Innovation and Technology in Sports

The rowing performance depends on how fast the boat moves during the race, contributed by the generated forces and rower kinematic. Based on that interest, the objective of the study is to investigate the kinematics during a stroke by including the generated forces and related speed. The study was carried out using a dynamic rowing simulator which simplified the rowing boat and mimicked the rowing biomechanics and the hydrodynamic condition. The experimental result found that extending the legs generated a higher force to boost the speed. Using legs, trunk and arms in their overlapping motion helps the rower instantly reach the peak effort and maintain the pressure. About 496–539 N of the handle forces were captured by the simulator, which generated a blade force peak between 222 and 234 N, with the average peak force achieved for these three strokes being 46% of the drive. The oar angular speed captured was from 1 to 1.5 rad/s. In return, accelerate the boat at two m/s for 25–27 stroke/min rowing stroke. In conclusion, the combination of the kinematic, force and speed related to the rowing during the stroke is important to consider during the study. The effect of rower kinematic can be observed directly in the forces generated and associated speeds.

This study aimed to clarify the differences on friction coefficient of footwear used in futsal when mechanically measured in two different sliding direction. Available Friction Coefficient (AFC) and Traction Force (TF) of three futsal footwear with different outsole design (S1, S2 and S3) were measured using a novel six-degree of freedom mechanical test in anteroposterior (AP) and mediolateral (ML) sliding direction. Results have shown differences of AFC value when measured in different sliding direction (AP and ML) for all three shoes. In addition, it was observed that S2 shoe was the least affected in terms of reduction of AFC value when compared between AP and ML direction. It was also observed that among the three shoes tested, S2 has produced the highest TF in both AP and ML direction as compared with other shoes. From these findings, it can be suggested that traction performance of sports footwear should be evaluated by multi-directional sliding approach, and conventional one directional footwear evaluation standard such as BE EN ISO 13287 is most likely not adequate to analyse sports footwear–sports playing surface traction performance in real world.

The purpose of this study was to determine the physical fitness profile, heart rate responses, technical and timing analysis of elite junior badminton players. Four elite junior badminton players (2 males & 2 females; Average age = 17.5 years) with a minimum of 12 h of training per week were included in this study. The players completed a series of fitness testing (cardiorespiratory fitness, speed, agility, upper strength & lower strength, and power) to obtain their physical fitness profile. Heart rate responses were recorded during the matches. Videos of both genders were analyzed using a customized badminton tagging panel with video analysis software. Results showed the fitness levels of male players were better than female players. During the matches, female players demonstrated a greater average % HR compared to male players. Match analysis demonstrated that male single favourite shots were net (31.68%) and smash (20.04%), while female favourite shots were clear (25.43%) and drop (19.58%). The winning shots for the males were mostly from smash, while the females’ winning shots were from drop. The results of the study provide an insight to the coaches to develop effective training plans for junior players based on gender. (197 words).

This study was carried out to observe the cues (leg and hand) and striding performance of the horse ridden by skilled and less-skilled riders in three types of equine gaits (walk, trot, and canter). This study used purposive sampling technique which includes 14 participants (7 less-skilled and 7 skilled riders). All riders were asked to ride the same horse between the distance consisted of a 12-m start-transition-finish point track. Two video cameras were placed 10 m from the sagittal plane of the horse's straight-line pathway. Results showed that there was a significant difference in the movement kinematics (knee angle) between skilled and less-skilled riders during the equine gaits. In addition, there were also significant correlations between the skilled rider's (elbow angle) and horse movement kinematics (head angle, stride length) with the horse speed. Therefore, it can be concluded that effective riding aids (rider's movement kinematics) will influence the better speed of the horse to achieve better performance.

