Aerodynamic drag dominates the resistive forces in many sports at racing speeds, and small athlete posture changes can produce practically meaningful changes in performance. Standard approaches like field testing, wind-tunnel testing and computational fluid dynamics (CFD) can provide accurate results, but are time-consuming. Recent advances in scientific machine learning have enabled surrogate models that predict flow quantities at a fraction of the computational cost of a full CFD simulation. However, most applications currently focus on industrial geometries, such as cars or aircraft, where the surface is morphed to generate large datasets for training. It remains unclear how well these methods transfer to athlete geometries where variability is dominated by articulated pose changes rather than smooth shape morphing. In this work, a dataset for cyclist aerodynamics was generated by combining 12 scanned athlete geometries with 20 postures per athlete. CFD simulations were performed for these 240 geometries, which were then used to train a state-of-the-art surrogate model. The generalization of the model to an unseen geometry was investigated, along with the balance between number of positions and number of unique geometries in the dataset. The surrogate model predicts drag area with a mean absolute percentage error of < 3
Protective padding is applied to fixed obstacles on ski slopes with the aim of reducing the injury severity experienced by snow sport participants during collision events. For high-velocity impacts, the protective performance of such protection devices may be limited. This study aimed to evaluate the head and thoracic injury risks during collisions with obstacle and to assess how these risks could be reduced by coupling multiple protective devices (pads and nets). Ten collisions with obstacle were reconstructed in a laboratory with an instrumented Hybrid III dummy projected at 30 km/h in a lying position against a fixed vertical pole. Two dummy orientations and five obstacle protections (coupling nets and pads) were investigated. Head linear accelerations, head angular velocities, and thoracic accelerations were recorded during the tests to estimate injury risks. Each test was filmed with two high-speed cameras. With one pad on the obstacle, the estimated severity of the collision ranged between moderate and severe head injury, depending on the dummy orientation. Coupling a pad on the obstacle with another pad or a net reduced the estimated severity of head and thoracic injuries. These tests provide insight into the performance and limits of the current pads to protect against injuries for impacts up to 25 km/h. They also provide insight into the added potential value provided by coupling different protection devices (pads and nets).
This study investigated the independent and combined effects of incline and speed on the kinetics and cycle characteristics of double poling. Seventeen trained male cross-country skiers completed a maximum speed, a maximum incline, and a constant-power-output test on a motorized treadmill. Axial pole force was measured throughout all tests using ski poles equipped with force sensors. The results indicated that the proportion of propulsive pole force relative to axial pole force (i.e., the pole force ratio) decreased linearly with increasing speed, which was explained by declining poling time. Contrarily, the pole force ratio and poling time barely changed by increasing incline; however, considerable inter-individual variation emerged in the former. A large proportion of variables describing cycle characteristics and poling kinetics differed significantly between the final completed stages of the maximum speed and maximum incline tests. Additionally, participants’ performance in these tests showed a moderate correlation ( r = 0.50 , P = 0.04 ). Both pole forces and poling impulse, as well as the pole force ratio, showed significant variation during the constant-power-output test, suggesting that, in cross-country skiing, external power output alone is insufficient to accurately capture the magnitude of effort, unless speed and inclination are also considered. The present study highlights the complexity of double-poling biomechanics and underscores the need for further research into physiology- and efficiency-related implications. Finally, it provides practical perspectives for enhancing sprint performance in cross-country skiing through targeted training sessions.
