activity of the biceps femoris and erector spinae, and less activity of the rectus femoris. Although the differences between average muscle activation levels of these prime movers were usually small and the standard deviations large, with respect to the impulse values from ground reaction forces (Table 4 and Results) one should suppose that these muscles are fundamental in shaping the take-off mechanism.

It is important to note that to jump vertically, there should be no horizontal component of the GRF. In fact, in order to induce the backward rotation of the body, the vertical push's centre of pressure (COP) should be in front of the body's COM (De Jaeger et al., 2004). However, if the direction of the force is not towards the COM this could lead to some ‘wasted effort’ and therefore, insufficient height may be reached. Is it even possible to exert a vertical push which is not directed towards the COM without inducing a horizontal force? To answer this question, the values of horizontal and vertical components of the GRF would have to be evaluated.

A comparison of the acrobatic jumps with rotation showed that in the take-off phase of the TS a significantly higher vertical impulse was created than in the PS and especially in the HS. Obviously, in the countermovement jump the value of the vertical impulse was the greatest. Leboeuf et al. (2003) mentioned that if a backward somersault was performed correctly, the force impulse would be around 200 N·s. Their values were much higher than those obtained in the current study. Our female gymnasts’ impulse ranged between 133.6 ±28.4 N·s and 128.1 ±25.4 N·s for the TS and PS, respectively (a comparison was performed for the same group of gymnasts). This difference may also be related to the fact that our gymnasts performed standing backward somersaults whereas in the Leboeuf’s et al. (2003) study, they performed it after a descending jump.

The opposite was found with the horizontal impulse. The smallest magnitudes of the horizontal force impulse were found in the countermovement jump and in the tucked somersault, followed by the piked somersault and handspring, respectively. As expected, the horizontal displacement was the largest during the HS, caused by a take-off of the COM and allowing rotation, as described by Medved et al. (1995) and Munkasy et al. (1996).

The magnitude of the horizontal displacement, the flight height and the velocity of the COM which was reached at the end of the takeoff phase result from the force impulses. Of course, the obtained values of these variables are arranged as in the case of the vertical and horizontal impulses.

In our study, the values of the COM vertical velocity are about half that of Hraski’s (2002) results. In Hraski's research, however, the backward somersault was performed after a typical acrobatic sequence: a run up, a round-off, a backward handspring, and the subject was a highly ranked, world class gymnast.

According to Lukjan and Parlak (2005), the smaller magnitude of vertical velocity, and thus, the flight height in somersaults, compared with the countermovement jump, are affected by:

Commencing the take-off phase by extending the hip joints before the knee joints and incomplete knee joint extensions at the beginning of the flight phase of the somersault was also confirmed in our study (Figure7). There were some differences, however, in the duration of subsequent phases, which may have resulted from the slightly different method of determining the beginning and the end of the take-off phase. In our study the beginning of the take-off phase was accepted as the moment of the return of the movement in the hip joints, after the countermovement phase. The end of this phase was accepted as the moment that the feet lost contact with the ground. In a study by Lukjan and Parlak (2005) the probable beginning of the take-off phase was at the lowest momentary location (position) of the COM at the end of the countermovement phase. The temporal ratio of the successive phases (the rhythm of the movement; Schnabel, 1998), as could be anticipated, was different for each of the jumps (Table 5).This was the most evident in the backward handspring. In the HS, the take-off phase was clearly longer, while the flight phase significantly shorter (almost two times shorter) than in the somersault and the countermovement jump.

The difference between the flight height in the standing countermovement jump and the standing somersault may be an indicator of the efficiency of the mechanism giving the angular momentum (Lukjan and Parlak, 2005). It can be assumed that the smaller this difference, the better the take-off to perform a standing somersault. In this way, the mechanism giving the angular momentum, which results from the similar nature of the take-off in both forms of jumping, is understood.

Besides the vertical velocity of the COM the angular momentum is another critical feature of the movement, and tends to be reversed. The great magnitude of the vertical velocity in the tucked backward somersault corresponds to the small angular momentum, while the small magnitude of the vertical velocity in the piked backward somersault corresponds to the high angular momentum. The magnitude of the angular momentum attributed to the COM at the take-off phase was determined during the flight phase – immediately after the feet lost contact with the ground.

The angular momentum achieved by gymnasts in our research for the standing tucked somersault was 53.1 ±8.2 kgm2/s, which was about 5.1 kgm2/s more than in the study by Lukjan and Parlak (2005). In the standing piked somersault, gymnasts achieved an angular momentum of about 19.9 kgm2/s greater than the athlete in the Lukjan and Parlak’s study (2005) in the tucked somersault.

In the study by Hwang et al. (1990), the magnitudes of the angular momentum were more than twice our results. However, in their study, seven top-class athletes performed the double somersault after a running start. The results of our study are consistent with the Hraski's thesis (2002) which states that greater angular momentum during the flight phase corresponds to greater horizontal velocity and smaller vertical velocity during the take-off phase.

As expected, the take-off that passed through the COM allowed for a better amplitude of movement than the take-offs thrown off the centre forward or backward. The CMJ showed the highest level of vertical impulse, velocity, and displacement followed by the TS, PS, and HS. This implies that for better performance of acrobatic jumps, it is necessary that the force passes close to the COM. This explains why in the take-off phase of acrobatic jumps with rotation (TS, PS and HS) the average muscle activation levels of the biarticular biceps femoris were significantly higher (a prime mover) and of the biarticular rectus femoris significantly lower (a stabilizer) than in the countermovement jump (CMJ). Thus, the hypothesis accepted in the introduction was confirmed. In practice, we recommend that coaches carefully monitor the position of a gymnast’s shoulders. Leaning backwards during the take-off while performing a standing backward acrobatic jump with rotation should be avoided.

Finally, although many variables influence sports success, including mental and physiological factors, biomechanical considerations reflected in correct or incorrect technique are crucial. This is especially true in such sports as gymnastics or figure skating. Despite the extensive use of acrobatic jumps in many forms of training regimes, there is still a paucity of published research directed toward the mechanical understanding of these movements. Gymnasts in particular, must master the following technical skills since they are included in most acrobatic jumps/skills: the ability to gain height, the ability to rotate, increasing or decreasing rotation by altering body configuration.

To better understand the cause-and-effect relationships between biomechanical factors, a computer simulation is often used (Yeadon et al., 1990). This is done through systematic manipulations of the key performance factors (critical features e.g. the vertical velocity of the COM, the take-off angle, and the angular momentum around the COM at the moment of the take-off). The quality of the simulation results depends on the accuracy of the input data and the complexity of the model used. This first factor depends, inter alia, on the class of the measuring devices. In our study, we used modern comprehensive research methodology consisting of eight pairs of surface electrodes, six infrared cameras and a force plate, all of which were synchronized with each other (multi-modular measuring system SMART-E). This system allowed the internal and external structure of acrobatic jumps to be explored. The often subtle differences between the mechanical variables of motion can also be recorded and a technique assessment can be conducted. Sometimes just small differences that cannot be initially noticed may strongly influence performance of certain gymnastic skills.