117 86 4MB
English Pages 186 [187] Year 2020
Agility in Sport
Agility in Sport By
Jaromír Šimonek and Pavol Horička
Agility in Sport By Jaromír Šimonek and Pavol Horička This book first published 2020 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2020 by Jaromír Šimonek and Pavol Horička All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-5275-4096-0 ISBN (13): 978-1-5275-4096-5
TABLE OF CONTENTS
Preface ....................................................................................................... vi Introduction ................................................................................................ 1 Chapter One ................................................................................................ 3 Definition of terms Chapter Two ............................................................................................. 19 Reactive versus pre-planned agility Chapter Three ........................................................................................... 33 Testing of agility Chapter Four ............................................................................................. 56 Age-related changes in agility Chapter Five ............................................................................................. 75 Sport-specific agility performance Chapter Six ............................................................................................... 88 Training of agility Chapter Seven......................................................................................... 170 Recommendations for coaches References .............................................................................................. 171
PREFACE
The purpose of this book is to introduce a new perspective of agility theory, a practice of training and testing. Recently, there has been a great deal of serious discussion concerning the methods of open-loop skills improvement in team games and combat sports. Experts have discussed methods of developing reactive and running agility. However, there is a lack of experimental work to prove these theories in sports practice. This book offers experimental research results as well as theoretical knowledge of both types of agility. Agility training methods and exercises are presented in the penultimate chapter of the book. This material can be used in training by coaches and trainers in sports games and combat sports. The lack of research samples forms the limitation of the book. In one sentence, describe your book: The book presents a comprehensive view of agility performance in sport. The comprehensiveness of elaboration on the topic is unique. It is based on the personal testing experience of the authors. Personal trainers, sports coaches, P.E. teachers, professional athletes, sports students at universities, recreational athletes, will be the potential audience for this book.
INTRODUCTION
Recently, a top sport has been characterized by high sports performance, perfect technique, high levels of motor skills and abilities, and as one that places high demands on the motor, psychological and physiological aspects of an athlete´s personality. Thus, the question of increasing effectiveness of sports preparation has come to the forefront. The core content of sports preparation has gradually passed from quantity to quality, from general to specific means. Trainers and coaches have been searching for more effective means of developing skills and abilities, focusing also on the more effective exploitation of training time. The quality of sports training rests on the exploitation of “sensitive periods” for the development of motor prerequisites crucial for the given sport. Sports games require a high level of specific movements, represented by perfect mastering of the technique of individual skills. The ability to move quickly, and to change direction and speed of movement while accelerating and decelerating, belong among the core ones. However, the athlete would not be successful if they did not use their cognitive skills to react to the constantly changing game situations in a match. Athletes with high levels of anticipation and mental processes, as well as high speeds of decisionmaking, show improved effectiveness in their motor performance. Similar to sports games, martial arts are sports where athletes´ degree of successfulness depends, besides perfect mastering of the technique of skills, on their ability to regulate their mental processes. Reactive agility also belongs among the fundamental qualities. Experts in the field of conditioning have elaborated a new theory of training of agility in both its forms: running and reactive ones. This book presents a comprehensive view of agility performance in sport. It has been designed for personal trainers and coaches in sports games and martial arts, for P.E. teachers in schools, professional athletes, sports students at universities, recreational athletes, researchers, as well as for the public interested.
2
Introduction
The aim of this work is to explain, to trainers and coaches, the difference in reactive and running agility and offer them methods of training agility separately, with a specific focus on the core skills in the given sport. The task of this monograph is to submit different training means, focusing on the development of agility for the stage of initial sports preparation and initial specialization as well as the specialized sports preparation stage. The application of the above-mentioned means should result in the improved motor display of a sportsperson, faster acquisition of motor skills, as well as a higher percentage of effectivity in a game.
CHAPTER ONE DEFINITION OF TERMS
In this chapter, the authors attempted at defining basic terms used in the area of sports preparation. Some parts of this chapter were adapted from the monograph of one of the authors (Šimonek, 2014) which was published in De Gruyter Publishing. Sports preparation itself represents a complex system of phenomena, connections, and behaviour of its components. If the coach wants to reach the goal of sports preparation – to nurture a top athlete able to reach an optimum performance in any conditions – he/she has to try, based on thorough knowledge of the individual, of specific age, gender, and developmental peculiarities, to create a comprehensive system of long-term sport preparation. This presupposes application of adequate and effective training means, methods and forms of work, optimum training loads and suitable frequency, and follow-up in training cycles (Šimonek, 2014). To solve this crucial task for the trainer (coach), they have to be familiar with the structure of performance in the given sport. Nowadays, rarely any sports expert doubts the contribution of factors like speed, quickness, explosiveness, the speed of frequency, coordination abilities to the sports performance in speed-strength events and sports games. In volleyball, a very high speed of reaction and reactive agility are required to be able to control balls on serve reception, especially in field defence. Many authors consider motor abilities, agility and explosive strength, along with pronounced longitudinal skeleton dimensionality, as the major characteristics for successful volleyball performance (Morales, 2002; Stamm, Veldre, Stamm, Thomson, Kaarma, Loko, & Koskel, 2003). Sports training focusing on the development of speed-strength and coordination abilities (mainly dynamic balance, reaction speed, spatial orientation) and reactive agility, contributes to an increase in sports performance and reduces the risk of injury. Unfortunately, many of the
4
Chapter One
known “classical” training programmes do not fulfil these aims. In reality, many athletes who have adhered to these old-fashioned training programmes do not reach required accruals and often suffer from recurrent injuries. Reasons for the above-mentioned conditions can be easily explained. All movement can be broken down into three planes of motion or directions – forwards and backward (the Sagittal Plane), side to side (the Frontal Plane), and rotational movement (the Transverse Plane) - and three muscle actions – acceleration (concentric), stabilization (isometric), and deceleration (eccentric). Most sports require the ability to move explosively in all three directions or to accelerate quickly, decelerate, stabilize functionally and accelerate explosively again. Yet, older and more ineffective forms of training have traditionally emphasized just one plane of motion (the sagittal plane – for example, sprints, squats, lunges leg presses and leg curls), and one muscle action – primarily acceleration. But functional movement and competitive sports are just not like this in practice, which is why close to 80% of all sports injuries occur without any contact with opponents, and usually when an athlete decelerates and rotates (such as during a change in direction). Athletes must train in all three planes of motion and with all muscle actions (acceleration, deceleration, stabilization) to create a much safer and more effective program. In addition, workouts should be both age-specific and sport specific. This is very important. A nine-year-old soccer player should not be using the same program as a fourteen-year-old basketball player or a nineteen-year-old hockey player. Trainers should construct a needs analysis of the sport- what are the dominant lanes of motion and muscle actions used by the sport and position of the athlete? We need to know the energy/endurance demands, the rest ratios, level of intensity demanded in each phase of the game and for each position. A program should be built around these components. Finally, the use of effective goal setting and training logs and charts to measure and monitor progress and improvements is an additional, often ignored, component that is very important to overall motivation levels, and thus to the overall success of any explosive speed, agility, and quickness program. Table 1.1 shows the position of indicators of functional preparedness of an athlete according to its importance in the given sport. In sports games and martial arts, we can find analysers in the first level of importance, while in the second one these are functional systems of the organism. It is inevitable that we focus our attention on these factors in the sports preparation of children and youth.
Definition of terms
5
Table 1.1 Distribution of indicators of functional preparedness of an athlete according to their importance (Nabatnikovová, 1982).
Level of importance Explosive Cyclic
Groups of sports Demanding a high degree of coordination
Martial Sports arts games
I
1,2
1,5,7,8,9
1,2,3,6
1,2,3,6
1,2,3,4
II
6
2,3,6,10
4,5,7,8,9
5,7,8,9
5,6,7,8,9
III
5
4
10
4,10
10
IV
3,4
Explanations: 1 – kinaesthetic analyser, 2 – vestibular analyser, 3 – visual analyser, 4 – acoustic analyser, 5 – endocrine system, 6 – peripheral muscle-nerves system, 7 – cardio-vascular system, 8 – respiratory system, 9 – the system of metabolism, 10 – thermal regulation system.
Let us define the crucial terms connected with effective sports preparation: Sport training is a process of complex biological, psychological, and social adaptations in which an athlete is systematically loaded by a set of specific stimuli, in order to improve reactions and form; develop motor abilities and personal qualities; acquire knowledge, motor skills, tactical acting, and behaviour; and to improve sport mastery (Šimonek, 2014). Agility is the key complex motor ability in team games. Sports with the highest level of this ability include soccer (8.25 points out of 10 maximum possible points), basketball (8.13), tennis (7.75), ice hockey (7.63), badminton (7.38), squash (7.25), volleyball (7.00), and ice-skating (6.88) (http://sports.espn.go.com/espn/). This term comprises the ability to stop, rapidly change direction, and accelerate in response to an external cue. In many sports games and combat sports (Bloomfield, Polman, O´Donoghue, & McNaughton, 2007; Gabbett, Kelly, & Sheppard, 2008a; Little & Williams, 2005), top athletes should have acquired a high level of agility. Some literature uses the term quickness synonymously with agility or change-of-direction speed (Moreno, 1995; Sheppard & Young, 2006a). However, Sheppard and Young (2006a) suggested that the definition of quickness has its limitations as it does not consider deceleration or a change
6
Chapter One
of direction and implies that quickness, in and of itself, contributes to agility. The literature indicates that agility must consider not only speed but also the ability to decelerate, change direction, and reaccelerate in response to stimuli. Agility is thus a complex set of independent skills that converge for the athlete to respond to an external stimulus with rapid deceleration, change of direction, and reacceleration (Sheppard and Young, 2006a; Young, James, & Montgomery, 2002). These experts suggest that agility is affected by the athlete´s perceptual and decision-making ability and the ability to change the direction quickly. In addition to the sensory and cognitive abilities involved in reacting to a stimulus and initiating the movement response, the constraints of agility may also demand elements of anticipation and decision-making. For example, when intercepting a ball, skilled players often employ a predictive movement strategy which is initiated in advance based on the anticipated trajectory of the ball (Gillet, Leroy, Thouvarecq, Megrot, & Stein, 2010). Recently, Young, James, and Montgomery (2002) outlined a comprehensive definition of agility in the context of running sports, such as soccer. The researchers addressed the multi-faceted influences involved in agility performance. In particular, the authors outlined that there are two main components of agility – change of direction speed, and perceptual and decision-making factors. Within these two main components, sub-components exist (Fig. 1-1). Sheppard and Young (2006a) redefined agility as a rapid whole-body movement, with the change of velocity or direction, in response to a stimulus. This definition implies three information-processing stages: stimulus perception, response selection, and movement execution. The first two components of agility performance can be estimated by measuring simple and multi-choice reaction time. Reaction time is an inevitable component of open-loop skills required in many sports games (for example - basketball, handball, soccer). Decision time strongly influences total reactive agility time. According to Young and Willey (2010), decision time has the highest correlation with the total time. Thus, decision time can be considered as the most influential, of the test components, for explaining the variability in total time. Most skilled elite players differ from less skilled players in this component (Farrow, Young, & Bruce, 2005). They show fast reactions thanks to the fast decision-making processes, which are based on anticipation and experience. The third component of agility performance is movement execution. This depends on the ability of a player to initiate the movement as quickly as possible by taking the first stride. We can assume a stronger correlation between maximal step velocity and agility time, over a shorter than a longer distance (Little & Williams, 2005).
Definition of terms
7
Fig. 1-1 Universal components of agility (Sheppard & Young, 2006a)
Zemková (2016b), in her recent study, found out that acceleration and deceleration phases impact more on agility performance over short than longer distances, but the reactive agility test only provides information on agility time, which includes both reaction time and movement time. The traveling distance thus should be adjusted to the real sport-specific situations. Šimonek and Kazár (2016) presented a unique structure of general motor ability, which, in line with Verkhoshansky (1996) and MČkota (2000), considers agility to be a hybrid and complex motor ability compound of different abilities, such as explosiveness, frequency of movements, action speed, reaction speed, dynamic balance, rhythm, spatial orientation, amongst others (Fig. 1-2).
8
Chapter One
Fig. 1-2 Position of agility within the hierarchical structure of motor abilities (Šimonek & Kazár, 2016).
Though agility requires the use of cognitive components, it is also composed of other qualities – namely ‘physical’ and ‘technical’. Together, these qualities (cognitive, physical, and technical) form agility (Fig. 1-3). This combination of independent qualities, plus the unplanned nature of agility, means agility has been referred to as a complex and open motor skill in its own right (Jeffreys, 2006).
