An interdisciplinary framework to optimize the anticipation skills of high-level athletes using virtual reality.

The ambition of our contribution is to show how an interdisciplinary framework can pave the way for the deployment of innovative virtual reality training sessions to improve anticipation skills in top-level athletes. This improvement is so challenging that some authors say it is like "training for the impossible". This framework, currently being implemented as part of a project to prepare athletes for the 2024 Olympic Games in Paris, based on the ecological-dynamics approach to expertise, is innovative in its interdisciplinary nature, but also and above all because it overcomes the limitations of more traditional training methods in the field designed to optimize anticipation skills in top-level athletes. The ambition is to tackle successive challenges ranging from the design of virtual partners and opponents to the deployment of training programs in virtual reality, while ensuring the acceptability and acceptance of such innovative virtual reality training protocols and measuring associated workloads.

There is no rush to upgrade the tennis racket in young intermediate competitive players: The effects of scaling racket on serve biomechanics and performance.

Introduction: Scaling the equipment of young athletes is justified by the constraints-led approach introduced in motor learning. The aim of the present study is to analyze the effect of racket scaling on the serve biomechanics and performance parameters for young tennis players (between 8 and 11 years-old). Methods: Nine young intermediate competitive tennis players (age: 9.9 ± 1.0 years) performed maximal effort flat serves with three different rackets (scaled 23 inches, scaled 25 inches and full-size 27 inches) in a randomized order. A radar measured ball speed while shoulder and elbow kinetics and upper and lower limb kinematics were calculated with a 20-camera optical motion capture system. Repeated measures ANOVAs were used to analyze the effect of the three rackets on ball speed, percentage of serve in, serve kinematics and kinetics. Results: No significant differences in ball speed, maximal racket head velocity and percentage of serve in were observed between the three rackets. The lowest maximal upper limb kinetics and the highest upper limb maximal angular velocities were obtained with the scaled 23 inches racket. Discussion: Using scaled rackets has the advantage to decrease shoulder and elbow loadings without reducing serve performance. Consequently, the present results incite tennis coaches and parents to not upgrade too soon the size of the racket in young intermediate tennis players to avoid overuse injury risks in the long term. Our results showed that the full-size 27 inches racket induced higher lower limb kinematics. As a consequence, occasionally serving with a fullsize racket can be a sparingly interesting intervention to help young tennis players to intuitively and immediately increase their leg drive action, allowing a more functional representation of the elite junior serve.

Influence of shoe torsional stiffness on foot and ankle biomechanics during tennis forehand strokes.

Tennis shoe characteristics need to minimise the risk of athletes suffering ankle injuries and improve players' feet performance. This study aims to evaluate the influence of shoe torsional stiffness on running velocity, stance duration, ground reaction forces and ankle biomechanics during two different tennis forehand runs and strokes. Ten right-handed advanced male tennis players performed two specific tennis forehand runs and strokes at maximal effort (a shuttle run with a defensive open stance forehand - SRDF and a lateral jab run with an offensive open stance forehand - JROF) with four different pairs of tennis shoes with different torsional stiffness. A force platform measured ground reaction forces (GRF). A motion capture system recorded the 3D trajectories of markers located on players' anatomical landmarks. The minimum, maximum angle value, and range of motion were computed using inverse kinematics for each rotation axis of the right ankle. Normalised maximal ankle torques were also computed using inverse dynamics. Shoe torsional stiffness had no effect on running velocity, on stance duration and maximal values of GRF. Shoe torsional stiffness influenced forefoot inversion which was significantly higher for the most flexible shoes. For SRDF, the maximal ankle inversion angle was significantly and largely increased for the stiffest shoe. The stiffest shoe may put the ankle at a higher risk of lateral sprains during SRDF while it was not the case during JROF.HighlightsShoe torsional stiffness has no effect on performance parameters (running velocity of the centre of mass, ground reaction forces, and stance duration) during tennis forehand strokes.Decreased shoe torsional stiffness increased the maximal forefoot inversion angle and the range of motion of forefoot inversion-eversion during tennis forehand strokes and movements.Increased footwear torsional stiffness causes higher maximal ankle inversion angle which may increase the risk for ankle sprains in SRDF.

Virtual reality boxing: Gaze-contingent manipulation of stimulus properties using blur.