This research aims to identify talented skateboarding athletes with reference to their bio-fitness indicators. A total of 45 skateboarders (23.09 ± 5.41 years) who were playing for recreational purposes were recruited for the study. Standard assessment of their bio-fitness as well as their skateboarding performances was performed. The bio-fitness investigated consisted of stork balance, star excursion balance test, vertical jump, standing broad jump, single-leg wall sits, plank and sit-up while the related-skill performances consisted of the observation on skateboarding tricks execution, namely Ollie, Nollie, Frontside 180, Pop-Shuvit and Kickflip. To achieve the objective of the study, a hierarchical agglomerative cluster analysis (HACA) was performed to cluster the athletes into groups in reference to the level of their bio-fitness markers. The clusters identified two groups of performance named High-Potential Skaters (HPS) and Low-Potential Skaters (LPS) following their skateboarding performance scores. An Artificial Neural Network (ANN) was conducted to ascertain the classified athletes into the clusters (HPS and LPS) based on the bio-fitness indicators evaluated along with the skateboarding tricks performance scores. The result demonstrated that ANN accomplished a high classification accuracy of 91.7% indicating excellent performance from the classifier in classifying the skateboarding athletes. Similarly, the area under the curve of the classifier was found to be 0.988 signifying further the validity of the model developed. Overall, these results suggest that the proposed technique was able to classify the skateboarding athletes reasonably well which will in turn possibly assist coaches to identify talents in this sport through the bio-fitness indicators examined.

Social network analysis reveals a significant set of essential information on the behaviour of the players and teams: such as passing sequences between players in the attacking third position of attacking performance. Despite their usefulness, network metrics related to expected goal values for soccer analysis are minimal. Thus, the study compared network approaches that led to Chelsea FC's goal and playing style in the English Premier League in season 21/22. Moreover, ‘expected goal values’ (xG) show the probability of the goal scored, which can be related to the goals scored passing network. The study used centrality in network analysis such as degree prestige, degree centrality, and betweenness centrality to find the significant contributor of the player's position independently during the match that led to the goal scoring and did not consider the playing style of Chelsea FC. Furthermore, the xG values of shots in every game were visualized using Tableau software. A set of adjacency metrics were computed using the highlight videos of goals for every match, and the results of network analysis found that the most received the ball from their teammates were left wing-back and right wing-back, defensive midfielders and right wing-back have the highest degree centrality and attacking midfielder and left wing-back have the highest betweenness centrality. Furthermore, the statistical data from the visualization such as cross percentage, passing percentage, shots and xG percentage can be used to enhance the team performance.

Previous research indicated that passing networks can increase the performances of players in a football team. This can be achieved with the aid of data visualization and analysis using post-match data. This paper provides a taxonomy of sports data in football visualization and summarizes the data from three aspects of data types, main tasks, visualization techniques, and visual analysis with the use of Tableau software. The objective of this paper is to identify the playing pattern for Liverpool FC during Jurgen Klopp’s era. To identify the playing pattern, this paper will display the diagram of the passing networks from the goals created in the match. Besides, networks and graph theory using Social Network Visualizer is to investigate social structures from the passes data that created goals from an open play. It describes networked systems in terms of nodes and the links between them. The playing pattern may thus be determined by examining the degree of centrality, degree of prestige, and betweenness centrality from nodes and linkages. This paper introduces a visual analysis of competitive football, using the social network from passes to construct degree centrality, and finally discusses the playing pattern for Liverpool FC. For this paper, collecting and flexibly presenting large and complex data is the main concern to increase the understanding of the analysis. In summary, it was feasible to draw the conclusion that network metrics can give sport analysts knowledge that is complimentary to traditional notational analysis by offering a novel visualisation and comprehension of team members' behaviour as well as by characterising particular play patterns.

The study aims to identify the effectiveness of five minutes Calisthenics exercise on the depression, anxiety and stress levels among teenagers. A total of 180 16 years old students in Kota Setar district secondary schools were examined using quasi experimental method. Depression, Anxiety and Stress Scale (DASS) instrument was used for pre and post tests for eight weeks. The data were analyzed using the Multivariate Analysis of Variance (MANOVA) and Multivariate Analysis of Covariance (MANCOVA) to see the relationship between Calisthenics exercises and mental health levels of depression, anxiety and stress in pre and post-tests of treatment and control groups. In overall, multivariate test results with wilks’ Lambda showed that there were significant effects on the three dependent variables [F (6, 524) = 2.20, p < 0.05) for post-test and pre-test [F (6,262) = 13.95, p < 0.05). The findings showed that students who practice five minutes Calisthenics exercise can reduce depression, anxiety and stress levels. In conclusion, the study showed that there is a significant relationship between five minutes Calisthenics exercise and depression, anxiety and stress levels. Female gender had higher depression, anxiety and stress levels in both tests.