Definition of terms
9
Fig. 1-3 The components of agility (Young, Dawson, & Henry, 2015)
Coordination can be defined as “cooperation of the central nervous system and skeletal muscles within some aimed movement procedure (Holmann & Hettinger, 1990). Quality of coordination depends principally on the processes of movement control and the connected nerve-muscular processes, as well as on the level of analysers. Movement coordination is defined as “temporal, spatial and power control of individual movements or complex motor expressions, which are executed with regard to tasks and goals handed over through senses (Mechling, 1983). Coordination abilities are defined by Hirtz (1985) as “complex, relatively independent prerequisites of performance regulation of movements, which are created and developed in motor activities based on dominant, inherited but influenceable neuro-physiological functional mechanisms and therefore, they can be improved by means of a methodical training. Kirchem (1992) states that the terms “skill and “agility, that were previously used, were not able to explain the complexity of coordination abilities and to describe their structure. Reaction speed is the ability to react quickly by an adequate (standard or non-standard) movement activity on a certain stimulus (acoustic, optic,
10
Chapter One
tactile, kinaesthetic) or actual change of situation (Hirtz, 1985). Impulse can be also a moving object (ball, puck, teammate, an opponent). We differentiate between a simple and complex motor reaction. In sports games and combat sports, complex motor reactions (a reaction with an option) are the most common requirement; this requires fast selection from various options of such motor reaction, depending on which is most appropriate and effective for the given situation, and which is most likely to lead to success. Gamble (2013) states that perception-action coupling and decision-making are critical elements in terms of developing the ability to express reaction speed and agility capabilities under match conditions. Space–orientation ability is an ability to learn fast and adequately change the position and movements of the body in space and in relation to the external environment (court lines, teammates, opponent, ball, goal) (Hirtz, 1985). This enables the player to have an accurate orientation in any game situation, and coordinate movements in compliance with the real movement task. This depends, to a great degree, on the quality of vestibular apparatus. Rhythmic ability is an ability to grasp and simulate temporal and dynamic segmentation of the course of movement (Hirtz, 1985). We speak mostly about accommodation of the movement to the given (external) rhythm or finding an optimum and effective internal rhythm, allowing for higher effectiveness of the motor activity. Related to this is also the ability to adapt to the motor rhythm of other athletes, the team, and the change in the rhythm of playing, and to enforce one´s own rhythm to the opponent. Balance ability is the ability of an individual to maintain or restore the balance of the body in situations where a fast or unexpected change in body position has occurred (Hirtz, 1985). Balance involves a host of sensorimotor capacities, comprising input from visual, vestibular and somatosensory systems (Bressel, Yonker, Kras, & Heath, 2007). It plays an important role, particularly in ice hockey. It depends on the size of the weight-bearing surface, the body's centre of gravity, and the state of the vestibular system and the CNS. Information from the vestibular systems is extremely important in terms of maintaining balance. We differentiate static and dynamic balance - from the point of view of sports games, a high level of dynamic balance is required. Elaboration of the program of sports preparation is a difficult and complex task, requiring thorough knowledge of the reality of this kind of sport, as well as honest preparation for its realization. Since motor activity
Definition of terms
11
in sports games and combat sports is of a non-standard character, it is very difficult to create a serious universal program of sports preparation. Longterm sports preparation, through the application of an optimum focus and content of preparation, should ensure a gradual development of all those factors of the structure of sports performance thus conditioning sports performance to a crucial degree (Šimonek, 2014). From this point of view, it is important to know all the factors that form the structure of sports performance in the given sport. This requires applying optimum focus and content of preparation, the procedure of reaching this target status, as well as information on desired changes of individual factors of performance in compliance with age-related developmental changes. An important prerequisite of the effectiveness of sports preparation is an adequate application of training loads, optimum in volume, intensity, coordination complexity, and psychological demands, as well as a gradual and sufficiently progressive increase in individual stages of the long-term sports preparation. A key part of sports training in sports games is the fulfilment of various tasks, these are called components of sports preparation. Only acquisition of all the components, which create a complex mosaic of the process that is called sports training, can lead to an optimum growth of performance level (Šimonek, 2014). Individual components are represented in sports training in various proportiona depending on the period in which the athlete is situated. Besides other components of sports preparation - such as technical, tactical, theoretical, psychological preparation, and medical observation conditioning plays the most important role as it is the decisive determinant for all sports activities. The importance of conditioning is manifested in various age categories in different proportions. Development of motor capacities is carried out based on adaptation changes in the particular physiological, functional systems and corresponding psychological processes, and is conducted in cooperation with the acquisition of motor skills and habits (Šimonek, 2014). A rational program of agility development essentially emerges from the knowledge of the factors structuring performance in sports games and combat sports. Sports performance forms a "complex system of factors, which are arranged in a certain way, there exist mutual relations among them and in their entirety, they are manifested in the level of performance” (Dovalil, Choutka, & Svoboda, 2002).
12
Chapter One
Zatsciorsky (1979 in Šimonek, 2014) underlined the great importance of the knowledge of factors in the structure of sports performance, while pointing to the need to solve the following problems: 1. Which factors underlie the performance in the given sport? 2. What are the mutual relations among these factors? 3. What is the degree of importance of individual factors for the performance in the given sport? 4. The knowledge of the structure of sports performance in time - in various age categories some factors are more important than others, but after time their importance can change. This means that the current level of preparedness and the state of the organism should be evaluated from the point of view of prospective requirements of the model structure of the sports performance. Due to the structural requirements of sports performance and the functional structure of the human organism, there exist two integrated levels on which sports performance evolves. According to Felix (1997), sports performance is the result of a cooperation of many factors (Fig. 1-4).
Fig. 1-4 Multifactorial character of sports performance (Felix, 1997).
Definition of terms
13
In the structure of sports performance, we can differentiate the following spheres:
Genetic predispositions (physiological, psychological and somatic); Prerequisites of personality; Motor predispositions (motor, conditioning potential); Coordination prerequisites and mastery of the sport technique; Sport tactics; Social environment and conditions.
Top sport poses higher and higher claims on sports preparation, accruing from the necessity of permanent improvement of its contents. Sports games in modern understanding are a highly dynamic and changeable phenomena, and knowledge of their essence requires a deeper analysis and insight into their finest structure (Šimonek, 2014). Choutka and Dovalil (1991) present the following classification of sport performance – noting that the majority of sport games belong in the group of collective performances - which require a high level of control by the central nervous system as well as analysers For example, anticipation, decision under time pressure, sensory-motor coordination in time and space, claims for a static and dynamic balance, claims for variability and application of the trained potential in the constantly changing conditions of a sports match (Table 1-2). Performance in sports games places a complex set of requirements on players. It consists of a large number of operations and acts, focused on the realization of a certain aim, logically structured according to time and controlled by voluntary processes. The majority of these game activities are realized in non-standard conditions, thus impeding the possibility of their acquisition and improvement in the training process. Similar requirements are also typical in combat sports. Regarding agility in combat sports, common features of the sports are the kinetic activity of acyclic character and variable intensity of loading (Singh, Sathe, & Sandhu, 2017). Whether you do judo, karate, boxing, wrestling or MMA, your fitness level directly determines how well you perform the skills of the given sport (Scott & Saylor, 2010). The specific skills of individual combat sports require high levels of basic motor abilities, such as flexibility, rhythm, dynamic balance, reaction speed (disjunctive reaction), dynamic strength, speed of action, as well as motor coordination. All of which depend on high levels of the functions of analysers and central nervous system regulation activity. Agility forms a special component of
Chapter One
14
sports performance, which underlies the technical mastery of an athlete. Footwork is very important from the point of view of the result of the match. However, only reactive agility, not the “closed-loop" skill, is the core element of the structure of sports performance in combat ports. This is why we focus on combat sports when characterizing agility performance in sport. Tab. 1-2 Characteristics of a motor activity in sports games and combat sports (Choutka & Dovalil, 1991 in Šimonek, 2014).
Team sports
Type of Characteristics of a motor activity sport Sport event Solved task perforPhysiologi- PsychologiMotoric mance cal cal Volleyball Overcoming The number Medium High level of Soccer active of power concentration, opponent movements uptake, controlled Handball by or skills is regulation aggression, Basketball individual, large, of motor creative Ice-hockey group or complex activity on tactical Fieldcollective movement quality thinking, the hockey means. structures, under the decision Water polo. creative long-term under time coordination, load, pressure, and large cardiovasc anticipation, variability. ular and team respiratory thinking and systems are acting, high loaded medium to maximum.
motivation, high claims for coordination in time and space, earnest attention and fast reaction, claims for dynamic balance.
Definition of terms Boxing
Overcoming active Wrestling opponent by Judo conditional, technical Fencing and tactical Karate means. Aikido Combat sport
Kendo Sumo Kung-Fu Taekwon-do Kickbox
The number of movements or skills is large, complex movement structures, oftencreative coordination, large variability.
15 Medium power uptake, regulation of motor activity on quality under the long-term load, cardiovascular and respiratory systems are loaded medium to maximum, aerobic to the anaerobic type of loading with medium to high intensity.
High level of concentration, controlled aggression, creative tactical thinking, the decision under time pressure, anticipation, high motivation, high claims for coordination in time and space, earnest attention and fast disjunctive reaction, claims for dynamic balance.
Although it appears that perceptual decision-making factors can affect competition agility, there is a paucity of scientific data on this relationship. Šimonek (2013a; 2013b) published some results of the research into agility development in soccer, but this is just one of the multiple studies that should be carried out in order to specify the components of the complex ability in sport – agility. Since, based on the literature analysis, contradictory findings have been reported around the extent of the relationship between the different speed components and agility we came to the conclusion that it is inevitable to go deeper in the research of various manifestations of speed and agility, especially in sports games. For example, Horiþka, Hianik, and Šimonek (2014), based on the theoretical analysis, carried out measurements of basic factors of speed abilities and agility in 14-17-year-old basketball, volleyball and soccer players (n=56). The results showed that, among the 3 sports games no statistical differences in the level of agility tested by Fitro agility test (basketball - p=0.189; volleyball - p=0.949; soccer - p=0.832) were observed. Spearman rank correlation test showed that no significant
16
Chapter One
correlation (p=0.786; p > 0.05) was found between the results of Fitro agility test and Illinois test measuring speed abilities. The results suggest that agility is not simply one of the speed abilities. Besides simple reaction of speed, acceleration, and deceleration, accompanied by the change in direction of movement comprises, also, of perceptual components determined by the complex reaction to unexpected, changeable stimuli occurring during a sports game.
Factors influencing agility performance Several factors have been reported as possibly influencing agility performance (Young et al., 2002; Sheppard & Young, 2006a). Cognitive and perceptual factors are considered the discriminating factor in agility performance; however, the majority of research has focused on the physical aspect (Paul, Gabbett, & Nassis, 2016). Cognitive and perceptual factors distinguish between high- and lowlevel agility performances (Scanlan, Humphries, Tucker, & Dalbo, 2014). The technique is considered a component of COD ability (Sheppard & Young 2006a); yet, the amount of empirical evidence is comparatively infrequent. Wheeler and Sayers (2010) have examined the differences in agility running technique between unplanned and pre-planned performance conditions in national and international rugby union players and concluded that the presence of a decision-making element limited lateral movement speed when sidestepping and, as such, the foot-placement patterns differed from pre-planned conditions. Less lateral movement speed during conditions was associated with greater lateral foot displacement at the COD step than in pre-planned conditions. Physical factors constitute the greatest proportion of total time to complete an agility test. They include strength and power qualities, and functional movement.
Anthropometric factors and change of direction speed Only a limited amount of research has been carried out that has attempted to find the relationship between anthropometric parameters and performance in the change of direction speed. Theoretically, factors such as body fat and length of body segments can affect agility performance. If we compare two athletes with the same body weight, the one with the higher 20
Definition of terms
17
percentage of body fat will have less active muscle mass contributing to speed requirements on agility performance. Additionally, they will have the larger amount of redundant fat tissue, which will require the bigger production of force per unit of muscle mass in order to perform the given change of speed or direction (Enoka, 2002). Gabbett (2002) tested 159 adolescent (13 – 16 ages) and junior (17 – 18 ages) rugby players. He measured body composition and then put participants through a set of tests including physical ability tests and Illinois agility test. Similar tests were carried out by Meir, Newton, Curtis, Fardell, and Butler (2001) and Reilly, Williams, Nevill, and Franks (2000). Results of all three studies revealed that, in sports like rugby or soccer, players reaching better results in the change of direction speed tests have a lower body fat percentage. No causal (only associative) relationship has been proven. Only one study found a poor correlation (r=0.21) of the above-mentioned variables (Webb & Lander, 1983). We can assume that the low percentage of body fat is a prerequisite of higher performance in the change of direction speed, however, the relationship between these two variables remains unclear. Further anthropometric factors which can potentially influence agility performance include body height, length of extremities, and gravity centre position. Some research studies claim (Cronin, McNair, & Marshall, 2003) that the length of legs affects certain types of movements, such as lunges (typical changes of direction for tennis players). An individual with lower body height should theoretically be able to employ horizontal power faster than the taller athlete is as they do not need so much time to lower the body centre during the preparation for the lateral change of direction). Another factor may be laterality. Testers should take into consideration the effect of laterality of players. In that regard, Rouissi, Chtara, Owen, Chaalali, Chaouachi, Gabbett, et al. (2016) reported that young elite soccer players had a better change of direction performance with the dominant leg vs. the nondominant leg. Therefore, the difference in time performance between various groups may be influenced by players´ laterality.
Technique Running technique plays an important role in agility performance (Sayers, 2000). Mainly, the tilt and body gravity centre seem to be crucial, both for optimizing acceleration and deceleration and increasing stability. Stability ensured by gravity centre lowering allows for changing the direction of movement more effectively. Sayers (2000) suggests that, in comparison with sprinting with a higher gravity centre (as it can be seen in the technique of middle- and long-distance runners), the change of direction requires
18
Chapter One
deceleration and postural adjustment (lowering the gravity centre and shortening the stride length). Athletes in sports which demand frequent changes of direction should find a compromise between higher gravity centre running, longer strides, and lowered gravity centre running, or shorter strides. Sayers (2000) also mentions great differences in the specificity of training in athletics and sports games or combat sports, which require fast changes of direction. For example, sprinters, having started from their blocks, keep their eyes set down, while athletes in sports games, where agility plays key role, must visually inspect the whole field and constantly react on the variable changing course of the game. In this case, technique can be also seen during the acceleration phase of the sprint, when a marked incline and gravity centre lowering occurs as in motor tasks containing changes of direction of movement. The further distinction between track and fielders and sport games players is that track and fielders can plan their sprinting action, while soccer players respond by sprinting only during the game which, in many cases, is not pre-planned. For example, during a soccer game, approximately 1,300 changes of movements are undertaken in offthe-ball conditions; players perform over 700 turns and swerves at different angles throughout the game (Stolen, Chamari, Castagna, & Wisloff, 2005). Besier, Lloyd, Ackland, and Cochrane (2001) investigated what role technique plays at onset speed in tasks containing changes of direction of movement. They examined the loading of the knee joint in planned and unplanned changes of direction. When athletes had to change direction in reaction to a light stimulus, the action of force in the knee joint increased. It is assumed from the study that unplanned changes of direction of movement brought up by a reaction to stimulus increases the action of force in the knee joint more than the changes of direction planned.
