Correlation between non- -dominant foot placement variability with acceleration and agility performance in handball players

The horizontal acceleration and agility are both an essential part of sports performance in handball. The foot placement variability has been shown to vary between differences in speed and direction of running. However, no studies have considered the connection between the foot placement variability, forward acceleration and agility performance. In the present study, a proposed repetitive single leg countermovement drift test was analyzed as a measure of the foot placement variability and correlated with forward sprinting and agility performance in handball players. Twenty-three male handball players performed a series of 10 consecutive single leg countermovement jumps, 10 m forward acceleration sprint, a handball adapted T-agility test and a single leg countermovement jump test. Correlations between the drift area, T-agility test duration, acceleration performance are measured as well as the time used to accelerate at a 5 m and 10 m distance and the height of the single leg countermovement jump. The drift area for the non-dominant leg had high and significant negative correlations with the first 5 m of forward acceleration sprinting and a positive correlation with the T-agility test to the non-dominant side. The countermovement jump height had no significant correlation to any other parameters. This data indicates that the foot placement variability of the jumps performed on the non-dominant leg could be an indicator of the ability to control stability during short forward acceleration sprints and a quick change of direction. On the contrary, these correlations disappeared when the direction change was performed under reaction time conditions.

The horizontal acceleration and agility are both an essential part of sports performance in handball. The foot placement variability has been shown to vary between differences in speed and direction of running. However, no studies have considered the connection between the foot placement variability, forward acceleration and agility performance. In the present study, a proposed repetitive single leg countermovement drift test was analyzed as a measure of the foot placement variability and correlated with forward sprinting and agility performance in handball players. Twenty-three male handball players performed a series of 10 consecutive single leg countermovement jumps, 10 m forward acceleration sprint, a handball adapted T-agility test and a single leg countermovement jump test. Correlations between the drift area, T-agility test duration, acceleration performance are measured as well as the time used to accelerate at a 5 m and 10 m distance and the height of the single leg countermovement jump. The drift area for the non-dominant leg had high and significant negative correlations with the first 5 m of forward acceleration sprinting and a positive correlation with the T-agility test to the non-dominant side. The countermovement jump height had no significant correlation to any other parameters. This data indicates that the foot placement variability of the jumps performed on the non-dominant leg could be an indicator of the ability to control stability during short forward acceleration sprints and a quick change of direction. On the contrary, these correlations disappeared when the direction change was performed under reaction time conditions.

Correlation between non--dominant foot placement variability with acceleration and agility performance in handball players Introduction
The horizontal acceleration is an essential part of the performance in team sports such as handball. Different factors are contributing to the horizontal acceleration, especially in forward sprinting performance. These studies varied from biomechanical characteristics or technique (Haugen, Thomas, McGhie, & Ettema, 2019) to strength and power (Rumpf, Lockie, Cronin, & Jalilvand, 2016). The distance acceleration is an important element in regards to sprinting biomechanics. It has been shown that shorter acceleration at a 10 m distance, depends more on strength and power of the lower limbs compared to sprints, where maximum speed can be achieved (Kawamori, Nosaka, & Newton, 2013). This is also important for the acceleration and sprinting in handball where average acceleration distance is usually short, between 5 and 10 meters depending on the player's position and the action performed (Karcher & Buchheit, 2014). Therefore, it is important to note that the first few steps are vital to gain a tactical advantage over an opponent.
It has been shown that the stride length, contact time and horizontal force are important contributors to the acceleration ability at distances for up to 10 m (Haugen et al., 2019). In addition, accelerations at shorter distances have been shown to produce higher demands on the strength and power production ability of the lower limb musculo-skeletal system (Sonderegger, Tschopp, & Taube, 2016). However, this cannot be directly transferred to the ability to quickly change the direction of movement.
Moreover, the speed of forward movement has been shown to have no or moderate correlation to the speed of lateral movements (Swinton, Lloyd, Keogh, Agouris, & Stewart, 2014;Tomáš, František, Lucia, & Jaroslav, 2014) or agility measured with a T-test (Alemdaroğlu, 2012). This suggests that movement in different directions demands specific movement controlling strategies as suggested by more basic studies on motor control (Seethapathi & Srinivasan, 2019). This is also partially supported by studies assessing the effects of training on horizontal or vertical power production ability in sprinting and agility performance. Although final conclusions cannot be drawn, a trend of direction of the specific training transferring to an on-field movement is suggested (Fitzpatrick, Cimadoro, & Cleather, 2019).
It has been shown that the horizontal component of a push-off force vector directed towards the sprinting direction is an important determinant of a better sprinting performance (Haugen et al., 2019;Moir, Brimmer, Snyder, Connaboy, & Lamont, 2018). The direction of the force vector orientation could be dependent on the placement of one's foot relative to the horizontal position of the center of gravity (Mehdizadeh, Arshi, & Davids, 2014).
Moreover, the foot placement has been shown to vary during running and thus influences the stability of running (Arshi, Mehdizadeh, & Davids, 2015). However, no studies have systematically studied the role of the

