Panoutsakopoulos Vassilios Kollias A. Iraklis KINEMATICS OF THE DELIVERY PHASE AND RELEASE PARAMETERS OF TOP FEMALE JAVELIN THROWERS KINEMATIČNI PARAMETRI FAZE META IN IZMETA PRI VRHUNSKIH METALKAH KOPJA ABSTRACT The introduction of the new-rules javelin for female athletes in 1999 was thought to be a cause of increased demand for excellence in the throwing technique and that a greater demand for biomechanical analysis was stressed. The purpose of the study was to investigate the kinematical parameters of the delivery phase and the release parameters of contemporary top female javelin throwers in order to support the above mentioned suggestion. Twenty-six throws performed during competitions by 16 right-handed top athletes (age: 28.5 years ± 4.3; body height: 1.75 m ± 0.05; body mass: 73.8 kg ± 6.3; average ± standard deviation, respectively) were recorded with a digital video camera (sampling frequency: 50fps). The delivery and release phases were examined with a 2D-DLT analysis method. The relationship of the extracted spatio-temporal, kinematical and release parameters with the official throwing distance was examined with a two-tailed Pearson correlation. The results indicated that the official distance (59.22 m ± 4.42) was strongly correlated (r = .909, p < .001) with the release velocity (22.9 m/sec ± 1.6) and negatively correlated (r = -.608, p < .05) with the knee angle of the right leg at its last touchdown (142.2° ± 13.1). The temporal parameters of the delivery phase were significantly (p < .05) correlated spatial parameters such as the delivery stride length, the inclination of the torso, the elbow angle of the throwing arm and the knee angle of the braking leg. Emphasis should be given in these parameters to ensure an improvement in female javelin performance. Key words: javelin, female, throwing technique, biomechan-ics, kinematical parameters Biomechanics Laboratory, Department of Physical Education & Sport Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece Corresponding author: Vassilios Panoutsakopoulos Biomechanics Laboratory Department of Physical Education & Sport Sciences, Aristotle University of Thessaloniki, Greece Telephone: +30 23 10 99 22 17 Fax: +30 23 10 99 22 18 E-mail: bpanouts@sch.gr IZVLEČEK Uvedba novih pravil za ženski met kopja leta 1999 naj bi povzročila vse več zahtev po odličnosti tehnike meta, obenem pa so se povečale tudi potrebe po biomehanskih analizah. Namen raziskave je bil preučiti kinematične parametre faze meta in parametre izmeta pri vrhunskih metalkah kopja v sodobnem času, da bi lahko podprli zgornji predlog. Šestindvajset metov kopja, ki jih je na tekmovanjih izvedlo 16 elitnih športnic, desničark (starost: 28,5 let ± 4,3; telesna višina: 1,75 m ± 0,05; telesna teža: 73,8 kg ± 6,3; kar pomeni povprečje ± standardni odklon), smo posneli z digitalno video kamero (frekvenca zajemanja slik: 50 fps). Fazi meta in izmeta smo preučili z analitično metodo 2D-DLT. Odnos med ekstrahiranimi prostorsko-časovnimi, kinematičnimi in izmetnimi parametri ter uradno doseženo razdaljo metov smo preučili z dvostranskim testom Pearsonove korelacije. Rezultati so pokazali močno pozitivno korelacijo (r = ,909, p < ,001) med uradno doseženo razdaljo (59,22 m ± 4,42) in hitrostjo izmeta (22,9 m/sek ± 1,6) in negativno korelacijo (r = -,608, p < ,05) s kotom desnega kolenskega sklepa v trenutku zadnjega dotika tal (142,2° ± 13,1). Časovni parametri faze meta so značilno (p < ,05) korelirali s prostorskimi parametri, kot so dolžina koraka med metom, naklon trupa, kot v komolčnem sklepu izmetne roke in kot v kolenu zavirajoče noge. Za izboljšanje ženskega meta kopja je treba poudarek nameniti tem parametrom. Ključne besede: kopje, ženske, tehnika meta, biomehanika, ki-nematični parametri INTRODUCTION After being introduced at the 1932 Olympic Games, the performance in female javelin throwing saw constant improvements in the past century (Bartonietz, 2000; Dyer, 1989; Jokl, 1984; Tsarouchas & Giavroglou, 1986). It is believed that it represents the most highly evolved throwing event in track and field since performances improved by 80% in the time span between 1932-1992 (Tipton, 1997) and the large throwing distances marked (~ 80 m) were actually achieved about 15 years earlier than predicted (Jokl, 1984). The International Amateur Athletics Federation (IAAF) altered its rules concerning the specifications of the positioning of the centre of gravity in women's javelin in 1999, a fact suggested as being to be the cause of increased demand for excellence in the throwing technique (Bartonietz, 2000). Given the release parameters of the implement, the new-rules javelin starts to decelerate earlier in the flight due to the shorter maintenance of the angle of attack and the decreased lift of the implement, which eventually results in a 10% smaller range than the old-rules javelin (White, 2011). The increased