CHAPTER TWO REACTIVE VERSUS PRE-PLANNED AGILITY
Sports games performance is characterized by high-speed actions, while athletes should take quick decisions and solve the sport-specific tasks occurring during the match. Based on this assumption we can conclude that complex reaction speed, acceleration, maximum speed, the speed of whole-body change of direction, and agility represent the basic components of sports performance mainly in sports games and combat sports (fencing, boxing, aikido, karate, etc.). Agility is one of the main determinants of performance in soccer, basketball, ice hockey and handball (Little & Williams, 2005). However, definitions of this quality differ among sports researchers. The basic movement patterns of team sports require the player to perform sudden changes in body direction in combination with rapid movements of limbs and the ability of the player to use these manoeuvres successfully will depend on other factors such as visual processing, reaction time, perception and anticipation. Speed and agility in team sports represent complex psychomotor skills (Verkhoshansky, 1996). They involve moving the body as rapidly as possible, but agility has the added dimension of changing direction. Speed is classically defined, as the shortest time required for an object to move along a fixed distance, which is the same as velocity, but without specifying the direction (Harman & Garhammer, 2008). In practical terms, it refers to the ability to move the body as quickly as possible over a set distance. However, in reality, the issue is slightly more complex because speed is not constant over the entire distance and can, therefore, be divided into several phases: acceleration, maintenance of maximum speed and deceleration (Plisk, 2008). Agility is most commonly defined as the ability to change direction rapidly (Altug, Altug & Altug, 1987). This can take many forms, from simple footwork actions, to moving the entire body in the opposite direction while running at a high speed. Thus, agility has a speed component, which is an important component of this trait, amongst others. The basic definition of agility is too simplistic because it is now thought to be much more
complex, involving not only speed, but also balance coordination and the ability to react to a change of the environment (Plisk, 2008). MČkota (2000) considers agility to be physical capability, which in its essence belongs among “mixed” physical capabilities. It is determined by the quality of regulation (CNS) and analysers, as well as the type of muscle fibre.
20
Chapter Two
Therefore, agility should be superior to speed, quickness and coordination abilities. In the past, this term used to be understood as the ability to change direction, or to start and stop the movement quickly (Gambetta, 1996; Parsons & Jones, 1998). Similar morphological and biochemical factors of maximal speed, acceleration speed, and agility, lead some authors to the assumption that the given abilities are related and interdependent. Despite that, Buttifant, Graham, and Cross (1999) did not succeed in finding a significant correlation between straightforward sprinting and agility in two different groups of Australian football players. Nor was there any correlation between agility, acceleration speed, and maximal speed found in the group of 106 Australian football players who were assessed for 10m sprint (acceleration), flying 20m sprint (maximum speed) and zig-zag agility performance (Little & Williams, 2005). Although performances in the three tests were all significantly correlated (p < 0.0005), coefficients of determination (r(2)) between the tests were just 39, 12, and 21% for acceleration and maximum speed, acceleration and agility, and maximum speed and agility, respectively. Based on the low coefficients of determination, it was concluded that acceleration, maximum speed, and agility are probably specific qualities and relatively unrelated to one another. The findings suggest that specific testing and training procedures for each speed component should be utilized when working with elite players. Young, Benton, Duthie, and Pryor (2001) proved, through their research, that if agility and speed abilities are connected with the performance of sport specific skill, inter-correlation decreases even more. This can also be caused by the fact that training methods for their development are specific to each of the types of speed abilities, thus minimum transfer of qualities between them occurs (Young, McDowel, & Scarlett, 2001). Traditionally agility has been defined as “the ability to change direction rapidly” (Bloomfield, Ackland, & Elliot, 1994; Mathews, 1973) “the ability to change direction rapidly and accurately” (Barrow & McGee, 1971), “whole body change of direction” as well as “rapid movement and direction change of limbs” (Baechle, 1994; Draper & Lancaster, 1985). However, this does not take into account that most changes of direction in sport are in response to a sportspecific stimulus. Sheppard and Young (2006a) stated that a definition of agility should not only recognize the physical and technical skills involved, but the
cognitive processes as well. However, agility is typically tested and trained by using set drills that require an athlete to navigate around a pre-planned course as quickly as possible (Draper & Lancaster, 1985; Semenick 1990), with these pre-planned drills being closed-skill drills with no response to a stimulus. Commonly employed agility tests, in fact, assess only the change
Reactive versus pre-planned agility
21
of direction performance. This stipulates that there must be some element of reaction and/or decision-making in any true assessment of agility. Some change of direction tests incorporate simple reaction cues, such as response to lights or similar. However, this does not represent a valid measure of the game-related information-processing and decision-making factors that contribute to team sports agility performance (Sheppard & Young, 2006a). Agility is now regarded to be more complex and as incorporating neuropsychological factors such as anticipation, intuition, sensory processing, and decision making with such physiological factors as response time, acceleration and maximum speed, change of direction speed and mobility. Moreover, these factors interact with each other to varying degrees, dependent upon the sport-specific context. It is now commonly accepted that visual cue processing, anticipation, and reaction time are all important to agility performance in team sports (Veale, Pearce, & Carlson, 2010). Lockie, Jeffriess, McGann, Callaghan, and Schultz (2014) try to verify, in their research, the assumption that planned and reactive agility are different athletic skills, also in basketball players. Results of this study reemphasized that planned and reactive agility are separate physical qualities. Reactive agility discriminated between semi-professional and amateur basketball players, while planned agility did not. Horiþka, Hianik, and Šimonek (2014) carried out measurements with the aim of finding out the correlation between agility and the ability to simply react, accelerate, decelerate and change the direction of movement. They presumed that there was no significant correlation between the results of two tests – Fitro Agility Check (FAC) and Illinois Test executed by young male soccer, basketball and volleyball players (n=56; Mage = 15.78 years, age range: 14 – 17 years) randomly recruited from the local basketball (V10), volleyball (V 13) and soccer (V33) teams in Nitra. Authors of the research expected that the performance of players of different sports games would not be significantly different in the test Fitro Agility Check. Fig.2-1 shows the different variability of values in the observed groups. The highest variance was registered in the values of volleyball players, followed by soccer, while the lowest variance was observed in basketball players. Authors presume that this fact could be induced by scattered extreme values mostly in volleyball and basketball players.
Chapter Two
22
Fig. 2-1 Variance of performances in Fitro agility check (FAC).
In the case of the comparison between the values in FAC in the groups of basketball and volleyball (F-test for variance), the value p=0.000323 was found. Since p 0.05 so it was necessary to use t-test with equality of variance. Table 2-1 Two-sample F-test for variance (basketball – volleyball).
Median
basketball 1339.30
volleyball 1333.25
Variance
6435.52
56907.26
n
13
13
Difference
12
12
F
0.1130
P(F 0.05/. b) Zero hypothesis could be accepted so basketball and soccer players did not statistically differ in the level of performance in the test Fitro agility check /p = 0.0501> 0.05/. In this case, the p-value is on the border of the opposite interpretation of the relationship between the performances of both groups. c) Zero hypothesis could be accepted so soccer and volleyball players statistically did not differ in the level of overall performance in the test Fitro agility check /p = 0.3173 > 0.05/.
24
Chapter Two
T-test for two samples showed that players of basketball, volleyball, and soccer did not have a statistically significantly different level of reactive agility (FAC). With the exception of the performances of basketball and soccer players, this statement is unequivocal. The character of the movement, mainly its reaction and speed-strength determinants are likely to be similar in all observed samples since no significant differences were found between players (Tables 2-4, 2-5, and 2-6). Table 2-4 Two-sample t-test with unequal variances (basketball – volleyball). basketball Median
15.65
Variance
6435.522
t Stat
0.086782
P(T 0.05
Spearman Rank Correlation2tailed Test rs DF p -0.037
57
0,786
p > 0.05
The second aim of the research was to look at whether any relationship between the observed parameters (FAC vs. Illinois) exists if we were to add together all performances in all sports games. The aim was to eliminate potential differences in the character of performances according to sports specialization, and to assess the given relationship between values of players observed as one team (VSG = Vb + Vv + Vf). In this case, using the procedure of Spearman (Hendl, 2006) (Table 2-10), non-linear Fig. 2-2) and zero characters of the relationship between the observed values were proven.
Fig. 2-2 Correlation Illinois vs Fitro agility.
After eliminating differences between the selected games, performances in the two tests did not correlate. Reaction ability to the visual stimulus and the following realization of movement of the body to the particular
Reactive versus pre-planned agility
27
destination in the shortest possible time occurs very frequently in the given kinds of sports games. However, it is a very complex structure of perception, coordination, speed-strength abilities and, in comparison with the movement without solving any movement task, the perception itself has a deceleration character with regard to the speed of movement. In agility training for sports games it is, therefore, necessary to implement adequate stimuli without the task solution known in advance. Results of the study by Horiþka et al. (2014) suggest that agility is not simply one of the speed abilities. Besides simple reaction speed, acceleration and deceleration accompanied by the change of direction of movement, it comprises also of perceptual components determined by the complex reaction to unexpected, changeable stimuli occurring during a sports game. Training to develop speed and agility would, therefore, appear to demand a high degree of neuromuscular specificity. Perceptual components that underpin speed and agility must also be accounted for when developing these qualities, these include anticipation and decisionmaking. These constraints will be specific to the sport and playing position. Exhibition of both speed and agility in team sports occurs in response to game situations. It follows that perception-action coupling and decisionmaking are critical elements in terms of developing the ability to express speed and agility capabilities under match conditions. When developing speed and agility, coaches should apply one of the two possible approaches: one approach involves the relatively closed skill practice of movement mechanics, often using specialized commercially available equipment such as ladders, mini-hurdles, and resistance belts; others advocate a more open skill approach in which agility movements are conducted in a training environment that is less structured and thereby closer to match conditions (Bloomfield et al, 2007). There is an increasing body of data that support the efficacy of training interventions to develop both change of direction abilities (Brughelli, Cronin, Levin, & Chaouachi, 2008) and the perceptual and decision-making aspects of agility (Serpell, Young, & Ford, 2011). Authors unite in the statement that in order to develop agility, planned change of direction movements executed in a static practice environment must be progressed to open skill conditions, requiring a response to a stimulus. Based on the found research facts we can state that:
28
Chapter Two
x No significant differences were observed in the level of reaction speed to various stimuli between the players of various sports games (basketball, volleyball, and soccer); x The influence of sports specialization is less marked than the one of training stimuli; x No significant relationship was found between the level of performance in agility with complex reaction (FAC test) and the speed components tested by Illinois test; x The above-mentioned facts show the dominance of perception in the character of movement action in game situations in sports games and its importance in the development of agility in the sports preparation; x The means used for agility development in the selected groups of athletes caused similar adaptation reactions even without the obvious compliance; x Based on a review of the current paradigm of agility classifications, training, and testing, there is a need within the sporting community to recognize what agility involves, how it is trained and what characteristics are being assessed using existing tests of agility. As noted above, many tests do not involve decision-making or reactive component and could be better described as the change of direction speed tests. Horiþka and Šimonek attempted to find factors of the structure of agility and their roles in team sports as well as combat sports. At the same time, they intended to find out the role of psychic processes, particularly cognitive processes in the realization of specific sports skills. Cognitive processes in cognitive psychology are defined as the ones processing information, which take their course from bottom to top or vice-versa (Sternberg, 2002). The task of cognitive abilities in movement control is evident, since the quality, amount and frequency of stimuli, which are formed by the game story, require particular adaptation processes also in this area. As stated by Barry (2007), executive functions allow us to predict the consequences of our behavior and adaptation to the changing situations. ýešková, Kuþerová, and Kašpárek (2006) classify physical abilities among seven areas of cognition (memory, attention, executive functions, physical abilities, language skills, visual perception, and social cognition). Executive functions are focusing on information processing and formulation of strategies upon solving problems.
Reactive versus pre-planned agility
29
One of the methods of finding out the level of executive functions is the Stroop test – the test of psychic loading. The test requires the interaction of cognitive abilities of an individual, and so predicates not only on his/her possibilities of coping with the perception loading, but also on the level of his/her cognitive abilities in general. Svoboda (1990) classifies the Stroop test among objective personality tests and presents its correlations with intelligence tests, tests of attention, memory and with methods oriented to detection of personality traits. Based on the above-mentioned knowledge, Horiþka and Šimonek carried out an experiment with elite female basketball players. 11 players of a top league basketball team from Nitra, Slovakia formed the experimental sample. The mean age of players was 21.58 years, the mean height – 177.22 cm, and mean body weight – 66.2kg. Players were asked to go through the Stroop test (Dyer, 1973), which consists of three elementary tables representing three basic subtests of the test. The observed factor was the time of realization of the task in the three subtests: S- Table - words (S-score interpreting personal tempo) F-Table - colours (F-score interpreting factor of perception) SF-Table – words, colours (SF-score interpreting factor of perception loading). For the explanation of the share of individual factors in the structure of performance in agility, testing of the following indicators was recognised: 1. Factor - reactive agility: FITRO agility check (Hamar - Zemková, 2001); 2. Factor - reactive agility: Y – Test (Fiorilli, Luliano, Mitrotasios, Pistone, Aguino, Calcagno, & DiCagno, 2017); 3. Factor – reaction and acceleration speed: modified 10 m sprint (3+7 m); 4. Factor - running agility: Illinois test (Getchell, 1979), 5. Factor – spatial orientation: Shuttle run (Hirz, 1985); 6. Factor – running speed: 30 m sprint; 7. Factor – perception load: Stroop test (Dyer, 1973).