This study is conducted to study Sports Advisors’ perspectives and level of satisfaction on Transformational leadership of Senior Assistants of Co-curriculum in primary schools in Selangor. This study uses survey causal-comparative with cluster random sampling involving 420 Sports Advisors (352 males and 68 females) in daily primary schools in Selangor state. Multifactor Leadership Questionnaire is used as research instrument to collect data on perspectives and levels of satisfaction from all the participants based on gender, location of schools and educational credentials. Statistic descriptive analysis shows very high mean values on perspectives and levels of satisfaction of Sports Advisors on Transformational leadership of Senior Assistants of Co-curriculum. Independent t-Test is used to identify differences between gender, location of schools and educational credentials in this study. Results show significant differences on perspectives and levels of satisfaction of Sports Advisors based on location of schools and educational credentials. However, there is no significant difference in gender on perspectives and levels of satisfaction of Sports Advisors on Transformational leadership of Senior Assistants of Co-curriculum. Therefore, these results clarify that Senior Assistants of Co-curriculum need to consider the difference in school locations and educational credentials of Sport Advisors to apply Transformational leadership in sport management in primary schools.

Research in soccer has shown that players’ technical, tactical, physical, and psychological abilities are required to meet the requirements of the competition. This study uses machine learning to develop a motivation model based on anthropometric, fitness, and soccer skills. Data were collected from 223 young Malaysian athletes consisting of Malaysia’s Sport School soccer athletes who play in various positions (defender, midfielder and forward) aged 13 to 17 years old who participated in this study. Athletes are required to complete the study's instrument, which consists of the anthropometric component test, Task and Ego Orientation in Sport Questionnaire (TEOSQ), technical skill component and physical fitness test. Data analysis was carried out using hierarchical agglomerative cluster analysis (HACA) and discriminant analysis (DA). Hierarchical agglomerative cluster analysis is used to divide groups according to their homogenous psychological attributes of the athletes and discriminant analysis used for determining the differences in player performance. Three groups formed and successfully discriminated three groups on 13 independent variables with 79.82% (forward stepwise) total variance resulting with Machine Learning method (Artificial Neural Network) 67 athletes predicted with potential. A group tends to have the taller player because of the highest significance in height variables than others. From the result, all groups show their characteristics with unique attributes and need to intervene to characterize their training program based on the group's performance.

The development of technology has increasingly reduced the practice of physical and motor activity use and it can lead to some health-related quality life problems such as social dysfunction and disease problems. The purpose of this research is to identify the development of motor performance index among 7 years old Malaysian kids. Datasets were collected from 1998 participants in research aged 7 years old in primary schools all around Malaysia. The participants completed multiple physical fitness tests (anthropometrics, standing broad jump, twenty-meter speed, sit and reach and hand wall toss). Data interpretation was carried out using Principal Component Analysis (PCA), Discriminant Analysis and Machine Learning. It was found that there is a small number of male kids that only have high physical fitness performance but female kids have a huge number that have a high physical fitness performance, also male kids have a dominant in some physical motor component and anthropometric and also female dominate some of the physical motor component. As a result, these models have the potential to reduce the number of kids with poor motor development. Furthermore, time and efforts can be saved because it is much easier to have concentrated parameters or those that have been extensively proven.