It has been reported that behavior of experts and novices in various sporting tasks is impervious to the introduction of blur. However, studies have used diverse methods of blurring the visual stimulus (i.e., dioptric blur and Gaussian blur), and tasks that did not always preserve the normal perception-action coupling. In the current study, we developed a novel experimental protocol to examine the effect of different levels of Gaussian blur on interception performance and eye gaze data using an immersive VR task. Importantly, this provided a realistic simulation of a real-world boxing scenario (e.g., the presence of a feint prior to the onset of different combinations of punches) in which expert combat athletes (n = 18) experienced a first-person, adaptive viewpoint of the visual environment, which could be blurred according to their gaze location (central blur, peripheral blur, no blur). We found that participants exhibited similar interception performance in the presence of central blur or peripheral blur compared to a control condition with no blur. However, interception performance was significantly better with a central blur compared to peripheral blur. Eye gaze data indicated that although participants fixated at similar areas of interest irrespective of the presence of blur, fixation duration was significantly longer with a strong level of blur in the peripheral viewing condition than all levels of central blur and the control condition. These findings can be explained by relocating attention to different areas of the environment, which thereby influenced the perception of salient information. Participants also performed better on the first punch of a sequence preceded by a foot feint compared to arm feint or no feint. Still, irrespective of feint type, performance was significantly better on the second and third punch compared to the first punch. These findings are consistent with participants using additional information from the opponent's body movements and situational probabilities to increase performance as the sequence of punches developed. Overall, these are the first evidence for the use of VR as a means to examine gaze-contingent manipulations of the environment, and hence highlight the potential for facilitating learning and transfer to a real sporting situations.

Multiple Players Tracking in Virtual Reality: Influence of Soccer Specific Trajectories and Relationship With Gaze Activity.

The perceptual-cognitive ability to track multiple moving objects and its contribution to team sports performance has traditionally been studied in the laboratory under non-sports specific conditions. It is thus questionable whether the measured visual tracking performance and the underlying gaze activity reflected the actual ability of team sports players to track teammates and opponents on a real field. Using a Virtual Reality-based visual tracking task, the ability of participants to track multiple moving virtual players as they would do on a soccer field was observed to pursue two objectives. (i) See the influence of different scenario types (soccer-specific trajectories versus pseudo-random trajectories) on the visual tracking performance of soccer (n = 15) compared to non-soccer players (n = 16). (ii) Observe the influence of spatial features of the simulated situations on gaze activity between soccer players and non-soccer players. (i) The linear mixed model regression revealed a significant main effect of the group but no interaction effect between group and the type of trajectories, suggesting that the visual tracking ability of soccer players did not benefit from their specific knowledge when they faced scenarios with real game trajectories. (ii) Virtual players' spatial dispersion and crowding affected the participants' gaze activity and their visual tracking performance. Furthermore, the gaze activity of soccer players differed in some aspects from the gaze activity of non-soccer players. Assumptions are formulated as to the implication of these results in the difference in visual tracking performance between soccer players and non-soccer players. Overall, using soccer-specific trajectories might not be enough to replicate the representativeness of the field conditions in the study of visual tracking performance. Multitasking constraints should be considered along with motor-cognitive dual-tasks in future research to develop the representativeness of visual exploration conditions.

Visual tracking assessment in a soccer-specific virtual environment: A web-based study.

The ability to track teammates and opponents is an essential quality to achieve a high level of performance in soccer. The visual tracking ability is usually assessed in the laboratory with non-sport specific scenarios, leading in two major concerns. First, the methods used probably only partially reflects the actual ability to track players on the field. Second, it is unclear whether the situational features manipulated to stimulate visual tracking ability match those that make it difficult to track real players. In this study, participants had to track multiple players on a virtual soccer field. The virtual players moved according to either real or pseudo-random trajectories. The experiment was conducted online using a web application. Regarding the first concern, the visual tracking performance of players in soccer, other team sports, and non-team sports was compared to see if differences between groups varied with the use of soccer-specific or pseudo-random movements. Contrary to our assumption, the ANOVA did not reveal a greater tracking performance difference between soccer players and the two other groups when facing stimuli featuring movements from actual soccer games compared to stimuli featuring pseudo-random ones. Directing virtual players with real-world trajectories did not appear to be sufficient to allow soccer players to use soccer-specific knowledge in their visual tracking activity. Regarding the second concern, an original exploratory analysis based on Hierarchical Clustering on Principal Components was conducted to compare the situational features associated with hard-to-track virtual players in soccer-specific or pseudo-random movements. It revealed differences in the situational feature sets associated with hard-to-track players based on movement type. Essentially with soccer-specific movements, how the virtual players were distributed in space appeared to have a significant influence on visual tracking performance. These results highlight the need to consider real-world scenarios to understand what makes tracking multiple players difficult.