30
Chapter Two
The 10 m sprint test was modified in order to find out the proportion of reaction speed (3 m sprint) and acceleration speed (10 m sprint), which we expected to play an important role in agility structure. Correlations between individual variables were determined by means of correlation analysis (Hendl, 2006). From the point of view of the factorial structure of reactive agility (measured by Fitro Agility Check test) no significant correlations with the observed physical abilities were found. Performances in the reactive agility test seem to be indifferent to other indicators with an exception of cognitive abilities (rs = 0.887 (p-value 0.006). Despite this, in the second test of reactive agility (Y-test) with simple (bipolar) selection of movement, statistically significant relationships with other observed indicators was detected, mainly with reaction speed (rs = 0.914), spatial orientation (rs = 0.887) and running agility (rs = 0.886). In these cases, we assess the relationship to be highly significant (Table 2-11).
.367 .332 .396 .357
.250 .516 .237 .498 .400 .286 .326 .433 .233 .546
Pearson Correlation rs Sig. (2-tailed) p
Pearson Correlation rs Sig. (2-tailed) p
Pearson Correlation rs Sig. (2-tailed) p
Pearson Correlation rs Sig. (2-tailed) p Pearson Correlation rs Sig. (2-tailed) p
3+7m sprint (10 m)
Shuttle run
Flying 30 m sprint
Y test
Illinois test
.402 .284
Pearson Correlation rs Sig. (2-tailed) p
3+7m sprint (3 m)
.383 .308
.251 .516
.317 .406
.367 .330
1 -
.887 .006
Pearson Correlation rs Sig. (2-tailed) p
Stroop test
.083 .831
1 -
Stroop test
Pearson Correlation rs Sig. (2-tailed) p
Fitro agility check
Fitro agility check
Tests
Table 2-11 Correlations between test results
.914 .003
.871 .018
.846 .009
.912 .003
.929 .002
1 -
.367 .330
. 402 .284
.566 .125
.763 .016
.917 .001
.237 .498
1 -
.929 .002
.317 .406
.250 .516
3+7m 3+7m sprint sprint (3m) (10m)
Reactive versus pre-planned agility
.887 .003
.496 .554
.617 .252
1 -
.237 .498
.912 .003
.251 .516
.237 .498
Shuttle run
.767 .027
.835 .004
1 -
.617 .252
.917 .001
.846 .009
.383 .308
.400 .286
Flying 30 m sprint
.876 .005
1 -
.835 .004
.496 .554
.763 .016
.871 .018
.367 .332
.326 .433
1 -
.876 .005
.767 .027
.887 .003
.566 .125
.914 .003
.396 .357
.233 .546
Illinois Y test test
31
32
Chapter Two
No relationship was found between the results in the Stroop test and Y-test (rs = 0.396), indicating that both tests measure different abilities. The above-mentioned findings point to the fact that a simpler selection of motor reaction (smaller number of stimuli) is more broadly conditioned by the observed motor abilities than the more complex, multidimensional motor reaction, in which this conditionality was not observed. Aside from motor abilities, it is also determined by the level of cognitive abilities. In a complex evaluation of the share of factors participating in the level of simple reaction agility (Y – test), in comparison with the more complex (disjunctive) reaction agility with a bigger number of stimuli (FAC), we can state the decreasing number in the share of conditioning motor abilities with the increase in the number of stimuli. Cognition plays the primary role in this type of agility, cognitive processes but also the ability to adequately react to the stimulus at perception loading. From the point of view of motor abilities, we gave evidence of the significant and limiting relationship between reaction speed (3 m sprint) and simple reaction agility (Y - test), but also with other observed motor abilities. On cue, the highly significant relationship between reaction speed and acceleration speed (rs = 0.929) was found.
CHAPTER THREE TESTING OF AGILITY
The aim of this chapter is to summarize the literature on agility testing and provide coaches with practical recommendations for the use and application of these tests. Testing in sport is typically undertaken with one of three broad aims in mind: -
To evaluate the abilities or the current state of preparedness of the athlete in the context of the demands of their sport; To monitor progression and evaluate the effectiveness of training designed; To set individualized training intensity parameters based on the evaluated capacities of each athlete (Impellizzeri, Rampinini, & Marcora, 2005).
Similarly, as in sports training, testing should be tailored to the specific needs and constraints of each sport. The principles of training specificity have implications when assessing athletic performance (Abernethy, Wilson, & Logan, 1995). Fundamentally, in order to be relevant, any physiological test selected must match the specific capabilities identified as contributing to performance in sport. For team sports, this requires consideration of not only the sport, but also the playing position. A wide variety of tests to measure the level of ability to change the direction of movement (Brughelli et al., 2008; Little & Williams, 2005; Sheppard & Young, 2006) are used in different kinds of sports. The test protocols differ in complexity and duration, which results in varying statistical significance of the correlation of assay scores for the individual criteria in changing the direction of movement (Sporis, Jukic, Milovanovic, & Vucetic, 2010). When choosing the assay protocol, it is necessary to take into account the following two main aspects: the extent to which the protocol is similar to the requirements of the competitive match, and the existence of standards for the given test, which would provide the possibility
34
Chapter Three
to compare the performance achieved by the players. Usually, agility tests offer good reliability, although this may be compromised in younger participants responding to various scenarios. A human and/or video stimulus seems the most appropriate method to discriminate between the standard of playing ability. The most commonly used tests of running agility (which do not include the reactive ability when making a choice) include: L-test, T-test, 22 m slalom run, and Illinois test (Sheppard & Young, 2006a).
Test No. 1: Illinois Agility Test (Getchell, 1979) Purpose: It is used to test running agility. Equipment required: A flat non-slip surface, marking cones, stopwatch, measuring tape, timing gates (optional). Procedure: The length of the course is 10 meters and the width (distance between the start and finish points) is 5 meters. Four cones are used to mark the start, finish, and the two turning points. Another four cones are placed down the centre, an equal distance apart. Each cone in the centre is spaced 3.3 meters apart. Subjects should lie on their front (their head to the start line and hands by their shoulders). On the 'Go' command, the stopwatch is started, and the athlete gets up as quickly as possible and runs around the course in the direction indicated, without knocking the cones over, to the finish line, at which point the timing is stopped.
Testing of agility
35
Fig. 3-1 Illinois agility test
Results: An excellent score is under 15.2 seconds for a male, and less than 17 seconds for a female. The table below gives rating scores from poor to excellent for males and females, the target group is national level 16 to 19year-olds (source: Davis B. et al; Physical Education and the Study of Sport: Text with CD-ROM, 5th edition. 2004). Table 3-1 Running agility norms Rating Excellent Above Average Average Below Average Poor
Males (s) < 15.2 15.2 - 16.1 16.2 - 18.1 18.2 - 19.3 > 19.3
Females (s) < 17.0 17.0 - 17.9 18.0 - 21.7 21.8 - 23.0 > 23.0
Test No. 2: 10-meter Agility Shuttle (4x10m) This test measures agility and speed while running between two lines, 10m apart, to pick up small blocks. Purpose: This is a test of speed, body control and the ability to change direction (agility).
36
Chapter Three
Equipment required: Two wooden blocks for each runner (each block should measure 10 x 5 x 5 cm), marker cones or marking tape, measuring tape, stopwatch, and a flat non-slip surface with two lines, 10 meters apart. Procedure: Mark two lines, 10 meters apart. using marking adhesive tape or cones. The two blocks are placed on the line opposite the starting line. On the signal "Ready", the participant places their front foot behind the starting line. On the signal "Go!", the participant sprints to the opposite line, picks up a block of wood, runs back and places it on or beyond the starting line. Then turning without a rest, they run back to retrieve the second block and carry it back across the finishing line. Two trials are performed. Scoring: Record the time to complete the test in seconds to the nearest one decimal place. The score is the better of the two times recorded. A trial is void if a block is dropped or thrown. Target population: This is a good test for children as a measure of general athleticism and sports in which agility is important, including tennis, soccer, and basketball. Comments: The blocks should be placed at the line, not thrown across them. Also, make sure the participants run through the finish line to maximize their score. In addition to running speed, turning technique and coordination are also significant factors in this test. Source: http://www.topendsports.com/testing/tests/agility-10m-shuttle.htm
Test No. 3: Zig Zag Test Equipment required: Marker cones, stopwatch, and a non-slip surface. Procedure: This test requires the athlete to run a course in the shortest possible time. A standard zigzag course is one with four cones placed on the corners of a rectangle 10 by 16 feet, with one more cone placed in the centre. If the cones are labelled 1 to 4 around the rectangle going along the longer side first, and the centre cone is C, the test begins at 1, then to C, 2, 3, C, 4, then back to 1. Source: http://www.topendsports.com/testing/tests/zigzag.htm
Testing of agility
37
Fig. 3-2 Zig Zag Test.
Test No. 4: 505 Agility Test The 505 Agility Test is a test of 180-degree turning ability. The test may also be adapted for sport-specific testing by having the subject dribble a soccer ball or hockey ball through the course or bounce a basketball.
Fig. 3-3 505 Agility Test.
Equipment required: Start/stop timing gates or stopwatch, non-slip running surface, and cone markers. Procedure: Markers are set up 5 and 15 metres from a line marked on the ground. The athlete runs from the 15-metre marker, towards the line (run in distance to build up speed) and through the 5m markers, turns on the line and runs back through the 5m markers. The time is started when the athlete first runs through the 5m marker and stopped when they return through these markers (that is, the time taken to cover the 5m up and back distance - 10m total). The best of the two trails is recorded. The turning ability on each leg
38
Chapter Three
should be tested. The subject should be encouraged to not overstep the line by too much. Source: http://www.topendsports.com/testing/tests/505.htm
Test No. 5: Agility T-Test Purpose: The T-Test is a test of agility for athletes, and includes forward, lateral, and backward running. Equipment required: Tape measure, marking cones, stopwatch, and timing gates (optional).
Fig. 3-4 Agility T-Test
Procedure: Set out four cones as illustrated in the diagram above (5 yards = 4.57 m, 10 yards = 9.14 m). The subject starts at Cone A. On the command of the timer, the subject sprints to Cone B and touches the base of the cone with their right hand. They then turn left and shuffle sideways to Cone C, and touch its base, this time with their left hand. Then, shuffling sideways, move to the right to Cone D and touch the base with the right hand. They then shuffle back to Cone B, touching with the left hand, and then running back to Cone A. The stopwatch is stopped as they pass Cone A.
Testing of agility
39
Scoring: The trial will not be counted if the subject crosses one foot in front of the other while shuffling, fails to touch the base of the cones, or fails to face forward throughout the test. Take the best time of three successful trials to the nearest 0.1 seconds. The table below shows some scores for adult team sports athletes. Comments: Ensure that the subjects face forward when shuffling and do not cross the feet over one another. For safety, a spotter should be positioned a few meters behind Cone A to catch players if they fall while running back through the finish. Table 3-2 Standards for Agility T-test:
Excellent Good Average Poor Excellent
Males (s) < 9.5 9.5 to 10.5 10.5 to 11.5 > 11.5 < 9.5
Females (s) < 10.5 10.5 to 11.5 11.5 to 12.5 > 12.5 < 10.5
Source: http://www.topendsports.com/testing/tests/t-test.htm
Test No. 6: Quick feet test This is a simple test of foot speed and agility and gives an indication of the amount of fast twitch fibres in the athlete's leg muscles. Equipment required: A flat, non-slip surface, stopwatch, 21 two-foot (60cm) long sticks or a 20-rung rope ladder. A soccer field with each yard marked can also be used. Procedure: Place the sticks 18 inches apart (or a similar size 20-rung stride rope ladder) on a flat surface, across a distance of 10 yards (9.14m). The subject starts at one end, and when ready, starts running along the ladder, placing a foot in each space without touching the sticks/rungs. The timing starts when their foot first touches the ground between the first and second stick and ends when they step beyond the last stick. Rest for two minutes and repeat the test. Results: Record the best result of two trials. Times of less than 2.8 seconds (males) and 3.4 seconds (females) are considered excellent for college athletes.
Chapter Three
40
Source: http://www.topendsports.com/testing/tests/quick-feet.htm
Test No. 7: Side-step test This side step test is a simple fitness test of agility, conducted over 1 minute. Equipment: A flat, non-slip floor with line markings (can use masking tape), tape measure, and a stopwatch. Procedure: Subject stands at a centre line, then jumps 30cm to the side (e.g. right) and touches a line with the closest foot, jumps back to the centre, then jumps 30 cm to the other side, then back to the centre. This is one complete cycle. The subject tries to complete as many cycles as possible in one minute. Scoring: One complete cycle is recorded as 1, and half a cycle as 0.5. The score is expressed as the number of repetitions in one minute. Some normative values are presented below. Table 3-3 Standards for Side-step test: Males (s) Poor Fair Average Good Excellent
< 37 38 – 41 42 – 45 46 – 49 50+
Females (s) < 33 34 – 37 38 – 41 42 – 45 46+
Source: http://www.topendsports.com/testing/tests/sidestep.htm
Test No. 8 Agility Cone Test The Agility Cone Drill is a lateral movement test that measures the agility of the athlete, specifically body control and change of direction.