Head impacts are common in youth collision sports such as ice hockey and can contribute to brain injury. Impact biomechanics are influenced by characteristics such as impact magnitude, frequency, inter-impact interval, and cumulative exposure duration. Player velocity, particularly closing velocity, affects collision energy transfer and is therefore a key kinematic variable for characterizing head-impact events. To enable large-scale extraction of head-impact characteristics from standard youth game video, this study presents a proof-of-concept automated pipeline to estimate planar player velocity from a single panning side-view camera. A YOLOv8 player detector pre-trained on approximately 80,000 National Hockey League images was fine-tuned on a youth player-detection dataset (4200 training images with light/dark jersey labels), achieving mean average precision at an intersection-over-union threshold of 0.5 of 0.97. Players were tracked using an intersection-over-union-assisted StrongSORT configuration, reducing identity switches from 172 to 53 across 100 head-impact–centered clips (3 s; 90 frames at 30 frames/s) and improving multiple object tracking accuracy from 89.0
The goal of this study was to develop a test methodology for use in studying cleat-turf mechanical interaction that permits laboratory testing, precision-controlled inputs, variation in normal load applied to the cleat, and imaging of the cleat-turf interaction. This new test method utilized a 6 DOF force/torque and position-controlled serial robotic test device to perform compress-then-shear tests between a cleated shoe surrogate and artificial turf samples. The test method proved to be repeatable in displacement-controlled inputs across variations in axial load. Force response data were also repeatable, but variations in shear response were indicative of inherent variability in turf construction. Precise identification of cleat-turf release was permitted through high-speed imaging and verified to occur at the time of peak horizontal force. A linear relationship was found between release shear force and normal force at time of release for both posterior and lateral shear test types under the loading regime investigated (R2 = 0.96, R2 = 0.97). This linear regression represents a “release traction” that can be used to help better understand how turf construction relates to cleat-turf interaction mechanics, which is necessary to optimize player safety and performance.
The use of microtechnologies in ice hockey is an emerging approach to the assessment of on-ice practice and match demands. This review aims to explore the use of wearable technologies to assess the locomotor and body external demands in ice hockey, provide position-specific normative data for practitioners and understand the in-depth insights these devices can measure. A scoping review was conducted to ensure methodological transparency and replicability. This design was selected to map the current literature and identify knowledge gaps regarding the integration of wearable technologies in ice hockey. The search strategy followed the specific framework to define eligibility criteria. Inclusion and exclusion criteria were applied to progressively focus on original research using wearable technologies that were used to quantify physical demands of ice hockey. A total of 22 articles were retained for the final analysis. Research predominantly focused on highly trained or elite male adult athletes, with a widespread 100-Hz inertial movement units application compared to Local Positioning System. Accelerations, decelerations, skating distance, PlayerLoad™, skating velocity zones distances, PlayerLoad·min−1, Explosive Efforts and skating velocity were the variables most assessed in the literature. Physical demand metrics were grouped into volume, intensity, and density categories, with volume-based being the most frequently reported. Wearable technologies are increasingly implemented in ice hockey to monitor on-ice performance across training and competition. Future research should aim to assess the validity and reproducibility of the variables extensively used and contextualize data more deeply to enhance training design and optimize match preparation.
There is limited understanding about players perceptions and physical demands of wheelchair tennis grass play. This study conducted a quantitative pilot study exploring wheelchair configurations, court hardnesses, and potential court damage. Insights informed the primary qualitative investigation into player perceptions and best practice for wheelchair tennis on grass. Performance data were collected from six players during three tests using different configurations on grass, with hardness values between 197-233Gmax. Sixteen semi-structured interviews with players and support staff offered insights about wheelchair tennis grass play. Qualitative interview findings revealed two themes: navigating performance and de-mystifying wheelchair tennis. On-court findings indicated thicker castors and tyres resulted in reduced rolling resistances (− 8
The purpose of this study was to present a method to convert spatiotemporal ball tracking data into equations of motion, thereby facilitating a greater understanding in how different variables influence the trajectory of a tennis ball. Interpreting a tennis ball trajectory via equations of motion enables characterising a tennis shot based on only the initial launch parameters off the racket (speed, direction and spin) and accounting for the influence of the ball inertial and aerodynamic (drag and lift) properties. This study used spatiotemporal ball tracking data (used for electronic line-calling purposes) collected at the 2022 Australian Open for demonstration and validation, with a focus on the tennis serve. From an assessment of 8654 serves, the median Mean Absolute Error (MAE) between the spatiotemporal ball tracking data and the equations of motion trajectory solution was 4.8 mm for the ball trajectory from racket impact through to the bounce, and 2.4 mm for the rebound trajectory after the bounce. The derived trajectory parameters from the study were consistent with expectations from research literature for drag and lift coefficients for a tennis ball, as well as the expected spin profiles for flat, slice and kick serves. The results also enabled quantifying the influence on the ball drag coefficient from repeated impacts, a phenomenon characterised by tennis balls progressively becoming fluffier with use.