demand for biomechanical analysis of female javelin throwers with the new-rules implement in order to redefine the recommendations for optimising the release parameters has been noted (Bartonietz, 2000). Studies generally conclude that release velocity is the single most important factor regarding the official throwing distance based on the existence of a strong correlation between them (Bartlett & Best, 1988; Bartonietz, Best, & Borgström, 1996; Bartonietz, 2000; Hay, 1985; Lehmann, 2010; Maier, Wank, Bartonietz, & Blickhan, 2000; Mero, Komi, Korjus, Navarro, & Gregor, 1994; Murakami, Tanabe, Ishikawa, Isolehto, Komi, & Ito, 2006; Viitasalo, Mononen, & Norvapalo, 2003). The velocity of the javelin is generated through its acceleration path during the delivery phase (Bartlett & Best, 1988; Morriss, 1995). It is suggested that further development of the javelin's velocity should be conducted through elongation of the acceleration path by keeping a relatively extended elbow (i.e. "straight throwing arm") during the early parts of the delivery phase (Morriss, 1995). The majority of information concerning women's javelin throw has been retreated from studies investigating the old-rules implement (Bartonietz et al., 1996; Menzel, 1986; Menzel, 1998; Mero et al., 1994; Shi & Tong, 2000; Tsarouchas & Giavroglou, 1986; Viitasalo et al., 2003; Xu & Nelson, 1988). The literature indicates that a limited number of biomechanical studies have analysed various aspects of the technique of female javelin throwers since the new-rules javelin was introduced (Jung et al., 2012; Kumar, 2005; LeBlanc & Mooney, 2004; Lehmann, 2010; Leigh, Liu, & Yu, 2010). A review of the above studies suggests that quantitative differences exist between the javelin technique with the previous (i.e. before 1999) and the current specifications concerning the spatio-temporal, kinematical and release parameters of the release. In detail, it seems that the current elite javelin release technique is characterised by an increased average release velocity (over 24 m/sec versus past reported values of 21-24 m/sec) and larger average angles of attitude (over instead of less than 40 degrees). These findings lead to the conclusion that further thoroughly inspected in order to establish the magnitude of the relationship of the key release and kinematical parameters with the throwing performance among contemporary top female javelin throwers. Based on the above, the primary purpose of the present study was to quantify the spatio-temporal and kinematical parameters of the delivery phase and the release parameters of the javelin throw executed by top female athletes in competition. The secondary aim of the study was to investigate the relationship of the examined parameters with the official throwing distance in order to establish the effect of those parameters on the throwing performance. MATERIALS AND METHODS Subject sample. Sixteen (n = 16) right-handed, top-level female javelin throwers (age: 28.5 years ± 4.3; body height: 1.75 m ± 0.05; body mass: 73.8 kg ± 6.3; average ± standard deviation, respectively) were examined during several major IAAF competitions held in Greece between 2006 and 2009. With the exception of one, all participants had competed at least once in the three most recent Olympic Games. The study was conducted in accordance with the Institutional Research Committee's Guidelines for the use of human subjects. Data acquisition. All trials of the participants were recorded from the right side of the athletes during the examined competitions. The recordings were acquired with a stationary JVC GR-D720E digital video camera (Victor Co., Japan), operating with a sampling frequency of 50fps and a shutter speed of 1/4000. A single camera set-up was selected since two-dimensional methods have been found to be adequate for evaluating basic javelin release parameters (Best, Bartlett, & Sawyer, 1995; Viitasalo et al., 2003). The camera was positioned on a fixed tripod in the stands at a distance of about 32 m from the middle of the runway and about 4 m before the foul line. The recorded area was calibrated by consecutively placing a 0.02 m x 0.02 m x 2.5 m pole in several predefined spots within the filming view following the guidelines suggested by Gervais et al. (1989) and Kollias (1997). This procedure was conducted in order to produce two-dimensional Figure 1. Determination of the cut-off frequency, based on the residual analysis method proposed by Winter (1990), using the SIDES2® (©: Iraklis A. Kollias) software (m: minimum value; M: maximum value). coordinates with the use of the 2D-DLT kinematic analysis method, where the X-axis was parallel to the javelin throw runway and the Y-axis was perpendicular and vertical to the X-axis. Data analysis. From all the recorded videos only the best attempt regarding the official throwing distance (SOFF) marked for the