Methods
In the present study twenty-three male handball players participated (22,4 ± 1,6 years of age, 94,8 ± 14,9 kg and 1,9 ± 0,08 m). All participants played in the EHF Champion league, had 4,1 ± 0,2 years of experience playing at the international level and had been involved in a systematic handball training regime for 10,7 ± 2,3 years. Participants played at different positions, but no goalkeepers were included. All the participants were required to sign an informed consent form. The study was performed according to the Declaration of Helsinki.
The exclusion criteria were: no reported locomotor system injuries in the last year, no neurological disease or vestibular pathology or any other disease/illnesses in the last 30 days. Instructions were given to the participants before the enrolment thus introducing them to the measurement tasks.

Measures
In the 10 m maximal forward acceleration task, intermediate times of first 5 m (t 1 ), second 5 m (t 2 ) and 10 m time (t 10 ) were measured using a photocell system (WittyManager 1.4.68., Microgate, Bolzano, Italy). Agility was measured using a modified T-agility test. Time Where Bh stands for the body height in meters and constant 0.565 is used to represent the relative height of the body's center of mass (Winter, 2009).
The single leg countermovement jump height (H) was calculated from the flight time, measured by the previously described photocell system.

Design and Procedures
On the first day, measurements of a 10 m maximal forward acceleration and an adapted T-agility test were performed, followed by a single leg countermovement repetitive jumping drift test and a single legged countermovement jump test on the second day. On each day, the measurements were preceded by a standardized warm-up, consisting of 10 minutes of low intensity running, performing five strengthening exercises for the lower limbs, torso and upper extremities. This procedure was followed by short stretches of the major muscle groups.
On the first day, participants performed a 10 m forward acceleration sprint test on a standard handball court.

Statistical Analysis
Statistical analysis was performed using the SPSS software (SPSS statistics 23, IBM, New York, United States). The correlations between different 10 m acceleration times and T-agility test times with other parameter were calculated using the Pearson correlation coefficient (r). The r was treated as no correlation for r < .3, small correlation for .3 < r < .5, medium correlation for .5 < r < .7 and high correlation for r > .7. The statistical significance (p) was set at p < .05.

Results
Results of the correlation analysis are presented in Table   1. High and statistically significant correlations were observed between Area _n and t 1 or Tt k_n . The Height of the countermovement jump had no correlation to any of the parameters describing 10 m maximal sprint or T-agility test. High and statistically significant correlations were also observed between t 1 and t 2 or Tt k_n .  p -statistical significance.