Using Blur for Perceptual Investigation and Training in Sport? A Clear Picture of the Evidence and Implications for Future Research.

Dynamic, interactive sports require athletes to identify, pick-up and process relevant information in a very limited time, in order to then make an appropriate response. Perceptual-cognitive skills are, therefore, a key determinant of elite sporting performance. Recently, sport scientists have investigated ways to assess and train perceptual-cognitive skills, with one such method involving the use of blurred stimuli. Here, we describe the two main methods used to generate blur (i.e., dioptric and Gaussian) and then review the current findings in a sports context. Overall, it has been shown the use of blur can enhance performance and learning of sporting tasks in novice participants, especially when the blur is applied to peripheral stimuli. However, while intermediate and expert level participants are relatively impervious to the presence of blur, it remains to be determined if they are positive effects on learning. In a final section, we describe some of the methodological issues that limit the application of blur and then discuss the potential use of virtual reality to extend the current research base in sporting contexts.

A musculoskeletal modelling approach of the assessment of the risk of hamstring injuries in professional soccer players: a pilot study.

Purpose: Evaluating and minimising the risk of hamstring injury remains complex as it lacks reliable field-testing. Kinematic analysis provides global external insights but fails to apprehend musculoskeletal loading. This study aimed to evaluate the association between hamstring function and prior injury using a novel functional test combined with a musculoskeletal approach.Methods: Twelve professional footballers, distributed in two groups (control or previously injured), performed a reactive functional test to one of four targets from a standing start and performed a knee and plantar extension on target. Joint kinematics served as input data of a musculoskeletal model, and joint angles and hamstring muscle lengths were calculated.Results: Biceps femoris long head (BFlh) was stretched to 150 ± 2% of initial length during the two conditions. Maximal BFlh length and time to stretch were significantly higher in the control group.Discussion: Kinematics and musculoskeletal parameters revealed that participants of the control group had higher maximal hip flexion, pelvis anterior tilt, and hip internal rotation than previously injured players. The combined approach of a hamstring functional test and musculoskeletal modelling gives new preliminary insights on the effect of previous history of hamstring injury on lower limb kinematics and BFlh muscle length.

Influence of path curvature on collision avoidance behaviour between two walkers.

Navigating crowded community spaces requires interactions with pedestrians that follow rectilinear and curvilinear trajectories. In the case of rectilinear trajectories, it has been shown that the perceived action opportunities of the walkers might be afforded based on a future distance of closest approach. However, little is known about collision avoidance behaviours when avoiding walkers that follow curvilinear trajectories. Twenty-two participants were immersed in a virtual environment and avoided a virtual human (VH) that followed either a rectilinear path or a curvilinear path with a 5 m or 10 m radius curve at various distances of closest approach. Compared to a rectilinear path (control condition), the curvilinear path with a 5 m radius yielded more collisions when the VH approached from behind the participant and more inversions when the VH approached from in-front. During each trial, the evolution of the future distance of closest approach showed similarities between rectilinear paths and curvilinear paths with a 10 m radius curve. Overall, with few collisions and few inversions of crossing order, we can conclude that participants were capable of predicting future distance of closest approach of virtual walkers that followed curvilinear trajectories. The task was solved with similar avoidance adaptations to those observed for rectilinear interactions. These findings should inform future endeavors to further understand collision avoidance strategies and the role of-for example-non-constant velocities.

Influence of the forehand stance on knee biomechanics: Implications for potential injury risks in tennis players.

