Kinematic differences between female national and provincial athletes in the tennis serve

Subject

The movements of the serve of the active national team and Tianjin female athletes were selected as the objects of study. A total of 10 athletes participated in the experiment, of which five of the national team were national athletes, and five of the Tianjin team were provincial athletes, all of which were right-handed (Table 1). According to the questionnaire, it is known that the athletes all use the foot-up serving technique, while the athletes use the slice serve on the first serve and second serve.

National athletes are the title of technical level of Chinese athletes, which is the highest level of sports level title in China, second only to international athletes. In addition, provincial athletes are second only to the national athletes. In addition, national athletes are the top three athletes in singles (singles, doubles, and mixed doubles) and the top two athletes in the team at the National Games. Provincial athletes are the athletes on the court of the 4th to 32nd (including ties) in singles, 4th to 16th (including ties) in doubles (including mixed doubles), and 3rd to 8th in teams at the National Games.

Prior to the experiment, all athletes were in good physical health and condition, enabling them to participate in the tests and perform powerful serves. Participants were informed of the risks involved in performing the experiment and provided informed consent before the formal test. The study design and procedures were in accordance with ethical standards and the Declaration of Helsinki and received approval from the Ethics Committee of the China Institute of Sport Science (20240111).

Experimental approach to the problem

At a hard tennis court, in Beihai City on 15 January 2024 and Tianjin City on 20 January 2024, China, from eight o’clock to twelve o’clock, two high-speed cameras (model: Sony DSC-RX10M4, frequency: 1,000 Hz, shutter speed: 1/1,000 s, main optical axis angle ≈ 90°, camera height: 1.5 m) were positioned behind the baseline and 8m to the side of the serving area of the player (Fig. 1).

The camera was calibrated by placing the entire calibration frame in the camera’s shooting frame. One of the calibration frames (Peak D20, calibration range: 1.6 × 2.2 × 3.2 m) was mounted on a tripod with a level, which was placed 1 m to the right of the midpoint of the base line for leveling, and the calibration frame covered the subject’s torso and limb range of motion.

The delineation of the serving zones employed in previous studies encompassed three distinct configurations: 1 × 1 m² (Abrams et al., 2014; Reid, Elliott & Alderson, 2007; Rosker & Majcen Rosker, 2021; Sakurai et al., 2007), 1.5 × 1.5 m² (Martin et al., 2013), and 2 × 1 m² (Reid, Whiteside & Elliott, 2011b; Reid et al., 2013). At the same time, a powerful serve from the outside corner can pull the opponent out of court, leaving the whole court open, so that the whole game is within their control. The outer corners of Zone 1, both for the first and second serves, were defined as the target area measuring 1.37 × 1.5 m² (Fig. 1).

Prior to the commencement of testing, athletes engaged in a 10-min warm-up session consisting of jogging and serving (Whiteside et al., 2013b). Upon attaining a state of preparedness, participants proceeded to the formal testing phase. Athletes were instructed to execute two serves to the designated target area in the prescribed order of the first serve, followed by a second serve. Sent into the target region and aligned with the line is deemed a valid ball. Among the criteria for a successful serve were that the ball landed within the designated area and that the player had the best feel for the serve. Subsequently, one successful example from each of the first and second serves was selected for further analysis.

Statistical analyses

The German sports video analysis software Simi Motion 9.1.1 (version number: 9.1.1; SIMI®, Munich, Germany) (Becker & Russ, 2015; Winiarski, 2003) was used for three-dimensional positioning of the captured video and export of data. The Hanavan body center of mass model was employed for the data analysis (Hanavan, 1964). Data processing and analysis were conducted using MATLAB 2016 and Excel 2019 software. IBM SPSS Statistics 27.0 (Chicago, IL, USA). Mann–Whitney U tests were conducted to determine differences between groups. Statistical significance was set at p = 0.05. In this study, the data results are expressed as mean ± standard deviation (M ± SD) according to the difference relationship.

Court and racket definitions

The origin of the two court coordinate systems is set at the midpoint of the baseline. The X-axis positive direction (left to right) is defined as pointing to the right and parallel to the baseline. The Y-axis positive direction (front to back) is defined as pointing forward and parallel to the single sideline. The Z-axis positive direction (up and down) is defined as pointing upward from the midpoint of the baseline. Furthermore, the racket’s rotational state at the moment of impact is described using Euler angles (Z-Y-X). The midpoint of the racket refers to the center point of the line connecting the left and right sides of the racket. The Z-axis direction of the racket points upward towards the racket head, the Y-axis direction points forward and is perpendicular to the racket surface, and the X-axis direction points to the right and is perpendicular to the midpoint of the racket. In three-dimensional space, the Z-axis, Y-axis, and X-axis are mutually perpendicular. The positive and negative values of the data only represent the direction and do not indicate magnitude (Fig. 2). The state of the racket is shown in Fig. 3.