Testing of agility
41
Fig. 3-5 Agility Cone Test.
Purpose: This is a test of speed, explosion, body control and the ability to change direction (agility). Equipment required: A stopwatch or timing gates, measuring tape or chalk, 5 marker cones, and a flat non-slip surface. Procedure: The cones are laid out, as per the diagram, with four marker cones placed in a diamond shape, and one in the middle. The outer cones are each placed 3 meters from the centre. The player crouches behind and with their left hand on the middle cone, facing forwards (towards Cone 5). The player then turns and runs to the right and touches Cone 2 with their hand. They then turn back and run to the centre cone, out to the next one (3), back to the centre, out to the next one (4), back to the centre, and then finally turn and finish by running through the finish line at Cone 5. The player is required to touch the cone with their hand at each turn. Timing starts when the hand comes off the centre cone and stops when the chest passes through the line of the final cone. Rest for three minutes, then repeat the drill, moving in the opposite direction (counter-clockwise, cones in order 1-4-3-2-5). Source: http://www.topendsports.com/testing/tests/agility-cone-drill.htm
Test No. 9: L-drill test Scoring: The time to complete the test in seconds is recorded. The score is the best time of two trials.
Chapter Three
42
Target population: This test is suitable for athletes involved in one of the many team sports where agility is important, such basketball, hockey, rugby, or soccer. Source: http://www.topendsports.com/testing/tests/3-cone-drill.htm
Fig. 3-6 L-Drill Test.
Test No. 10: Balsom agility test Balsom Agility Test is a test of agility designed for the soccer player, in which the subjects are required to make several changes of directions and two turns. The test was developed by Paul Balsom in 1994.
Fig. 3-7 Balsom Agility Test.
Purpose: This is a test of speed, body control and the ability to change direction (agility).
Testing of agility
43
Equipment required: A stopwatch or light gates, measuring tape, marker cones, and a flat surface. Procedure: Set up the cones as illustrated in the diagram to mark the start, finish, and the three turning points. The length of the course is 15m (the distance to cones at B, C, and D have not been defined). The subject starts at A and runs to the cones at B, before turning and returning to A. Subject then runs through cones at C, turns back at D, and returns through C. The subject turns to the right and runs through cones at B, and through the finish. Two trials are allowed and the fastest time recorded. Scoring: The best (fastest) total time is recorded. Target population: The test was designed for soccer players, but the test would also be suitable for many team sports where agility is important. Source: Balsom, P. (1994). Evaluation of Physical Performance. In Ekblom, B. (ed.) Soccer (soccer). Oxford, UK: Blackwell Scientific, p.
Test No. 11: Shuttle Cross Pick-Up This test is a hockey specific, hand-eye coordination and agility test, performed off-ice, though set up to replicate some of the on-ice abilities required of ice hockey players. It is one of the tests of the SPARQ rating system for hockey, the SPARQ protocol details are listed here. Purpose: This is a test of hand-eye coordination and the ability to change direction (agility). Equipment required: A stopwatch or timing gates, measuring tape, 2 marker cones, 3 octagonal (SPARQ Agility Web) rings, 2 SPARQ Quick React Balls are preferred (tennis balls are also okay), and a flat non-slip surface. Procedure: The cones and rings are laid out in the cross pattern shown in the diagram. Figure 3.8 shows the distances to the centre of each ring or cone. A ball is placed in the ring at position C and E. The athlete stands at the starting line (A) in a two-point athletic stance, holding the set position for 2 seconds. If using hand-timing, the athlete begins in a 2-point (standing) athletic stance on the start line. Hand-timing begins on the athlete's first movement. If timing with an electronic start/finish beam,
44
Chapter Three
the athlete should start from a line marked 50 cm (20 inches) behind the timing beam to minimize premature breaking of the beam due to arm swing. On the command “Go”, the athlete sprints forward to centre cone (B) and left around it to ring (C) to pick up the ball. They then sprint back across the centre cone (B) to the opposite ring (D) to place the ball in the ring. They then sprint back to the centre, around the cone to ring (E) for second ball pick-up and finally sprint back through the start/finish line (A) with the 2nd ball in their hand. Scoring: Two attempts are allowed. Record the fastest time, in seconds, to complete the circuit as described by the protocol. A dropped ball disqualifies the trial. The times are also disqualified and not recorded if the athlete uses a rolling start, does not hold for 2 seconds, starts in a 3-point or 4-point stance, or starts with foot across the start line. The trial is also nullified if the athlete fails to perform the left and right hand turns as instructed by protocol, and if they fail to place the ball in the ring or fail to carry the 2nd ball through the finish line.
Fig. 3-8 Shuttle Cross Pick-Up.
Target population: Ice hockey Comments: The total running distance is 50 meters, and the test requires both the right and left turns with 3 stop/starts.
Testing of agility
45
In sport, agility movements are typically reactive with few circumstances occurring where a change of direction is pre-planned with no decision making (Safaric & Bird, 2011). In light of this, reactive agility tests (RAT) were designed as a means of assessing the athlete´s physical, technical and cognitive qualities. According to the latest study by Zemková and Hamar (2018), agility time strongly correlates with the choice reaction time, regardless of sports specialization of athletes or their previous experience with agility training. This indicates that perception and decision making are the most influential components of agility performance. Therefore, sportspecific methods should be addressed in both agility testing and training. Different reactive agility protocols have been investigated with various teams in various sports. To test reactive agility, reactive agility tests should also include anticipation and decision-making components in response to the movements of a tester. Loureiro and Freitas (2016) introduced a new reactive agility test called Badcamp, designed for badminton players. The test is performed in a rectangular area of 5.6 by 4.2 m (L×W). In the centre of this area, the participant's starting position is demarcated with a rectangle measuring 0.7 by 1.4 m (L×W). This rectangle is divided into two parts (0.7 by 0.7 m square) in which participants must place each foot. Six targets, composed of 1.2 m high inflatable towers, are placed on the edges of the test area; four on each corner and two in the middle of each longer side and aligned with the centre line of the rectangular area. A luminous panel measuring 0.67 by 0.52 m is placed in front of the starting position, half a meter out of the testing rectangular area (Figure 3-9). This panel contains six LED arrows that indicate the direction the participant should run in order to reach the target. The arrows pointing upward correspond to the front targets, the ones pointing downward correspond to the rear targets, and the ones pointing to the left and right correspond to the left and right middle targets, respectively. The panel is connected to an integrated circuit that controls the arrows lighting. The arrows light randomly, without repetition, one at the time, until all six arrows have been lit. A push button connected to the circuit is placed just in front of the starting position area and is used to control the arrows’ lighting, and to start and end a digital chronometer attached to the panel and used to register the time to perform the test.
46
Chapter Three
Fig. 3-9 The Badcamp reactive agility test (Loureiro & Freitas, 2016).
The participant is instructed to stay in the starting position area, start the test by pressing the button switch when he/she is ready to do so, and run as fast as possible. The test starts when the participant presses the push button switch for the first time. This first touch lights the first arrow and the participant runs towards the corresponding target, touches it, immediately runs back to the starting position area, and presses the switch again. The second pressing of the switch lights the second arrow and shows the new target that should be touched. The test ends when the participant finishes touching all six targets and presses the switch for the seventh and last time. Altogether, each participant performs 12 movements, six going toward the target and six going back to the starting position, resulting in a running distance of around 36.4m. Loureiro and Freitas (2016) recommend this test, which was created to assess badminton players by simulating specific movements and conditions of uncertainty. It is a specific agility test for badminton players, recognising that the differences between badminton players and other groups of athletes occur only when these athletes have their agility evaluated by the Badcamp agility test. In an effort to make the tests as specific as possible, Farrow, Young, and Bruce (2005) developed a RAT for netball, which was based on a video projection simulating real movements of an opponent in a game. The video image was projected so that the image of the player was life size, with the screen being positioned 6 m from where the participant had to react (Fig.3-
Testing of agility
47
10). Researchers found that it was able to differentiate between high and lesser skilled netball players whereas a pre-planned agility test was not. Movement times were, on average, slower in the reactive condition due to the perceptual component. As a result, test measures recorded under reactive conditions also appeared to be superior in differentiating elite athletes from sub-elite competitors in these sports. A similar design was used by Serpell, Ford, and Young (2010), where the life-size stimulus was projected approximately 9m from the start position and 6 m from where the participant had to react. Henry, Dawson, Lay, and Young (2011 and 2013) tested the reactive agility of Australian rules football players with the same distinctive Y pattern run, with the video screen being positioned approximately 16 m from the start line, and 9 m in front of where the participants had to react. Despite the different traveling distances, all the above-mentioned test protocols were similar in that the participants had to run forward and cut either to the left or the right at a 45o angle when presented with the stimulus. As with any electronic stimulus device, the participant is exposed to a twodimensional and generic stimulus. The light image does not resemble a stimulus that is present in sport. Moreover, the generic pattern of targets does not stimulate sporting movements. Cognitive research suggests that the anticipatory cues of high-performance athletes are linked directly to specific cues displayed by opponents within their sport (Sheppard & Young, 2006). When considering the human information-processing model (Fig. 3-11), a stimulus produces specific mental operations that are based on the individual´s retrieval of stored memory information before initiating a response. The accuracy and speed of this response will be dependent on the previously stored information, specific to the situation (Cox, 2002). Taking the above-mentioned assumptions into consideration, the new protocols, using pre-recorded videos of various sporting movements, have been elaborated for the sake of reactive agility performance testing.
Chapter Three
48
Fig. 3-10 Video-Based Reactive Agility Test (Henry, Dawson, Lay, & Young, 2011).
Stimulus
Mental operations
Response
Fig. 3-11 The information-processing model (Cox, 2002).
However, the first to introduce the new RAT test were Sheppard and Young (2006a) who evaluated the reliability and validity of a new test of agility (Fig. 3-12), the reactive agility test (RAT), which included anticipation and decision-making components in response to the movements of a tester. They designed their test to measure the reactive agility of Australian rules football players. The design involved timing gates being
Testing of agility
49
placed 5 m to the left, 5 m to the right, and 2 m forward of the start line where the participant stood. The tester was positioned opposite the participants and would initiate a movement, with the participant reacting to the movement by sprinting in the direction they anticipated the tester was going. They proved that traditional closed-skill sprint and sprint with direction change tests may not adequately distinguish between players of different levels of competition in Australian rules football. It is not just the physical qualities that differentiate higher and lesser-skilled players, but more importantly their cognitive qualities or decision-making abilities. The authors recommended including evasions drills and pursuant games for reactive agility rather than performing directional changes around stationary objects.
Fig. 3-12 Live Tester Reactive Agility Test (Sheppard & Young, 2006a).
Similar results of reactive agility test (RAT) have been found in different sports such as rugby league (Gabbett & Benton, 2009; Gabbett, Kelly, & Sheppard, 2008a; Gabbett, Sheppard, Pritchard-Peschek, Leveritt, & Aldred, 2008b; Serpell et al., 2010), Australian rules football (Henry, Dawson, B., Lay, B., & Young, 2011, 2012, 2013; Sheppard et al, 2006b) and hockey (Morland, Bottoms, Sinclair, & Bourne, 2013). Collectively, these studies found that highly-skilled players performed significantly better
50
Chapter Three
at the RAT, while there were no significant differences between groups in the pre-planned agility tests. This test can help coaches distinguish between players, while traditional closed-skill sprint and sprint-with-directionchange-tests did not. Inglis and Bird (2016), in their review, summarized information on reactive agility tests and provided coaches with recommendations on the best way to test and develop agility in athletes. They concentrated on agility tests that used a sport-specific stimulus and tested higher and lesser-skilled athletes. Ten articles were identified that met the criteria for inclusion, with half of the studies incorporating a video-based stimulus and the other half using a tester as the stimulus. It was found that reactive agility tests were a valid and reliable method of testing agility compared to traditional preplanned agility tests. Reactive agility tests can also be used as a training drill to improve an athlete´s perception and response times by using a sport specific stimulus, where pre-planned agility drills may not. Young and Willey (2010) introduced a new test of reactive agility. One week before assessment, the participants were provided with information about the test and performed eight practice trials. The reactive agility test layout and procedures were the same as previously used with rugby league players (Gabbett et al., 2008a, Gabbett & Benton, 2009), incorporating eight trials presented in a random order. A single tester, who was experienced in agility movements, provided the change of direction stimulus to all players. Four-foot patterns, described previously, were performed by the tester. The total time for the agility test was recorded with a dual beam infra-red timing light system (Speedlight Timing System, Swift Performance Equipment). The time commenced when the tester moved forward and departed from a light gate and finished when the participant ran through a stop gate to the left or right after they had changed direction. This total time comprised of the tester's movements, the time required by the participant to decide which direction the tester was going to move, and finally, the time necessary for the participant to change direction and sprint to the finish gate. These components of the total time were determined by counting frames from video recordings. A high-speed digital video camera (Redlake PCI 2000S), operating at 125 Hz, was positioned behind and to the side of the tester so that the field of view could clearly identify the placement of the feet of both the tester and participant for all trials. The following times were obtained:
Testing of agility
51
1. Total time (described above from the timing light system). 2. Tester time was the time from the first forward movement of the tester, when the body left the beam, to the instant the foot was planted for the final side-step. 3. Decision time was the time from the moment the tester planted their foot for the side-step to change direction to the moment the participant planted their foot to change direction. Although the participant must produce some movement after, the decision has been made prior to his/her foot plant, the instant of foot plant was required to develop an operational definition of decision time. This meant that a negative value was possible when the participant planted his/her foot to change direction before the tester planted his/her foot to produce the stimulus. 4. Response movement time was the time from the moment the participant planted their foot to change direction, to the moment they ran through the finish gate. Based on the definitions of these components, the total time was equal to the sum of the tester time, decision time and response movement time. The means of the eight trials for all measures were retained for analysis. Pearson correlations were calculated to determine the relationships among the test variables. The coefficient of variation (CV) was calculated to determine the variability associated with the tester times. Hamar (1997) introduced a new device for measuring reactive agility performance called Fitro Agility Check (FAC), created by the firm Fitronic, which consisted of 4 square mats (35 x 35 cm) 3 m apart, placed on the floor and connected with the computer (Fig. 3-13). The player stands in the middle of the testing area, their task is to quickly react to the visual stimulus (red circle on white background), appearing alternately in one of the corners of the display, by stepping on the correct mat (front-right, front-left, rearright, rear-left). The test protocol included 16 to 60 (4 to each direction) randomly generated stimuli appearing in the time interval of 2000 to 6000 ms. Reaction time is registered using the software by Zemková and Hamar (2009). The value of the arithmetic mean of 16 (60) recorded times (ms) is included in the record protocol.