The open stance forehand has been hypothesized to be more traumatic for knee injuries in tennis than the neutral stance forehand. This study aims to compare kinematics and kinetics at the knee during three common forehand stroke stances (attacking neutral stance ANS, attacking open stance AOS, defensive open stance DOS) to determine if the open stance forehand induces higher knee loadings and to discuss its potential relationship with given injuries. Eight advanced tennis players performed eight repetitions of forehand strokes with each stance (ANS: forward run and stroke with feet parallel with the hitting direction, AOS: forward run and stroke with feet perpendicular to the hitting direction, DOS: lateral run and stroke with feet perpendicular to the hitting direction) at maximal effort. All the trials were recorded with an optoelectronic motion capture system. The flexion-extension, abduction-adduction, external-internal rotation angles, intersegmental forces and torques of the right knee were calculated. Ground reaction forces were measured with a forceplate. The DOS increases vertical GRF, maximum knee flexion and abduction angles, range of knee flexion-extension, peak of compressive, distractive and medial knee forces, peak of knee abduction and external rotation torques. Consequently, the DOS appears potentially more at risk for given knee injuries.

Can the Open Stance Forehand Increase the Risk of Hip Injuries in Tennis Players?

BACKGROUND: The open stance forehand has been hypothesized by tennis experts (coaches, scientists, and clinicians) to be more traumatic than the neutral stance forehand as regards hip injuries in tennis. However, the influence of the forehand stance (open or neutral) on hip kinematics and loading has not been assessed. PURPOSE: To compare the kinematics and kinetics at the hip joint during 3 common forehand stances (attacking neutral stance [ANS], attacking open stance [AOS], defensive open stance [DOS]) in advanced tennis players to determine whether the open stance forehand induces higher hip loading. STUDY DESIGN: Descriptive laboratory study. METHODS: The ANS, AOS, and DOS forehand strokes of 8 advanced right-handed tennis players were recorded with an optoelectronic motion capture system. The flexion-extension, abduction-adduction, and external-internal rotation angles as well as intersegmental forces and torques of the right hip were calculated using inverse dynamics. RESULTS: The DOS demonstrated significantly higher values than both the ANS and AOS for anterior (P < .001), medial (P < .001), and distractive (P < .001) forces as well as extension (P = .004), abduction (P < .001), and external rotation (P < .001) torques. The AOS showed higher distractive forces than the ANS (P = .048). The DOS showed more extreme angles of hip flexion (P < .001), abduction (P < .001), and external rotation (P = .010). CONCLUSION: The findings of this study imply that the DOS increased hip joint angles and loading, thus potentially increasing the risk of hip overuse injuries. The DOS-induced hip motion could put players at a higher risk of posterior-superior hip impingement compared with the ANS and AOS. CLINICAL RELEVANCE: Coaches and clinicians with players who have experienced hip pain or sustained injuries should encourage them to use a more neutral stance and develop a more aggressive playing style to avoid the DOS, during which hip motion and loading are more extreme.

Uncovering EEG Correlates of Covert Attention in Soccer Goalkeepers: Towards Innovative Sport Training Procedures.

Advances in sports sciences and neurosciences offer new opportunities to design efficient and motivating sport training tools. For instance, using NeuroFeedback (NF), athletes can learn to self-regulate specific brain rhythms and consequently improve their performances. Here, we focused on soccer goalkeepers' Covert Visual Spatial Attention (CVSA) abilities, which are essential for these athletes to reach high performances. We looked for Electroencephalography (EEG) markers of CVSA usable for virtual reality-based NF training procedures, i.e., markers that comply with the following criteria: (1) specific to CVSA, (2) detectable in real-time and (3) related to goalkeepers' performance/expertise. Our results revealed that the best-known EEG marker of CVSA-increased α-power ipsilateral to the attended hemi-field- was not usable since it did not comply with criteria 2 and 3. Nonetheless, we highlighted a significant positive correlation between athletes' improvement in CVSA abilities and the increase of their α-power at rest. While the specificity of this marker remains to be demonstrated, it complied with both criteria 2 and 3. This result suggests that it may be possible to design innovative ecological training procedures for goalkeepers, for instance using a combination of NF and cognitive tasks performed in virtual reality.

Virtual reality to assess and train team ball sports performance: A scoping review.

Virtual reality (VR) is a widespread technology drawing an increasing interest for players and coaches, especially in team ball sports as it offers a simple tool to simulate, analyse and train situations that are often too complex to reproduce in the field. In this review we aimed at (1) providing an overview of methodologies and outcomes of research studies using VR in team ball sports; (2) better evaluating the potential interest of VR to analyse or train team ball sports situation and (3) identifying limitations, gaps in knowledge and remaining scientific challenges. The MEDLINE and Web of Science Core Collection databases were searched, using predefined combinations of keywords. Thirty articles were retained and analysed. VR can be an interesting tool to assess or train team ball sports skills/situations as it allows researchers to control and standardise situations and focus on specific skills/subskills. Studies that used VR in team ball sports still have some limitations, mainly due to technical issues or study design. This paper also describes the way VR should be used to enhance understanding of performance in team ball sports. Additional suggestions for future research and study design are proposed.