Videogrammetry can quantify head acceleration events in sport, but because standard datasets lack the large rotations, rapid motion, and frequent occlusion characteristic of sports collisions, the accuracy of modern deep learning pose estimators in this context remains unclear. This study addresses this gap by benchmarking three models for monocular head pose estimation during controlled football headers: a direct head pose regressor, an end-to-end face reconstruction model, and a full-body human mesh recovery model. Ten participants performed linear and rotational headers. Synchronised 1000 Hz infrared motion capture provided ground-truth orientations, while dual 50 Hz video cameras supplied frontal and side views. Model outputs from standardised detections were temporally smoothed and evaluated using geodesic and incremental geodesic error metrics. All models achieved single-digit mean geodesic (4°–8°) and incremental geodesic (<4°) errors. SAM 3D yielded the lowest mean errors (4.59° and 1.99°, respectively) and showed lower sensitivity to occlusion and temporal impact phase. By uniquely comparing head-only and full-body approaches, results demonstrate that modern full-body human mesh recovery models outperform dedicated head pose estimators under the heavy occlusion and dynamic conditions typical of sports collisions. Errors increased for side-view footage, rotational trials, and low facial visibility. These findings support using deep learning, particularly full-body mesh recovery, for semi-automated videogrammetric reconstruction of head acceleration events.
The aim of this study was to determine the differences in characteristics between traditional waterfowl-feather and officially approved synthetic-feather shuttlecocks. The following tests were conducted as part of this study: (1) physical measurements of the synthetic-feather shuttlecock; (2) human tests; (3) wind tunnel tests; and (4) 2D flight simulations. The synthetic-feather shuttlecock was significantly heavier by 0.1 g (1.9
Taekwondo is a martial arts sport involving kicks to the head. Participants must wear headgear to reduce injury risk, yet no test standards or performance thresholds exist for this protective equipment. The aim of this study was to test headgear under simulated use conditions to quantitatively compare scenarios relevant to head injury. Four commonly used brands of Taekwondo headgear were subjected to seven different tests: simulated aging, typical use, ultraviolet light, repeated laboratory impacts, temperature, moisture, and simulated mechanical damage. Headgear performance was assessed using a drop tower and an instrumented headform from which a Gadd Severity Index was found. The severity index differed by up to a factor of four between the headgear tested. Simulated aging was found more likely to decrease a headgear’s performance than use or simulated mechanical damage. Environments occurring in play, such as elevated temperatures and moisture, also led to a temporary decrease in headgear performance.
Player and ball tracking data derived from broadcast offers a cost-effective and scalable alternative to multi-camera optical tracking systems in football. However, the practical adoption of broadcast tracking systems depends critically on the accuracy of the data they produce. This study examined the accuracy of using broadcast-derived player and ball position data for automatic event detection for a 90 min match from the 2022 FIFA World Cup. The results were compared against events generated from a high-definition multi-camera optical tracking system (TRACAB Gen 5, ChyronHego, New York, USA) and manually tagged events from FIFA’s Data Collection Unit. The results showed that broadcast-derived auto-events, particularly from Camera 1, have the potential to reach the accuracy of multi-camera optical tracking systems for certain events. The best-case performance for most events examined in this study either exceeded, matched, or fell within 0.05 F1-score of the multi-camera system performance. However, performance varied across the systems and was limited by instances of player visibility, ball tracking errors and subjectivity in event definitions, particularly around set pieces and shots. The findings of this study highlights both the capabilities and current limitations of broadcast tracking technologies, provides guidance for their appropriate use and informs future efforts to improve data accuracy in applied contexts.