52
Chapter Three
Fig. 3-13 Fitro agility check
A very new approach was offered by Serpell et al. (2011). This approach of testing reactive agility performance was based on a stimulus shown on the screen. Players had to react to the skills performed by the players on the screen (video clips). There were twelve different actions taken by the players, which represented a variety of stimuli. The experimental protocol included two groups: one experimental and one reference group. Differences between the two types of experimental stimuli, represented by 15-minute-long sessions of reactive drills using the video clips (3-times a week for 3 weeks), were statistically significant. However, there is no evidence that the improvement of reactive agility performance in players who went through the reactive agility training was transferred from laboratory conditions into the real sports performance of players. This issue remains unclear. Spasic, Krolo, Zenic, Delextrat, & Sekulic (2015) recommended reactive agility tests be used for finding the level of agility in elite handball players. The aim of their study was to evaluate the reliability and validity of the new test when used to assess reactive agility of handball players. With regard to the validity issue, they hypothesized that players whose reactive-agility is more challenged during the game (for example, defensive players; see later for more details) would outperform those who are not frequently involved in reactiveagility-tasks (for example, offensive players). However, to the best of their knowledge, all investigations done on handball athletes investigated change of direction speed, and not reactive-agility performance (Cavala & Katic, 2010; Iacono, Eliakim, & Meckel, 2014; Vieira, Veiga, Carita, & Petroski, 2013). Although the change of direction speed is an important quality in handball (mainly in the offense), in defence, reactive-agility is almost exclusively challenged. Most specifically, defensive players have to quickly respond to opponents' actions, and therefore, agility performance cannot be pre-planned. Previous studies noted the importance of sport-specific tests in testing reactiveagility to replicate real-sport environments (Gabbett & Benton, 2009; Morland,
Testing of agility
53
Bottoms, Sinclair, & Bourne, 2013; Sekulic, Krolo, Spasic, Uljevic, & Peric, 2014). With this in mind, Spasic, Krolo, Zenic, Delextrat, and Sekulic (2015) have developed a novel reactive agility test that would be appropriate in defining true game reactive-agility for handball athletes. Additionally, some studies did not measure decision/response time (Morland et al., 2013; Sheppard et al., 2006b; Veale, Pearce, & Carlson, 2010) and only measured total time. As discussed by Gabbett et al. (2008b), if decision time was not recorded, some athletes may be incorrectly classified as having superior anticipatory skills when in fact they had superior movement time. Finally, as suggested by Serpell et al. (2010), there are inconsistencies in terminology with decision time in RAT. Decision time in all RAT studies was defined as the “time between stimulus presentation and the initiation of the change of direction by the participant”. However, a more accurate description would be perception and response time.
The following table (Table 3-4), adapted from Farrow, Young, and Bruce (2005), demonstrates how a performance can be broken down to further identify the athlete’s strengths and weaknesses. Table 3-4 Diagnosis and training prescription for two athletes with different results on the reactive agility test (Farrow et al., 2005).
Generally speaking, we can consider an agility test to be assessed as being valid if it is able to differentiate participants from different skill levels, while it is considered reliable if it is able to find similar results from their tests on more than one occasion.
Chapter Three
54
A wide variety of agility tests with numerous modifications in agility testing protocols do not allow for the comparison of research study findings. Therefore, Zemková and Hamar (2015) investigated a variety of test alternatives to determine the Agility Index and related methodology based on stimuli number and traveling distances. Zemková (2015) proposed the Agility Index, which is defined as the ratio of reaction time and agility time divided by the previously determined coefficient for each traveling distance. The coefficient is the fold of the minimum traveling distance of 0.4 m estimated by analysis of the data of all examined subjects (Table 3-5). Table 3-5 Formula for calculation of the Agility Index (Zemková, 2015). Traveling distance (m) 0.4 0.8 1.6 3.2
Agility Index calculation 1 - (RT /AT) 1 - [RT / (AT/1.3)] 1 - [RT / (AT/1.9)] 1 - [RT / (AT/3.1)]
Table 3-6 Example of the specification of contact mats at the creation of agility test protocol for selected groups of athletes. Specification of contact mats Number
Adjustment
Recommendations for the given sports
2
Tennis
4
Badminton, basketball, soccer and hockey players
0.4 m 0.8 m Distance
Arrangement in space
Children and youth, older people Karateists, taekwon -doists
1.6 m
Badminton, basketball, hockey players
3.2 m
Soccer players
square
Handball, hockey players
Semi-arch
Handball, hockey goalies
Testing of agility
Placement
On the ground At breast height 6.5 x 6.5 cm
Size 35 x 35 cm
55
Soccer, hockey players Basketball, handball players Karateists, tae-kwon -doists, fencers Basketball, soccer, hockey, tennis players
CHAPTER FOUR AGE-RELATED CHANGES IN AGILITY
Overall and special conditioning preparation form a unit and that is why overall preparation must also be related to the given sports branch. According to Dovalil et al. (2002) “…correctly focused conditioning can result in an optimum state of physical and psychic preparedness of an athlete”. The importance of conditioning is manifested in specific proportions in various ages, in various sports branches and on different performance levels. Youth training requires a specific and quite different approach to design and implement physical preparation. As famously stated by Tudor Bompa, young people cannot merely be considered "mini-adults" (Bompa, 2000). The physiological makeup of children and adolescents is markedly different from that of mature adults (Naughton, Farpour-Lambert, Carlson, Bradley, & van Praagh, 2000) – it follows that the parameters applied to the training design should reflect these differences. The young athlete´s neural, hormonal and cardiovascular systems develop with advances in biological age, leading to corresponding changes in neuromuscular and athletic performance (Quatman, Ford, Myer, & Hewett, 2006). Rates of development of a number of physiological and physical performance parameters measured in young team sports athletes are shown to peak at approximately the same time as they attain peak height velocity (Philippaerts, Vaeyans, & Janssens, et al., 2006). The age at which this occurs is highly individual; “typical” ages are around 11.5 years for females (Barber-Westin, Noyes, & Galloway, 2006) and for males in the range of 13.8-14.2 years (Philippaerts et al., 2006). However, this can vary considerably – levels of biological and physiological maturation can be markedly different between young athletes of the same chronological age (Bompa, 2000; Kraemer & Fleck, 2005). In early childhood there prevails overall sports preparation (development of an overall fitness), with an increasing age and growth of sports performance
Age-related changes in agility
57
the ratio between overall and special physical preparation equalizes and, in the stage of top training, special physical preparation represents only a small proportion. Full development of generic coordinative abilities provides a range of motor skills that can be adapted to deal with sport specific movement demands. There are not many studies dealing with reactive agility related to the age of athletes. In the designing of new reactive agility tests, researchers start to investigate changes in reactive agility in different sports games and combat sports depending on the age of athletes. Contrary to pre-planned agility tests, for these tests, there is no information on agility times in subjects of different ages. Such information on age-related changes in agility time may be useful also for comparison with the agility performances of school-age children. The first attempts to compare agility times in athletes of different sports games according to age were carried out by Horiþka, Šimonek, and Broćáni (2018), Zemková and Hamar (2014b), HĤlka, Tomajko, and Šajna (2008), amongst others. Šimonek and Horiþka, in their research, measured reactive agility times of a sample of soccer and handball players and observed trends in developing agility depending on age. The level of reactive agility of players was assessed using the Fitro Agility Check device from Fitronic.sk. The FiTRO Agility Check test device (Fig. 4-1) consisted of four "pressure plates" connected to the computer. The test subject (hereinafter TS) stood (dead centre) between the 4 square bases (plates) sized 35x35cm. When testing the reactive speed and abilities, we distributed the bases 3m away from each other with their closest (internal) boundaries, and each base (plate) acted as a timer. The task of the TS was to respond to the stimuli (visual stimuli displayed on the PC monitor as a red circle on a white background) and step on the respective base as quickly as possible. The stimuli P(16), 4 on each side, were generated randomly by the software in a 3000ms interval. The reaction time was automatically measured by the software (Zemková & Hamar, 2009).
58
Chapter Four
Fig. 4-1 Equipment for Fitro Agility Check.
The level of running agility was measured by the Illinois test (Getchel, 1979). The traveling length was 10 meters, 5 meters wide (distance between start and finish). The start and finish line, and the two turning points, were indicated by 4 cones. Four more cones were placed equidistantly in the middle of the track. The central cones were placed 3.33m from each other. The player ran out when triggered by the signal as indicated in the diagram. The time was measured in seconds. The measurement of time in the Illinois test was performed by precision electronic measuring devices - Witty photocells and Witty timer with an accuracy of 0.01s. Šimonek and Horiþka, in their latest research, examined both reactive and running agility of 112 young (categories U11 through U16) male soccer players of the Soccer Club FC Nitra (Slovakia). They observed trends in the performance of 6 teams as well as a relationship between the two types of agility. They came out with the assumption that both reactive and running agility are limited by distinctive factors. In the given youth period, individuals come through a dynamic physical and psychic development which is reflected in the realization of abilities and skills in the sports game. In order to fulfil the aim of the research, they selected two tests of reactive (Fitro Agility Check; Hamar & Zemková. 2001) and running agility (Illinois agility test). Results of the testing are presented in Tables 4-1 and 4-2.
Age-related changes in agility
59
Table 4-1 Basic characteristics of the measure of position - Illinois test. Illinois test Minimum [s] 25th Percentile [s] Median [s] Average [s] 75th Percentile [s] Maximum [s]
U11 16.72 17.79 18.48 18.51 19.10 21.09
U12 16.615 17.072 17.44 17.71 18.32 19.175
U13 15.91 16.33 16.5 16.77 17.13 18.43
U14 15.89 16.32 17.02 16.90 17.47 17.74
U15 15.78 16.27 16.44 16.45 16.66 17.01
U16 15.39 16.19 16.14 16.11 16.32 17.16
Table 4-2 Basic characteristics of the measure of position – Fitro agility check. Fitro agility Check
U11
U12
U13
U14
U15U16
Minimum [ms] 1538.75 1425.94 25th Percentile [ms] 1618.28 1585.28
1291.24
1377.62
1169.98 1112.64
1371.31
1388.10
1332.31 1225.17
Median [ms]
1657.46
1651.44
1415.62
1416.06
1370.86 1299.75
Average [ms] 75th Percentile
1687.39
1680.58
1442.66
1419.95
1393.17 1292.26
1702.73
1755.03
1509.62
1431.82
1396.43 1355.75
Maximum [ms] 2180.31 1984.44
1687.25
1554.62
1541.87 1475.56
When comparing the performances of players in both tests (Illinois test vs. Fitro agility check) according to their age, we can see that there is an increasing trend in the level of performance of players in each age category. An increase in the level of performance in running agility from the initial 18.51s in category U11, to the level of 16.07s in category U16 was recorded, which represents an increment of 2.44s. However, the most dynamic growth was recorded between 12th and 13th year of age (from 17.71s to 16.77s). After a slight decrease in the performance in year 14 (on average by 0.13s), an increase in the level of performance in the test starts after the age of 14 with the peak level at the age of 16 (16.07s) (Fig. 4-2).
60
Chapter Four
Fig. 4-2 Illinois test U11-U16.
When assessing the level of reactive agility in young soccer players, we can state that there is a similar trend in the changes in the level of this indicator (Fig. 4-3). The level of reactive agility remains almost identical in categories U11 and U12 (1687.39 resp. 1687.58ms), while in the U13 category, a rapid growth was recorded (1442.66ms). Starting from this age category, a slight increase in the level of performance was observed up to the category of U16, where the level is markedly the highest (1292.26ms).
Fig. 4-3 Fitro agility check U11 – U16.
Age-related changes in agility
61
A gradual increase in the level of performance through the ages 11 and 16, and dynamic growth at the age of 13 and peak values at the age of 16, are the common feature of both reactive and running agility. At the age of 13, a person obviously comes to the stabilization of motor and nervous systems, sensory organs, and perception. This stabilization is marked mainly in running agility, which is limited mostly by nervous and muscular systems. On the contrary, a gradually increasing trend through ages in reaction agility was observed. From the point of view of dispersion of values in running agility, a higher incidence of extreme values was recorded in lower age categories, while lower oscillation of values around the mean was observed in higher age categories. A certain consistency in performance in the observed groups (Fig. 4-4) can be seen in the oldest categories (U15 and U16). Whilst, in younger categories, a higher degree of variability of values (VAR in 16Illinois = 0.141 through VAR in 11 Illinois = 1.061) can be seen.