Virtual Reality Gaming Elevates Heart Rate but Not Energy Expenditure Compared to Conventional Exercise in Adult Males.

Virtual reality using head-mounted displays (HMD) could provide enhanced physical load during active gaming (AG) compared to traditional displays. We aimed to compare the physical load elicited by conventional exercise and AG with an HMD. We measured energy expenditure (EE) and heart rate (HR) in nine healthy men (age: 27 ± 5 years) performing three testing components in a randomised order: walking at 6 km/h (W6), AG, and AG with an additional constraint (AGW; wrist-worn weights). Although we found that HR was not significantly different between W6 and the two modes of AG, actual energy expenditure was consistently lower in AG and AGW compared to W6. We observed that playing AG with wrist-worn weights could be used as a means of increasing energy expenditure only at maximum game level, but ineffective otherwise. Our findings indicate that AG in an HMD may not provide a sufficient stimulus to meet recommended physical activity levels despite increased psychophysiological load. The differential outcomes of measures of HR and EE indicates that HR should not be used as an indicator of EE in AG. Yet, adding a simple constraint (wrist-worn weights) proved to be a simple and effective measure to increase EE during AG.

Detection of deceptive motions in rugby from visual motion cues.

Frequently, in rugby, players incorporate deceptive motions (e.g., a side-step) in order to pass their opponent. Previous works showed that expert defenders are more efficient in detecting deceptive motions. Performance was shown to be correlated with the evolution of the center of gravity of the attacker, suggesting that experts may rely on global motion cues. This study aims at investigating whether a representation of center of gravity can be useful for training purposes, by using this representation alone or by combining it with the local motion cues given by body parts. We designed an experiment in virtual reality to control the motion cues available to the defenders. Sixteen healthy participants (seven experts and nine novices) acted as defenders while a virtual attacker approached. Participants completed two separate tasks. The first was a time occlusion perception task, occlusion after 100ms, 200ms or 300ms after the initial change in direction, thereafter participants indicated the passing direction of the attacker. The second was a perception-action task, participants were instructed to intercept the oncoming attacker by displacing medio-laterally. The attacker performed either a non-deceptive motion, directly toward the final passing direction or a deceptive motion, initially toward a false direction before quickly reorienting to the true direction. There was a main effect of expertise, appearance, cut off times and motion on correct responses during both tasks. There was an interaction between visual appearance and expertise, and between motion type and expertise during the perception task, however, this interaction was not present during the perception-action task. We observed that experts maintained superiority in the perception of deceptive motion; however when the visual appearance is reduced to global motion alone the difference between novices and experts is reduced. We further explore the interactions and discuss the effects observed for the visual appearance and expertise.

Biomechanical analysis of the "waiter's serve" on upper limb loads in young elite tennis players.

Waiter's serve (WS) is a specific tennis serve posture frequently observed in young players, and commonly considered as a technical error by tennis coaches. However, biomechanical impact of WS is unknown. The aims of this study were to identify the potential consequences of WS in young elite players relating to performance and injury risk, and to explain the kinematic causes of WS. Serve of 18 male junior elite players (Top 10 national French ranking, aged 12-15 years) was captured with a 20 camera, 200 Hz VICON MX motion analysis system. Depending on their serve technique, the players were divided into two groups (WS versus Normal Serve [NS]) by experienced coaches. Injury data were collected for each player during a 12-month-period following the motion capture. Normalized peak kinetic values of the dominant arm were calculated using inverse dynamics. In order to explain WS posture, upper limb kinematics were calculated during the cocking and the acceleration phases of the serve. Shoulder internal rotation torque, wrist proximal and anterior forces (P < .05) and elbow varus torque (P < .01) were significantly higher in WS group, with no difference from NS group concerning serve velocity. Moreover, significant lower shoulder abduction and higher wrist extension (P < .05) were observed for WS players during the cocking phase. Even if no significant difference was found between groups concerning injuries, higher upper limb joint loads suggested WS could be considered as pathomechanical in young elite players and could lead to upper limb joint injuries.

Walking with Virtual People: Evaluation of Locomotion Interfaces in Dynamic Environments.