Divisions of key moments of the serve

Previous researchers have divided the technical action of serving into four moments: ball release, trophy position, racket low point, and impact (Whiteside et al., 2013a, 2013b). This article refers to previous research and divides the serving action into four moments: the moment the ball leaves the hand (T1), the moment of the left knee’s minimum flexion (T2), the moment of the lowest point of the racket head (T3), and the moment of impact (T4). See Fig. 4 for details of the key moments of the serve.

Comparison of differences between first and second serves

The speed of the racket and ball

There was a correlation between the speed of the racket and the speed of the ball at T4 (Gordon & Dapena, 2006; Tanabe & Ito, 2007), which contributed positively to overall service performance (Fleisig et al., 2003; Girard, Micallef & Millet, 2007). It could also be found that the swing speed of the racket has determined the speed of the ball to a certain extent. Usually there will be more topspin on the second serve, and more “brushing” when the dribble hits (Abrams et al., 2014). This study demonstrated that the athletes featured in this article exhibit slower ball speeds in their first serves compared to the findings of Whiteside et al.’s (2013a) previous research (national athletes, provincial athletes: 145.234, 142.358 km/h vs. postpubescent: 156.24 km/h), it is possible that this discrepancy can be attributed to differences in serving styles.

Vorobiev, Ariel & Dent (1993) studied Andre Agassi’s serve and investigated the biomechanical differences between the flat (first) serve and kick (second) serve. They found that for this specific high-level player, the horizontal center of gravity velocity was greater in the flat serve, whereas the kick serve demonstrated a larger vertical center of gravity linear velocity. This investigation also showed that the kick has a more medial (over the head) contact point than the flat serve (Vorobiev, Ariel & Dent, 1993). The data on ball position and swing speed on the first and second serves showed that national athletes and provincial athletes the ball more by increasing the vertical and lateral racket speeds and decreasing the forward speeds from the first to the second serve (Table 2), throwing the ball closer to the body and imparting topspin and sidespin on the ball (Tables 3 and 6), which led to a decrease in the overall ball speed.

In terms of the ball

The stability of the tennis ball throwing action is crucial for the player’s later strokes, and the height of the ball throw and the trajectory of the ball will affect the accuracy and success of the serve (Brody, 1987; Reid, Whiteside & Elliott, 2011a). The reasonable throwing technique is to properly adjust the elbow joint of the throwing arm and control the movement of the wrist joint during the throwing process, so as to ensure that the ball is thrown vertically. Studies have shown that from the first to the second serve, the player throws the ball closer to the body, resulting in a decrease in the forward position of the ball before hitting and an increase in the vertical and lateral velocities of the racket before T4, resulting in kick and slice spin (Chow et al., 2003; Gordon & Dapena, 2006).

According to this study, both the first serves of national athletes and provincial athletes were positioned further forward than their second serves, which aligns with prior research indicating that first serves are typically thrown more forward. Furthermore, the research revealed that the first serve batting position (0.373 ± 0.153, 0.489 ± 0.147 m) for national athletes and provincial athletes was more advanced than the second serve batting position (0.296 ± 0.088, 0.446 ± 0.094 m), According to the data, it appears that the variation in batting distance between the initial and subsequent serves of the national athletes (0.077 m) is larger than that of the provincial athletes (0.043 m), which may explain the disparity in the speed of their second serves.

At the same time, through the data, it can be seen that on the X-axis, the ball flies from right to left from T1 to T4, and on the Y-axis, the ball flies from the front (away from the body) to the back (close to the body), and does not move straight up and down. This is consistent with the finding of Chow et al. (2003) that during the flight phase of the toss, the ball moved forward and to the left on all.

In terms of shoulder-hip angle

The approximate order of the main contributors to racket speed between T2 and T4: external rotation of the shoulder, wrist extension, torsional rotation of the lower torso, torsional rotation of the upper trunk relative to the lower trunk, shoulder abduction, elbow extension, ulnar deflection rotation, second torsional rotation of the upper trunk relative to the lower trunk, and wrist flexion. Among them, elbow extension and wrist flexion contribute particularly well (Gordon & Dapena, 2006). Among them, the relevant rotation corresponds to the angle of shoulder rotation, hip rotation angle and shoulder-hip angle in this study. There have been reported that the lateral rotation of the torso makes a moderate contribution to the batting speed of the racket (Elliott, Marshall & Noffal, 1995). It has also been proposed that high-level men’s tennis serves to transfer momentum to the power arm through lateral (torsion), frontal (shoulder-to-shoulder) and sagittal plane trunk rotation (Bahamonde, 2000).