The analysis of football has always offered great interest and attract many people among experts, researchers, pundits, and fans. Football data have been observed and studied from various perspective aiming for several objectives whether for predicting matches results or goals as well as analyzed team or player performance. This paper presents an analysis and discussion for team ranking in Malaysia Super League (MSL) 2021 based on football rating system. The football dataset is limited to seven seasons of MSL football data between 2015 and 2021, however, the study mainly focuses on final ranking in MSL league table for season 2021 based on football rating system consist of Elo rating, pi-rating and Poisson model. Each of the football rating system presented have their own unique calculation to introduce the football team rating whether as whole team for every round of league’s matches (Elo rating), rating while home or away (Pi-rating) and rating for attack and defense (Poisson model). The football rating system mainly rely on the number of goals scored, goals conceded and match results whether at home and away to be evaluated for rating the football team strength in term of attack, defense, or team as whole. Thus, the findings show that Johor Darul Ta’zim Football Club successfully become prominent football club that dominate MSL 2021. Moreover, it is suggested to include other tier of Malaysian Football Leagues (MFL) as well as adjustment of football rating system for optimization to suit the MFL environment.

Predicting football match results or goals is unexceptionally buzzworthy. Football prediction can be classified into two clusters which are statistical and machine learning. Despite successfully introducing numerous statistical and machine learning techniques to predict football match outcomes, flaws still exist. This paper attempt to present football matches outcomes prediction models based on an Elo rating system and machine learning algorithms using limited data of football matches result for Malaysia Super League. The dataset used for the prediction is the MSL football data which consists of 7 seasons played between 2015 to 2021 that contain several basic features such as date, name of home team, name of away team, home team scored, away team scored and the matches results (Win, Draw, Loss). The football data were calculated to Elo rating that rate the strength of MSL football team before are divided into training and testing set. Machine learning (ML) algorithms such as Naïve Bayes (NB), Logistic Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) have been selected in this paper to predict football matches outcomes for MSL 2021. The accuracy and average of Rank Probability Score (RPS) are used as performance matric to evaluate the prediction models. Based on the comparative analysis conducted, all the models were able to predict the outcomes for more than 50% accuracy of the matches except RF which only obtained 49.24% accuracy. The NB is the best ML algorithm compared to SVM, LR and RF for predicting MSL football matches outcomes by achieved highest accuracy of 54.55% and lowest value of average RPS by 0.2025.

Badminton is known as one of the fastest racket sports in the world. Therefore, it is important for the player to have excellent reactive agility to stay competitive in a badminton match. The purpose of this study was to investigate the validity and reliability of the React-Run, a new reactive agility assessment system for badminton players. Six subjects were divided into two groups, three male amateur players with experience in representing badminton clubs for competitions and three recreational players, who used to play badminton, were recruited for the experiment. The measurements were performed with the use of the React-Run system, where the experiment was conducted on an actual badminton court. The result of eight averages (in milliseconds) indicates reactive agility from different sensor locations (node) (i.e., Front, Front-Right, Right, Back-Right, Back, Back-Left, Left, and Front-Left) were stored and analyzed. The mean, standard deviation, and range were calculated for each outcome variable. Pearson correlation method and an independent t-test analysis were used to evaluate the construct validity of the React-Run system meanwhile the reliability (within-subject variation) was established using Cronbach’s alpha value. Results indicated that React-Run system showed good construct validity and reliability as it was statistically able to distinguish the performance of amateur and recreational badminton players (p-value < 0.05) and had internal consistency between the node sensors’ positions (Cronbach’s alpha value > 0.9).

Woodball is a rising regular sport that garnered a lot of attention from people with different ages and genders due to its easy playstyles and adaptability to play it almost everywhere. Without any teaching and guides from professional, some of novice players may unconsciously practicing wrong swing postures which is the most important fundamental in woodball. Having a good swing posture helps players to dictate how the mallet swings around the body, promotes good balance and helps the body to turn correctly. In this study, we proposed and verified a simple regression method using a portable, miniature Inertial Measurement Unit (IMU) to classify swing phases to assist the novice players to have a correct swing postures as professional players does. The IMU was attached to the player’s left hand wrist to collect the angle, acceleration and angular velocity of the swing. The signal data were processed using second-order Butterworth filter with cutoff frequency of 10 Hz, estimated using 2nd order Least Square and the swing phases were classified automatically with labels. The proposed system yields high classification performance with average accuracy of 99% for all swing phases.