Box Plot - Illinois test
22 21
/s
20 tim e
19
18
17 16 15 U11
U12
U13
U14
U15
U16
Categories Fig. 4-4 Box Plot - Illinois test.
When assessing the degree of stability in the Fitro agility check test, the tendencies are slightly different. The smallest variance can be seen in the U14 and U11 categories, while the highest in the U12 category (Fig. 4-5). The degree of stability of performance was similar in the oldest categories (U15 and U16). This is probably caused by the degree of sports preparation
62
Chapter Four
and long-term effect of training means focusing on the development of agility.
Fig. 4-5 Box Plot - Fitro agility check.
Both types of agility showed common, reduced incidence of extreme values towards older age categories. Reactive agility is influenced by the level of cognition and perception. These are the most decisive intervening components, which play an extremely important role in the control and realization of fast movements with rapid changes of direction of movement and reaction to a fast-changing stimulus. From the point of view of the dynamism of changes in the level of agility, we have to also take into consideration sensitive periods of development of individual kinds of motor abilities, which would probably limit agility performance most markedly. Much research since the 1980s has evidenced that, first of all, speed, speed-strength, and coordination abilities reach their peak development at the age of 13. This means that adequate intervention would have the greatest impact on their improvement. Another intention of the latest research by Šimonek and Horiþka was to find out the existence of a relationship between both kinds of agility in individual age categories (U11 through U16) of soccer players. Regarding the observed facts, we can state that no relevant relationship between
Age-related changes in agility
63
reactive and running agility, in any age category (Table 4-3), was observed. Low values of correlation coefficients (r= -0.570 through 0.5029) point to indifferent determinants in running and reactive agility. We assess this finding as fundamental. These facts underline the causes of the distinctive dynamism of changes in both types of agility, but also distinctive, conditional indicators of changes in agility in individual age categories. Of course, a certain role can also be played by different degree and focus of fitness training and match preparation of individual teams, specifications of players in teams, individual peculiarities of players, or other facts. Table 4-3 Illinois vs Fitro agility check correlation. Illinois vs Fitro agility check correlation Category
U11
U12
U13
U14
U15
U16
Fitro agility check test Illinois test
-0.0481 -0.0027
0.47484
0.50294
-0.5709 -0.199
In a detailed investigation into the characteristics of the relationship between both types of agility, we can observe a noticeable distribution of correlation coefficients in all age categories. This evidenced the low causalconsequential relationship between reactive and running agility in the observed soccer teams (Fig. 5 – 10).
Fig. 4-6 Fitro agility test vs Illinois test correlations in U11.
64
Chapter Four
Fig. 4-7 Fitro agility test vs Illinois test correlations in U12.
Fig. 4-8 Fitro agility test vs Illinois test correlations in U13.
Age-related changes in agility
Fig. 4-9 Fitro agility test vs Illinois test correlations in U14.
Fig. 4-10 Fitro agility test vs Illinois test correlations in U15.
65
66
Chapter Four
Fig. 4-11 Fitro agility test vs Illinois test correlations in U16.
A similar research investigation was carried out by Šimonek and Horiþka in handball. The research sample was formed by 36 male handball players divided into three categories: younger cadets (mean age 14.54y, mean height 164cm, mean body weight 51kg), older cadets (mean age 16.80, mean body height 169cm, mean body weight 64kg), and senior team (mean age 25y, mean body height 171cm, mean body weight 64kg). Basic characteristics of the measure of the position of teams are presented in Table 4-4. Marked differences in performances were observed in the results in all tests as well as in all age categories. Average value of performance in the acceleration 10m dash was 2.09s, 1.97s, and 2.03s respectively (Table 4-4). In 30 m sprints, the following values were observed: 4.39s in the younger cadets, 4.86s in the older cadets, and 4.43s in the senior team. Senior handball players reached the best average performance in Triple jump (5.52m), Illinois test (15.83s) and in Fitro Agility Check (1568.42ms). Large variability in performances was surprisingly observed also in individual age categories, primarily in indicators, which are limited exclusively by motor abilities. On the contrary, in the Fitro Agility Check, median values in all categories were rather balanced (Fig. 4-12 to Fig. 4-16).
Age-related changes in agility
Table 4-4 Basic characteristics of the measure of position – handball.
Fig. 4-12 10 m sprint.
Fig. 4-13 30 m sprint.
67
68
Fig. 4-14 Triple jump.
Chapter Four
Fig. 4-15 Illinois test.
Fig. 4-16 Fitro Agility Check.
When assessing the relationships between the observed indicators in the category of younger cadets, a statistically significant relationship was observed in two cases: Between the results of tests 30 m sprint (running speed) and 10 m sprint (acceleration speed) with the values of R= 0.688; p= 0.013 0.05; on 5% level of statistical significance (LSS). The relationship between the results in 30 m sprint test and Illinois test with the values of R= 0.766, p= 0.003 < 0.01, on 1% level of significance (Table 4-5).
Age-related changes in agility
69
Table 4-5 Spearman test - younger cadets.
FAC
Illinois
Spear man test 30 m Triple Jump
10 m
10 m
30 m
Triple Jump
Illinois
FAC
r
1.000
.688
-.123
.305
-.074
Sig. (2-tailed)
-
.013
.702
.335
.820
N
12
12
12
12
12
r
.688
1.000
-.158
.776
.217
Sig. (2-tailed)
.013
-
.623
.003
.499
N
12
12
12
12
12
r
-.123
-.158
1.000
-.309
-.541
Sig. (2-tailed)
.702
.623
-
.328
.069
N
12
12
12
12
12
r
.305
.776
-.309
1.000
.175
Sig. (2-tailed)
.335
.003
.328
-
.587
N
12
12
12
12
12
r
-.074
.217
-.541
.175
1.000
Sig. (2-tailed)
.820
.499
.069
.587
-
N
12
12
12
12
12
When assessing the relationships between the observed indicators in the category of older cadets, statistically significant relationships were found in three cases: between the 30 m test (running speed) and the 10 m sprint (acceleration speed) with the values of R= 0.587; p= 0.045 < 0.05; on 5% LSS, between the 10 m sprint and Illinois test with the value of R= 0.762; p= 0.004 < 0.01; on 1% LSS, and between the 30 m sprint and Illinois test with the value of R= 0.832; p= 0.004 < 0.01; on 1% LSS (Table 4-6).
70
Chapter Four
Table 4-6 Spearman test – older cadets.
When evaluating the relationships between the observed indicators, the most statistically significant relations were found in the category of seniors (Table 4-7): specifically, the relationship between acceleration speed (10 m sprint) and maximum running speed (30 m sprint) with the value of R= 0.813; p= 0.001 < 0.01; 1% level of statistical significance, and the relationship between acceleration speed (10 m sprint) and Illinois test with the value of R= 0. 694; p= 0.012 < 0.05; 5% level of significance. The negative correlation was found between acceleration speed (10 m sprint) and explosive strength of legs (triple jump) with the values of R= -0.598. p= 0.040 < 0.05; 5% level of statistical significance. No relationship was found between acceleration speed (10 m sprint) and reactive agility (Fitro Agility Check). The 30 m sprint correlated with the 10 m sprint and also with the Illinois test with the value of R= 0.839; p= 0.001 < 0.01; 1% LSS. 30 m sprint correlated with Fitro agility check with the value of R= 0.846; p= 0.001 < 0.01; on 1% level of significance.
Age-related changes in agility
71
Table 4-7 Spearman test – senior players.
In correlation analysis, the difference between performances in younger and older cadets was highly significant in the case of the triple jump (0.003 < 0.01) (Table 4-8). There was also significant statistical correlation between the performances of older cadets and seniors (Table. 4-9) (0.019 0.05) in the Illinois test, and also when comparing the results in the Illinois test between younger cadets and seniors (0.009 < 0.01) (Table 4-10).
Chapter Four
72
Table 4-8 Mann-Whitney U- test: younger cadets older cadets. 10 m run Mann-Whitney U Wilcoxon W Z Asymp. Sig. (2Exact Sig. [2*(1tailed Sig.)]
30 m
43.500 45.500 121.500 123.500 -1.647 -1.530 .100 .126 .101 .128
Triple jump
Illinois
FAC
20.000 98.000 -3.005 .003 .002
71.500 66.000 149.500 144.000 -.029 -.346 .977 .729 .977 .755
Table 4-9 Mann Whitney U- test: older cadets seniors. 10 m Mann-Whitney U Wilcoxon W Z Asymp. Sig. (2Exact Sig. [2*(1tailed Sig.)]
30 m
run 61.500 47.500 139.500 125.500 -.607 -1.415 .544 .157 .551 .160
Triple jump 51.000 129.000 -1.214 .225 .242
Illinois
FAC
31.500 72.000 109.500 150.000 -2.339 .000 .019 1.000 .017 1.000
Table 4-10 Mann Whitney U- test: younger cadets seniors.
Mann-Whitney U Wilcoxon W Z Asymp. Sig. (2Exact Sig. [2*(1tailed Sig.)]
10 m 30 m run 50.000 70.500 128.000 148.500 -1.272 -.087 .203 .931 .219 .932
Triple jump 41.000 119.000 -1.793 .073 .078
Illinois
FAC
27.000 63.000 105.000 141.000 -2.598 -.520 .009 .603 .008 .630
The differences in the variance of values in the analysed samples could have been caused by the rarely occurring extreme values, and/or by the instability of performance. Given these facts, the authors concluded that the dependence between the running and reactive agility decreases with age. This fact clearly shows that the response to stimuli and subsequent
Age-related changes in agility
73
execution of the movement seem to be limited by other factors, such as agility, with a previously known structure of the movement. Similar results were arrived at by other authors (Zemková & Hamar, 2014a; HĤlka et al., 2008; Kail, 1991). Zemková and Hamar (2014a) proved that agility time in children (n=553) decreased with increasing age up to early maturity. This decrease in agility time was divided into three phases. There was a rather steep decrease in agility time from 7 to 10 years of age (27.1%), and from 10 to 14 years of age (26.5%). Afterward, there was a slow decrease during puberty, from 14 to 18 years of age (16.5%). More specifically, agility time decreased at a rate of approximately 241.4ms/year in the first period, 172.1ms/year in the second period, and 78.7ms/year in the third period. Moreover, the within-subject variability was highest among the youngest children and diminished with age. However, significantly higher intraindividual variability for females than for males in 4-choice agility time was observed in the last period from 14 to 18 years of age. Based on this research, the authors suggest adjusting the distance between mats in the test Fitro Agility Check according to the height of children/athletes. HĤlka et al (2008) came to the conclusion that reactive agility times in soccer and basketball players (n=83) aged 10 to 15 decreased linearly with the growing age. This increase in the reactive agility was statistically significant. Kail (1991), in his study, presents that children and adolescents´ reaction times increase linearly as a function of adults’ reaction times under corresponding conditions, and that the size of the increase becomes smaller with age in an exponential manner. To sum up, reactive agility is greatly limited not only by the level of speed abilities but also by the level of perception, the status of perception organs, sensory and autonomic functions, spinal and supraspinal level of the motor system, amongst other factors. Their impact on the quality of reaction agility increases with the increasing level of their development. A significant role is played by the ability to respond to the changing visual, auditory and tactile stimuli in the game (Balkó, Borysiuk, & Šimonek, 2016; Balkó, Wasik, Chytrý, Dunajová, & Škopek, 2017). Running agility with a fixed and predetermined structure of the movement is most likely determined by different conditions. The absence of the need to respond to the stimuli and make decisions points to a major
74
Chapter Four
influence of innate speed predispositions with a lower possibility for major changes resulting from sports training. The above is, however, hardly applicable in sports games since the course of the game is constantly changing at any moment and the player must react appropriately to this development. In sports training, we recommend using open-loop skills and focusing on the separate development of reaction and running agility.
CHAPTER FIVE SPORT-SPECIFIC AGILITY PERFORMANCE JAROMÍR ŠIMONEK
In line with the increasing knowledge of reactive agility, new methods of assessment of reactive agility have been verified, besides the ones traditionally used for measuring the pre-planned change of direction speed. These tests can assess both the cognitive and the physical component of agility. As Zemková and Hamar (2018) report, such testing is more sensitive in discriminating athletes of different performance levels, compared to the pre-planned change of direction speed tests. Adding reaction task to certain stimuli into agility tests would correspond to sport-specific situations more accurately. Sports which require changes of movement direction while responding to a certain stimulus (or stimuli) represented by an opponent or the object of the game (ball, shuttlecock, etc.) should be tested and enhanced under competitive conditions. It has been found (Zemková, 2016b) that the contribution of decisionmaking processes and change of direction speed varies with the specific actions athletes have to perform during competitions, so it is sport-specific. Agility time consists of running time and decision-making time, however, in combat sports, the contribution of movement time is lower than in sports games (soccer, badminton, tennis, squash, etc.). The longer the travel, the less significant the correlation between agility time and movement time is. Zemková suggests using the so-called Agility Index for an estimation of the contribution of movement time to the agility performance. It is defined as a ratio of reaction time and agility time which is divided by the previously determined coefficient for each individual distance travelled. Another component of agility time in sports games with longer traveling distance (tennis, soccer, basketball) is acceleration, which can be assessed independently by measuring the time on a short distance (3 up to 10 m) using wireless timing gates. This kind of accurate measurement would show even minor differences between individuals and between athletes of
76
Chapter Five
different levels of performance. In highly skilled athletes there will be a weak relationship between the change of direction speed and straight sprinting speed. In most sports games and combat sports, assessment of agility performance requires a specific approach, taking into consideration sportspecific conditions. For this reason, different sports use different tools to measure agility (both pre-planned and reactive). Agility time decreases with the growing age of athletes up to maturity (Zemková & Hamar, 2014a). These authors reported a 27.1% decrease between ages 7 and 10, 26.5% decrease between ages 10 and 14, while during the ages 14 and 18 the decrease was slighter (16.5%). Zemková and Hamar (2015) found that the variability among subjects showed high F values for agility time (F1.280 = 34.48, p < .001) indicating that the subjects differed significantly in their performance. As shown in Fig. 5-1, the best agility times have been found in table tennis players, badminton players, fencers, tae-kwon-do competitors and karate competitors (< 350ms), followed by ice hockey, tennis, soccer, volleyball, basketball, and hockey players (350–400ms), then aikidoists (400–450ms), and finally judoists and wrestlers (450–500ms). Accordingly, the authors divided these sports into four basic categories (Fig. 5-2) that can be used for the comparison of individual athlete data and changes in the data during training.