Previous results showed that the contact with the ground during the tennis serve caused the front leg to have greater rotation, and then the trunk and pelvis rotation accelerated longer, so that the pelvic maximum angular velocity and the trunk maximum angular velocity were produced during the tennis serve, and the maximum angular velocity of the tennis serve would be higher (Wagner et al., 2014), so the importance of the horizontal angle between the shoulder and hip can be seen. There are also studies that show that at the end of the buffer (the moment of the left knee’s minimum flexion), the angle between the shoulders and the hips is 20–30° (Elliott, Reid & Crespo, 2003), the disparity between the shoulder-hip angle of athletes in both groups as presented in this article is noteworthy, and it is essential for athletes to be cognizant of this characteristic.

In addition, this study also found that there was a significant difference in the vertical surface of the shoulder-hip angle at T2 in the first serve, and the national athletes were smaller than those of provincial athletes, which meant that the amplitude of shoulder and hip movement of provincial athletes was greater. At T3, there was a significant difference in the horizontal plane of the shoulder-hip angle, with national athletes turning the shoulder forward towards the hip and provincial athletes turning the shoulder backward towards the hip.

In terms of shoulder and hip rotation angles

There were significant differences in the vertical rotation angles of the first and second serves at T2, which was mainly manifested in the fact that the national athletes’ hip tilts more downward, which may be the result of the need for a hip top. At T3, there was a significant difference in the rotation amplitude of the vertical plane of the shoulder, and the rotation angle of national athletes was smaller than that of provincial athletes. At the same time, it can be observed that there was a difference in the rotation angle of the vertical plane of the first and second serves, and the national athletes’ rotation angle was smaller than that of provincial athletes. The torso horizontal rotation from T2 to T4 increases the distance from the racket to the striking position and the speed of the racket at T4. In addition, these rotations pre-stretch the trunk muscles and store elastic energy (Tubez et al., 2021). The angle of rotation of the shoulder and hip horizons of national athletes from T2 to T4 was greater than that of provincial athletes, indicating that national athletes had a better grasp of this aspect when serving.

In terms of body angles

Leg drive played a crucial precursor role in generating powerful torso rotation (Whiteside et al., 2013a). The knee and hip velocities generated by the lower limb drive were synchronized with the shoulder and elbow velocities generated by the torso and upper arm movements, followed by elbow extension and finally forearm internal rotation and wrist flexion (Elliott & Marsh, 1989). Studies have shown that knee flexion helps to generate more power and speed when the leg is driven (Bonnefoy et al., 2009; Girard, Micallef & Millet, 2005; Hornestam et al., 2021). In this study, the left and right knee angles of national athletes were larger than those of provincial athletes at T2. In addition, elbow extension and wrist flexion were particularly important contributors to racket speed (Gordon & Dapena, 2006), and the change in national athletes right elbow angle from T3 to T4 was greater than that of provincial athletes in this study. Whiteside et al. (2013a) found that the ankle extension was greater when the leg was driven during the serve than the knee or hip. Plantar flexors were major contributors to the forward propulsion of gait (Elliott & Reid, 2004), and this muscle group played a crucial role in leg drive in women. When the foot was fixed in the trophy position on the ground, knee flexion prior to leg drive was essential for placing the plantar flexors (Elliott & Reid, 2004). There was a significant difference in the left ankle angle at the moment of the first and second serves, with the athlete having a larger left ankle angle, which also meant that national athletes’ legs were more fully extended, and their bodies were better stretched when batting the ball.

In terms of body center of gravity

From the moment of minimal left knee flexion to the moment of impact, the provincial athletes body center of gravity is more forward, which leads to a more forward strike position. Dossena et al. (2018) worked out that reaching higher impact point during tennis serve could allow to serve faster. Through Z, it was found that the center of gravity of the national athletes was lower than that of provincial athletes at T1, T2, and T3, but was higher than that of provincial athletes at T4, which showed that national athletes had always maintained a low center of gravity before hitting the ball, obtained upward energy through the rapid push and extension of the knee joint, and then hit the ball at a high position to obtain faster ball speed and spin.

Study limitations

This study not only found significant differences between the two groups of athletes in terms of shoulder-hip and shoulder-rotation and hip-rotation angles, but also found significant difference in body center of gravity. However, a small sample size is considered a limitation of this study, future studies should focus on larger samples of participants. Other factors that may affect the effectiveness of the serve (psychological stress, external environmental disturbances) are not added to the experiment, and it is expected that these disturbances will be added in future studies. In addition, intervention studies can be included to observe differences in athletes and explore training pathways for technique optimization.