Sepak Takraw players have intensively practiced heading the ball as the game’s primary movement. Repetitive takraw ball heading can result in head injuries such as concussion, internal bleeding, and dizziness. The head injury criterion can be measured practically, yet the mechanical properties of the takraw ball are not well examined. The primary aim of this study is to investigate the mechanical properties of takraw balls at different orientations based on a quasi-static compression experiment of two takraw balls, GE511 and MT908. Each ball has been subjected to compression tests in three distinct orientations of woven layers, namely Orientations 1–3. MT908 has a greater ultimate force and stiffness than GE511. It is also discovered that Orientation 1 has the highest values for both the mean ultimate force that occurred and the mean stiffness, followed by Orientations 2 and 3. These discoveries are particularly relevant to the creation of modern takraw balls and head protectors.

Table tennis is a dual sport in which two teams compete against an opposing side. In training conditions, it needs consistently receive the ball with varying ball spins and angles to improve the athlete’s skills. Therefore, this study aims to develop an automatic table tennis robot that can launch the ball in a different way of rotation. Two 12 V DC motors were placed against each other to produce the opposite direction of ball spin. While four types of rotations which are topspin, backspin, right spin, and left spin, were considered in the evaluation. The microcontroller was used to control the system, including the motor speed and launcher’s angle itself. Hence, the combination of variables applied can be customised and increasing the difficulties of training level. In addition, the setting of robot movement can be set up via the control board or wirelessly using Android apps. The reliability study was concerned with the consistency of ball bouncing, ball rotation as well as ball launching. The performance of this robot launcher is satisfactory when the error is less than 5% from the entire repetitive testing. In the experimental session, it is shown that the capability of ball shooting distance, the feed rate of a ball launched, and the ability of the robot launcher to do various ball spins are achieved and suitable to the player. Thus, this table tennis robot launcher benefits the athlete’s self-training to improve their skills and technique.

The golf swing involves the movement of numerous body parts and can be viewed in several phases. Understanding how each phase behaves is critical to improving the game. The improvement is commonly quantified in terms of accuracy, which can be determined from the trajectory of the ball. The relationship between golf swings and ball trajectories is normally obtained in a laboratory setting, which is both limited in access and expensive. Golf swings were studied in this study using a wearable device called MetaMotionS (MMS) from MbientLab. The MMS measures the acceleration of a golfer’s wrist and waist during a swing. According to the findings, the ball trajectory was influenced by the wrist and waist acceleration. Consequently, the golfer may apply the findings to improve his game.

The study of object tracking has substantially advanced thanks to the development of deep learning visual recognition and tracking methods. However, because to the additional difficulties they provide, such as the difficulty in tracking small, swiftly moving objects like a ball or shuttlecock due to the fast camera movement and the existence of swings and spins, sports videos are still understudied. To access these massive archives of sports video data and automatically tag and analyse its properties, such as player performance and stroke and shot analysis, an effective end-to-end solution is needed. The aim of this research is to create a complete deep learning based model that can do object detection and tracking in sports movies as well as classify the played stroke. We employed the SF-YOLOv5 model, a lightweight model for the identification of swiftly moving small objects, for this. Then, we utilised the Deep-Sort algorithm and zero shot learning to follow the objects that had been detected. Finally, we classified the played shot using the CNN classifier.

Machine learning techniques are often used for sports analytics, such as player health prediction and avoidance, appraisal of prospective skill or market worth, and predicting team or player performance. This reshapes the sports performance and helps in coaching the teams and individuals. This research focuses on football analytics, which can help football managers and coaches for reshaping the performance of players to target the goal with higher accuracy and precision. The match results depend on the successful number of goals; any minor mistake may lead to failure. Other statistics, like shots on target and game possessions, have been gaining popularity in recent years. Several attributes are utilized to train an anticipated goal model formed by monitoring football data to evaluate the chance of a shot being a goal. Using historical data and advanced analytics, a credible prediction of a goal, as well as player and team performance, can be deduced. Furthermore, we address the identification and recording of personal talents and statistical categories that distinguish an exceptional goal scorer from the worst goal scorer through football analytics. Feature selection, data size, and parameters used may impact the results of the model. Our research proposes a Goal Prediction Model (GPM) with player analysis trained on data from 9,074 games, including 941,009 events from Europe's top 5 leagues containing the information of five seasons. Our model will explain the observations on expected goals through football analytics and monitor the performance of the players with respect to anticipated goals. This research could benefit football team managers and coaches by reshaping the performance of players.