Sport-specific agility performance
77
Fig. 5-1 Agility time in groups of athletes of different sports specializations (Zemková & Hamar, 2015).
Fig. 5-2 Agility time (± SD) in different sports, divided into four basic categories (Zemková & Hamar, 2015).
78
Chapter Five
Numerous authors unite in considering complex reaction speed, acceleration, maximum speed, the speed of whole-body change of direction, and agility to be the basic components of sports performance mainly in sports games and combat sports. However, contradictory findings have been reported as to the extent of the relationship between the different speed and agility components. Šimonek, Horiþka, and Hianik (2017), in their study, aimed at assessing the rate of dependence among individual speed abilities, explosiveness, and reactive agility performance. The sample comprised of 117 players (soccer – 56, basketball – 17, volleyball – 20, and handball – 24) playing in youth leagues U15-U17, who were assessed for 10-m sprint (acceleration), flying 30-m sprint (maximum speed), triple-jump (special explosiveness) performance, Illinois agility test (speed of whole-body change of direction) and Fitro Agility Check (reactive agility). In the test assessing the level of simple reaction and acceleration speed (10m sprint; Fig. 5-3), we found the highest level in cadet volleyball players (mean value = 1.85 s), handball and basketball players (mean value = 1.89 s), and soccer players (mean value = 2.06). A relative steadiness in performances, like in the FAC test, was observed in both age categories of handball players. Rather surprisingly, this seems to be the reverse ranking of players in the category of pupils, where best performances recorded handballers (1.94 s) and the worst basketballers (2.285 s).
Fig. 5-3 Comparison of U15 (pupils) and U17 (cadets) performances in 10 m sprint among sports.
Sport-specific agility performance
79
In the test of maximum running speed, flying 30 m sprint (Fig. 5-4), soccer players clearly dominated (3.73s), which was probably caused by the character of play in a match, where players move within a wider area, and also by a different structure of training load. Handball players ranked second (3.88s), volleyball players third (3.95s), and basketball players ranked last (4.02s). The surprising order (2nd and 3rd places) of players could be a consequence of a worse quality of maximum running speed of cadets in basketball since the worst results, as expected, were recorded by volleyball players (4.25 s).
Fig. 5-4 Comparison of U15 (pupils) and U17 (cadets) performances in Flying 30 m among sports.
In the test assessing jumping explosiveness (triple jump; Fig. 5-5), the highest level of explosiveness was recorded in cadet volleyball players (10.13m), basketball players (9.97m), while the lowest was in soccer players (9.31m). In the pupils´ category, handball players dominated (9.01 m), which could be caused by the lower rate of adaptation to loading in the early sports age. In the test assessing speed of changing direction as a reaction to a standard stimulus (closed skill) – the Illinois test – the best results were observed in cadets in volleyball (15.76s), basketball (15.82s), soccer (16.35s), and handball (15.355s). In the pupils´ category, volleyball players dominated (16.23s), followed by soccer, basketball and handball players (Fig. 5-6).
80
Chapter Five
Fig. 5-5 Comparison of U15 (pupils) and U17 (cadets) performances in Triple jump among sports.
Fig. 5-6 Comparison of U15 (pupils) and U17 (cadets) performances in Illinois test among sports.
In the diagnostics of agility and perception in the test, requiring quick decision-making and choosing the adequate motor reaction (FAC, Fig. 57), we observed a smaller variability in the performances of players. Differences in players´ performances were negligible, except for the ones between the age categories. The best performances were recorded in soccer players (1284.6ms), followed by volleyball players (1294.7ms), basketball players (1344.3ms), and handball players (1420.9ms). In this test, a smaller variability between both age categories was observed, mainly in handball. This is probably because of the fact that coordination (reaction speed and perception) is not limited by the level of fitness factors (speed, strength, special endurance), but by the quality of analysers and central nervous
Sport-specific agility performance
81
system and its role in the movement control from the point of view of the age of sportspersons.
Fig. 5-7 Comparison of U15 (pupils) and U17 (cadets) performances in FAC test among sports.
When assessing the rate of the dependence of selected indicators, we can observe low values of correlation coefficients (r) in soccer players (Table 51). Negative polarity in the test triple jump (in all cases) can be interpreted as negative to the low relationship of explosiveness with other speed and coordination abilities. Table 5-1 Descriptive statistics and correlation coefficients - Soccer.
Pearson Correlation Results for:
SOCCER
Descriptive Statistics
Correlation Matrix (R)
Variable
Mean StdDev
StdE rr
N
Illinois
FAC
10m
30m
Triple jump
Illinois
16.42
0.51
0.078
56
1
0.42
0.28
0.24
-0.62
FAC
1390.4
116.4
15.46
56
0.45
1
0.11
0.34
-0.39
10m
2.08
0.35
0.048
56
0.28
0.11
1
-0.53
-0.29
30m Triple jump
3.85
0.32
0.041
56
0.24
0.34
-0.53
1
-0.28
9.05
0.77
0.095
56
-0.66
-0.39
-0.29 -0.28
1
Chapter Five
82
In the case of the basketball players, a similar character of relationship can be seen, however, with higher values of correlation coefficients (Table 5-2; r = 0.586 – 0.631), mainly in the case of the relationship between the result in Illinois test and the ones in the remaining tests. Even higher values of r can be observed in the test triple jump, mainly in the relationship with the speed of changing the direction of movement (Illinois test). Similar results were found in the case of the relationship with maximum running speed (10 m vs 30 m tests). Table 5-2 Descriptive statistics and correlation coefficients – Basketball.
Pearson Correlation Results for:
BASKETBALL
Descriptive Statistics Variable Mean StdDev StdE rr
Correlation Matrix (R) N
Illinois
FAC
10m
30m Triple jump
Illinois
16.65
1.44
0.35
17
1
0.63
0.58
0.62
-0.69
FAC
1463.5
218.9
53.0
17
0.63
1
0.35
0.01
-0.51
10m
1.99
0.15
0.03
17
0.58
0.35
1
0.32
-0.41
30m Triple jump
4.03
0.20
0.05
17
0.62
0.01
0.32
1
-0.51
9.38
0.92
0.22
17
-0.69
-0.51
-0.41
0.51
1
Similar tendencies can also be found in volleyball (Table 5-3). High values of correlation coefficients (r) - exceeding the value of 0.5 – suggest that there is a high relationship between the observed variables – speed of changing the direction (Illinois test) and other speed abilities, with an exception of triple jump test. This relationship can be seen mainly in the case of 10 m and 30 m tests (r = 0.768), and FAC and 10 m tests, where the value r reaches also high level (r = 0.708). Similar to soccer and basketball, correlation coefficients in the test triple jump are negative (r = -0.635 – 0.829).
Sport-specific agility performance
83
Table 5-3 Descriptive statistics and correlation coefficients - Volleyball. VOLLEYBALL
Pearson Correlation Results for: Descriptive Statistics
Correlation Matrix (R)
Variable
Mean StdD ev
StdE rr
N
Illinois
FAC
10m
30m Triple Jump
Illinois
15.95 0.615 0.138
20
1
0.569
0.592
0.626 -0.635
FAC
1361 121.3 27.14
20
0.569
1
0.708
0.609 -0.641
10m
2.007 0.217 0.049
20
0.592
0.708
1
0.768 -0.790
30m Triple jump
4.068
0.26
0.058
20
0.626
0.609
0.768
9.493
1.18
0.264 20
-0.635
-0.641 -0.790 -0.829
1
-0.829 1
When assessing the relationship of indicators of performance in handball we observe a relatively high relationship in maximum running speed (30 m sprint) and agility, or acceleration speed (10 m sprint) with the value r = 0.686. In the case of the remaining abilities, this dependence is just slight (Illinois vs. 10 m sprint; r = 0.597), or low with the value of r lower than 0.5. Similarly, as in the previous sports games, explosiveness showed the negative relationship with speed and agility performances. All values of correlation coefficients showed negative polarity (Table 5-4).
Chapter Five
84
Table 5-4 Descriptive statistics and correlation coefficients – Handball. Pearson Correlation Results for:
HANDBALL
Descriptive Statistics
Correlation Matrix (R)
Variable
Mean
StdDe StdE v rr
N
Illinois
FAC
10m
30m Triple Jump
Illinois
15.632
1.021
0.20
24
1
0.46
0.59
0.68
-0.58
FAC
1447.0
117.4 23.97
24
0.46
1
0.43
0.41
-0.44
10m
1.919
0.090 0.018
24
0.59
0.43
1
0.68
-0.57
30m Triple jump
3.952
0.183 0.037
24
0.68
0.41
0.68
1
-0.56
9.335
0.625 0.127
24
-0.58
-0.44
-0.57
-0.56
1
To sum up, low (0.112-0.425) correlation coefficients between the factors were found in soccer, while in the other sport games they were medium (0.329-0.623 in basketball; 0.414-0.686 in handball) to high (0.569-0.768 in volleyball). A negative relationship was observed between triple jump and all other test performances in all sports games. By comparing the levels of individual speed factors in the observed groups of cadets the authors assumed that volleyball players dominated in four out of the five indicators (Illinois, FAC, 10 m sprint and triple jump). In the category of pupils, there was not as great a dominance and the differences between the sports games were wiped away. This fact can result from the shorter period of adaptation to the training and competition load, and probably also the result of the fading away of sensitive periods for the development of speed factors in the younger age category. The dominance of volleyball players in the speed-strength component is clear only in the cadet category. Another conclusion was that there were very slight differences in the level of reactive agility among the players in the four different sports games. This was probably due to the demands of all sports games on perception and quality of reaction in open skills performances. The smallest variability in both age categories was found in handball players.
Sport-specific agility performance
85
The worst results were observed in cadet handball players, but in pupils the opposite was true. Low values of correlations in soccer were found in almost all the relationships reported. On the contrary, in basketball, a high rate of dependence was found, mainly in the Illinois test and the other tests of abilities. A negative relationship was observed in all examined sports games in the case of explosiveness and the remaining speed abilities. This relationship points to the negative impact of this ability on reaction and realization speed, as well as reactive agility. In accordance with the findings of Horiþka et al (2014), a very weak relationship between the speed of changing the direction of movement, tested by Illinois test, and agility (complex multi-choice reaction), tested by Fitro Agility Check, was observed, with the exception of volleyball players. This suggests that reactive agility does not appear to be strongly linked with straight-speed components. Speed and agility are distinct physical qualities, and speed training does not appear to enhance change of direction speed. Therefore, training for the change of direction speed and agility must involve highly specific training that recognizes the specific demands of the sport. The findings of this study suggest that specific training procedures for each speed and agility component should be utilized even in junior ages. It is interesting that specific agility tests have shown no significant difference between futsal and soccer players, despite the fact that the size of the ball used in futsal significantly differs to the one used in soccer (Milanoviþ, Sporiš, Trajkoviþ, & Fiorentini, 2011). This leads researchers to the conclusion that in both, the elite soccer and futsal, it is necessary to have very skilful players. That was indicated by an STB test of dribbling the ball with the inside of the foot as well as by an SB90° test in which all the leading skills of controlling the ball have been examined (dribbling the ball with the inner and outer side of the foot, etc.). Bloomfield et al. (2007) stated that, during a soccer match, each player performs approximately 305 turns of 0-90° to the right side and 303 turns of 0-90° to the left side. Since the test S90° includes a change of direction at an angle greater than 90° in practical terms, there was no difference between futsal and soccer players. The research results demonstrate that both futsal and soccer players have quite similar motor characteristics of agility type. The majority of futsal players, at first, go through the soccer schools and, after that, they are exposed to the futsal training. Based on our results, we can conclude that the futsal and soccer players differ in the intensity exertion during the game, but not in motor activities such as agility. Agility is a very important component of futsal and soccer and it represents a common characteristic.
86
Chapter Five
Based on that fact, it can be said that the players in these two sports are very similar in agility performance. Recently, researchers (Goral, 2015; Spasic et al., 2015; Zemková & Hamar, 2013) have tried to find out the level of reactive agility in players playing in different positions in sports games. In one of these studies, Goral (2015) measured the change of direction speed in sixty-eight adult soccer players using the T-test and Illinois Agility Test. The statistical analysis revealed that the times in Illinois agility test were significantly lower in midfielders compared to goalkeepers (p