participants in each competition was selected for further analysis. In all, 26 throws of the 16 participants were analysed. Twenty-two anatomical points of the body (tip of the toe, 5th metatarsal, heel, ankle, knee, hip, shoulder, elbow, wrist and 5th metacarpal on both sides of the body, 7th cervical vertebra and the top of the head), the grip and both ends of the javelin were manually digitised in each field. The coordinates of the body centre of mass (BCM) were calculated for every field using the segmentational data proposed by Dempster (1955). A second-order low-pass Butterworth filter with a cut-off frequency ranging from 3.5 to 6 Hz, based on the noise calculated with residual analysis (Winter, 1990), was selected for smoothing (see the example in Figure 1). The digitisation, smoothing and analysis were conducted using the DIASDIG® and SIDES2® software (©: Iraklis A. Kollias). The accuracy of the 2D reconstruction was determined by Root Mean Square error, after randomly re-digitising 5% of the captured frames. An error of 0.4 cm and 0.3 cm was found for the X- and Y-axes, respectively. The coordinates of the digitised points were used to calculate the kinematical parameters presented in this study. The extracted parameters were examined regarding three key instants of the delivery phase: the last right foot touchdown (T1), the last left foot touchdown (T2) and the release of the javelin (T3) and were defined as illustrated in Figure 2 and described below. Figure 2. Definition of the spatial parameters of the delivery phase which were examined in the present study (see text for the abbreviations displayed). • Delivery stride length (S): the horizontal distance between the toes of the right foot at T1 and the toes of the left foot at T2 • Distance from foul line (S): the horizontal distance between the toes of the left leg and the middle of the foul line at T2 • Javelin acceleration path (Sy): the two-dimensional displacement of the javelin's grip • Knee angle: the angle formed by the longitudinal axes of the thigh and the shank of the right (ÖRknee) and kft (0^ kg • Elbow angle (0elbow): the angle formed by the longitudinal axes of the arm and the forearm of the throwing side • Torso inclination (%orso): the angle between the horizontal level and the line connecting the midpoints of the hips and shoulder joints' axis • Leg inclination (^leg): the angle between the horizontal level and the line connecting the hip and ankle joints • Release velocity (V0): the resultant velocity of the javelin's centre of mass at the instant of release • Height of release (h0): the vertical distance between the ground and the javelin's mass centre at the instant of release • Release angle (90): the angle between the direction of the vector of the release velocity and the horizontal level • Angle of attitude (0 ): the angle between the horizontal level and the longitudinal axis of the javelin • Angle of attack (0ß): the angle between the javelin's longitudinal axis and the direction of the vector of the release velocity SLD, Sy and h0 were also expressed relatively to body height in order to normalise data by excluding the effect of the anthropometric factors. To fulfil the descriptive nature of the present study, the researchers also obtained the number of the strides of the running and the acyclic parts of the approach by visual observation. Statistical Analysis. Results are provided for the group of the analysed throws in the form of basic descriptive statistics (mean ± standard deviation). After testing the normality of the distribution for each parameter with the Kolmogorov-Smirnov test, a two-tailed Pearson correlation analysis was used to investigate possible correlations between the examined spatio-temporal and release parameters and the official throwing distance, along with the intercorrelation among the examined parameters. All statistical procedures were conducted using the SPSS 10.0.1 software (SPSS Inc., Chicago, IL), with the level of statistical significance set at p = 0.05 for all analyses. RESULTS The average SOFF of the analysed throws was 59.22 m ± 4.42 (range: 49.78 m - 67.78), which corresponded to 93.2% of the participants' season best. 26.1% of the attempted throws were fouls. On average, the approach consisted of 7.7 running strides, followed by the acyclic part which was accomplished with 7 strides by all the participants. The values of the parameters extracted for the instants of the last right foot touchdown (T1), last left foot touchdown (T2) and javelin release (T3) are presented in Table 1. The delivery phase (T1^T3) had a duration of 0.335 sec ± 0.033, in which the ratio between the single-support phase (T1^T2) and the double-support phase (T2^T3) was approximately 60%:40% (± 6.2). SLD was about 80% (± 7.8) of the body height and was correlated with the time interval between T1 and T2 (r = .495, p < .05), with