Navigating in virtual environments requires using some locomotion interfaces, especially when the dimensions of the environment exceed the ones of the Virtual Reality system. Locomotion interfaces induce some biases both in the perception of the self-motion or in the formation of virtual locomotion trajectories. These biases have been mostly evaluated in the context of static environments, and studies need to be revisited in the new context of populated environments where users interact with virtual characters. We focus on a situation of collision avoidance between a real participant and a virtual character, and compared it to previous studies on real walkers. Our results show that, as in reality, the risk of future collision is accurately anticipated by participants, however with delay. We also show that collision avoidance trajectories formed in VR have common properties with real ones, with some quantitative differences in avoidance distances. More generally, our evaluation demonstrates that reliable results can be obtained for qualitative analysis of small scale interactions in VR. We discuss these results in the perspective of a VR platform for large scale interaction applications, such as in a crowd, for which real data are difficult to gather.

Collision Avoidance Behavior between Walkers: Global and Local Motion Cues.

Daily activities require agents to interact with each other, such as during collision avoidance. The nature of visual information that is used for a collision free interaction requires further understanding. We aim to manipulate the nature of visual information in two forms, global and local information appearances. Sixteen healthy participants navigated towards a target in an immersive computer-assisted virtual environment (CAVE) using a joystick. A moving passive obstacle crossed the participant's trajectory perpendicularly at various pre-defined risks of collision distances. The obstacle was presented with one of five virtual appearances, associated to global motion cues (i.e., a cylinder or a sphere), or local motion cues (i.e., only the legs or the trunk). A full body virtual walker, showing both local and global motion cues, used as a reference condition. The final crossing distance was affected by the global motion appearances, however, appearance had no qualitative effect on motion adaptations. These findings contribute towards further understanding what information people use when interacting with others.

Influence of a Prolonged Tennis Match Play on Serve Biomechanics.

The aim of this study was to quantify kinematic, kinetic and performance changes that occur in the serve throughout a prolonged tennis match play. Serves of eight male advanced tennis players were recorded with a motion capture system before, at mid-match, and after a 3-hour tennis match. Before and after each match, electromyographic data of 8 upper limb muscles obtained during isometric maximal voluntary contraction were compared to determine the presence of muscular fatigue. Vertical ground reaction forces, rating of perceived exertion, ball speed, and ball impact height were measured. Kinematic and upper limb kinetic variables were computed. The results show decrease in mean power frequency values for several upper limb muscles that is an indicator of local muscular fatigue. Decreases in serve ball speed, ball impact height, maximal angular velocities and an increase in rating of perceived exertion were also observed between the beginning and the end of the match. With fatigue, the majority of the upper limb joint kinetics decreases at the end of the match. No change in timing of maximal angular velocities was observed between the beginning and the end of the match. A prolonged tennis match play may induce fatigue in upper limb muscles, which decrease performance and cause changes in serve maximal angular velocities and joint kinetics. The consistency in timing of maximal angular velocities suggests that advanced tennis players are able to maintain the temporal pattern of their serve technique, in spite of the muscular fatigue development.

Influence of Playing a Prolonged Tennis Match on Shoulder Internal Range of Motion.

BACKGROUND: Shoulder range of motion (ROM) deficits have been identified as an injury risk factor among tennis players. It is well known that shoulder internal rotation deficit increases with age and years of play, but there is a lack of knowledge regarding the influence of a prolonged tennis match on shoulder ROM. PURPOSE: To examine changes in shoulder ROM during a prolonged tennis match. STUDY DESIGN: Descriptive laboratory study. METHODS: Shoulder passive internal and external rotation ROM were measured on 8 male tennis players before, every 30 minutes during, and just after a 3-hour tennis match. Total ROM was calculated as the combination of shoulder internal and external rotations. Ball velocity on the serve was measured with a radar gun before, at midmatch, and just after the match. RESULTS: Decreases in shoulder internal rotation (-20.8°; P = .005), total ROM (-24.6°; P = .001), and serve velocity (-1.8 m/s; P = .002) were observed at the end of the match. No statistically significant difference was observed for shoulder external rotation after the match (P = .460). CONCLUSION: Passive shoulder internal rotation and total ROM are significantly decreased during a 3-hour tennis match. CLINICAL RELEVANCE: The results show that a prolonged tennis match play can modify values of shoulder ROM.