Discussion
In the present study, the foot placement variability measured with a drift during the repetitive single leg countermovement jumps proved to have an important correlation to sprinting and agility performance. The correlation between the Area _n and t 1 proved to be negative. On the contrary, correlation between the Area _n and Ttk _n was positive. Interestingly, no significant correlations were observed between the maximal single leg countermovement jumping height and t 1 or any Tt times.
As suggested by our hypothesis, the correlation between the results suggest that the increased foot placement variability is beneficial for a better performance in short forward acceleration sprints. This implies that the foot placement variability cannot be directly related to common propulsive force vector. It could be suggested, that it is to a larger extent connected to a constant demand to preserve stability of the body's center of mass trajectory during running. This is the primary variable that needs to be controlled during forward running at high speed (Arshi, Mehdizadeh, & Davids, 2015).
In addition, a human body is a complex multicomponent system. During a push-off, the muscular system of the trunk and the lower limbs must produce a coordinated and explosive contraction. This has to be accomplished with a locomotor system characterized by multiple degrees of freedom. For example, each joint of the lower limbs is Scatter plots for the statistically significant pairs of correlations are presented in Figure 1. able to move in more than one axis. This is controlled by multiple muscles. As suggested by the more recent motor control theory such as dynamical systems theory and theory of muscle synergies, muscles connect to muscle groups or muscle modes with the goal of effectively executing the task, enabling adaptability and the stability of the performance (Latash, 2010). The research using this concept has shown that simple movements can consist of more than one muscle mode. Such example is a simple forward and backward body oscillation task during an upright stance where two modes are concerned with puling the body forward or backwards respectively and the third mode is concerned with holding the mediallateral stability (Danna-Dos-Santos, Slomka, Zatsiorsky, & Latash, 2007). Studies of running indicate that similar strategies are used to control body's stability during forward or lateral running where anterior-posterior and medial-lateral stability are controlled differently (Arshi, Mehdizadeh, & Davids, 2015). This would suggest that medial-lateral stability during running is not necessarily connected to the forward propulsion. In addition, the foot placement variability has been hypothesized to be larger than the movement variability of the more proximal body parts during running which confirms its importance in controlling the proximal stability of the body (Mehdizadeh, Arshi, & Davids, 2014). On the other hand, studies analyzing isolated explosive tasks of the lower limbs have shown that muscle synergies must be abruptly destroyed in order to achieve an explosive increase in force (Sarabon, Markovic, Mikulic, & Latash, 2013) which could be the case in the push-off phase during acceleration. Considering the above theories, one could speculate that this abrupt falling apart of propulsive muscle synergies due to push-off, might also affect medial-lateral stabilization synergy. This fine-tuned control of medial-lateral stability synergy has not been studied up to date, but could provide an  Interestingly, no correlations between a single leg vertical jump height, 10 m forward acceleration and T-agility test were observed. This is somehow surprising as it is not in line with the general guidelines presented in the literature (Sonderegger, Tschopp, & Taube, 2016). In addition, no correlations were found between the foot contact position variability and the single leg vertical jump height. Our data indicates that the ability to control foot contact position variability is not necessarily linked to the propulsive force production and could be a measure of mechanisms responsible for controlling stability of the moving body. This suggests that forward acceleration is also dependent on the ability to preserve stability of the athlete's body rather than on the power production capacity.
As could be expected, the T-agility test where the change of movement direction was performed in a nonanticipatory manner had no correlation with any of the observed parameters. This is in line with the literature stating that an important aspect of agility is also the ability to perceive the in-game situation and react accordingly (Paul, Gabbett, & Nassis, 2016). This study indicates that, although the motor control perspective of changing direction of movement is important and should be studied further, the cognitive decision training represents an important, but separate entity. It would be interesting to better understand how these two properties interact together and affect sports performance as well as the injury prevention.
The present study presented important limitations such as the small sample size, possibly effecting the correlation assessment. Future studies should use a bigger sample size in order to gain higher statistical power. In addition, comparisons to athletes from other sports disciplines and a control group should also be made. Another important limitation was the nature of the repetitive vertical jumping test, which did not demand the production of horizontal push-off forces as it is the case in sprinting and agility tests used in this study. This discrepancy could have influenced the results and limited the applicability of the data. In the future, consecutive horizontal jumps should be studied in order to enable more specific testing and comparisons.

STATEMENT
In their statements, the authors confirmed the absence of any conflict of interest.