Muscle exhaustion is one of the most common injury types affecting most athletes and even ordinary people. This study aims to understand more about athletes’ muscular fatigue, especially track and field athletes such as sprinters. The electromyography (EMG) shall capture the muscle activation signals, specifically on the leg muscles, as both gastrocnemius muscles quickly became fatigued during the calf raise exercise. The electricity flow through the calf muscle can be measured via EMG, reflecting the amount of fatigue experienced. In this research, three selected athletes were recruited as the test subject. The EMG sensor was attached to the calf muscle of the athlete during the calf raise exercise and the subject was instructed to maintain composure until fatigue set in. Using the ProEMG software, the data was transferred to the computer for further analysis. The preliminary result found that the average median frequency dropped to 6.13 ± 0.03%, and the root mean square amplitude increased by 11.98 ± 0.02% during fatigue conditions. The endurance time for each subject varies, depending on the physical properties of the body and performance experience. The understanding from the findings is pivotal for achieving optimal muscular strength and endurance development and reducing the prospect of training-related injuries to sport-person.

This study aims to investigate the effect of different input images, namely raw data (RAW) and Continuous Wavelet Transform (CWT) towards the discriminating of street skateboarding tricks, i.e., Ollie, Kickflip, Shove-it, Nollie and Frontside 180 through a variety of transfer learning with optimised k-Nearest Neighbors (kNN) pipelines. Six amateur skateboarders participated in the study, executed the aforesaid tricks five times per trick on an instrumented skateboard where six time-domain signals were extracted prior it was transformed to RAW and CWT. It was shown from the study that the CWT-InceptionV3-optimised kNN pipeline could attain an average test and validation accuracy of 90%.

The Ball-Oriented Soccer Simulation (BOSS) is an innovation from soccer match-play simulations previously used in scientific studies that aim to replicate soccer match-play demands. Several shortcomings from the previous simulations may include the utility of unidirectional treadmill running, large spatial requirement, or lack of individual ball-handling actions. With the utility of the multidirectional overground running, compact, BOSS in soccer studies, much of these shortcomings may be easily overcome in a safe, controlled, laboratory setting. The ball-handling activities in the BOSS were designed to closely follow match-play frequencies and distances to create an ecologically valid soccer match-play simulation. Mean heart rate and rating of perceived exertion during the BOSS was found to be similar to previous overground soccer match simulations (HR: 155±6 bpm; RPE: 14±1). Practical applications of the BOSS include its utility in multiple research disciplines such as exercise physiology, nutrition, sports psychology, sports biomechanics as well as skills and motor control. The BOSS may also be used in real-world settings such as using the protocol on injury risk screening during pre-season or as a return-to-play biomechanical assessment protocol for players following post-injury or post-surgical rehabilitation. In the future, perhaps the BOSS may be used as a blueprint for the development of different sport-specific simulation protocol for the betterment of sports research as well as sports equipment production.

The Internet of Things (IoT) now plays an important role in sports especially in tracking athlete’s development. IoT offers limitless opportunities to monitor athlete’s growth by creating personalized measuring instrument for athletes. In sports, track and field routine, protocol as well as teaching events is one of the areas that could benefit from IoT. Speed and agility training for sport requires a variety of training approaches, much as strength and power training does. A few approaches can be used to measure speed for athletes, including timing gate. However, the cost of developing a timing gate system is on the high side, hence there is a need to develop a cost-effective timing gate that is reliable and serve the same purpose. The development of timing gate device is used in this study to assess and enhance all aspects of athletic training, including speed, acceleration, responsiveness, power, and elevation. The prototype of the timing gate system is successfully developed by combining hardware and software. This project seeks to improve the architecture of a smart timing gate system for sports applications by utilizing ESP32 Wi-Fi module as the microcontroller and infrared sensors as input to achieve high accuracy. All data are transferred to Blynk apps and later displayed on LCD. This paper outlines the design and execution of the timing gate system application in sports.