The inclusion of Coastal Rowing Beach Sprints in the 2028 Olympic program is reshaping the sport by mainstreaming racing in challenging and unpredictable open-water environments, characterized by waves, sharp turns, and constantly changing conditions. Despite its rapid growth in popularity, further systematic research on how to enhance performance, excitement and enjoyment while maintaining athlete safety remains limited. As a result, athletes, event organisers, equipment manufacturers, and governing bodies face a steep learning curve. This work highlights the key technical and safety aspects of the discipline that warrant further development and proposes conceptual solutions grounded in science and engineering. It also outlines a preliminary roadmap for future research aimed at advancing safety, performance, and the overall athlete and spectator experience in coastal rowing—an endeavour that demands collaboration between the sporting, scientific, and engineering communities.
The playing surface in football plays a pivotal role in player performance, safety, and overall game quality. With evolving regulations and continuous technological advances in football surfaces (natural, hybrid, and artificial turf), understanding their performance characteristics is essential. This study evaluated the vertical compliance and rotational stiffness of football surfaces using two test devices adopted by the FIFA Quality Programme: the Advanced Artificial Athlete and the Rotational Traction Athlete. Measurements were taken from 126 football pitches and turf samples, assessed in-situ and in the laboratory, respectively, which encompassed various natural, hybrid, and artificial turf systems. Results indicated significant differences in vertical compliance, with artificial turfs exhibiting higher peak shock absorption, greater peak deformation, and increased energy return compared to natural and hybrid turfs, suggesting that artificial turfs are more compressible and more elastic. Rotational stiffness measurements revealed that while peak torque values were similar across most surface types, natural and hybrid turfs demonstrated significantly higher resistive torque at 10 degrees of rotation than artificial systems. The findings of this study informed revisions of the performance thresholds adopted within the FIFA Quality Programme, which specify the global quality requirements for football playing surfaces.
This study compared the effects of artificial football surfaces incorporating different vegetal infills on players’ physical performance. It also examined how these systems differed from natural grass and from traditional third‑generation artificial turf using recycled rubber from end‑of‑life tyres. A quasi-experimental field design was conducted with 30 amateur male football players who completed a standardised battery of sprint, agility, and fatigue tests across seven surface types: five artificial turfs with vegetal infills (processed olive pits, pinecone granules, corn cob, cork cob, and wood chips), one with rubber infill, and one natural grass pitch. Performance metrics included sprint velocity, acceleration, force, repeated sprint ability, fatigue countermovement jump, and change of direction time. All artificial turf systems complied with the FIFA Quality Programme for Football Turf and were verified using FIFA‑standardised mechanical test procedures for surface–player interaction. Mechanical properties were assessed at five locations per field using standardised test methods, including shock absorption, deformation, energy return, and rotational resistance. Sprint performance was broadly comparable across all surfaces, with players achieving similar maximal sprint velocities regardless of infill type or reference surface (natural grass or ELT). Wood chip infill was linked to lower force production and slower sprint times. No single vegetal infill outperformed others across all variables, highlighting the need for individual evaluation. These findings indicate that some vegetal infills may provide performance characteristics comparable to existing systems, although their suitability for wider implementation will depend on further long‑term, environmental, and context‑specific evaluations.
This study aimed to propose and apply a validation method for a novel three-axis force sensor integrated into a cycling pedal, which offers several innovative features, notably compatibility with any bike or stationary cycle and fully wireless operation. A measurement rig was designed to compare the outputs of the pedal forces sensor with those of a reference sensor. A functional loading protocol—during which the pedal forces sensor was tested in seven different orientations—was performed to assess the pedal forces sensor validity in both directions of each axis. Data from the pedal forces sensor and the reference sensor were compared using difference plot, mean bias, intra-class correlation, correlation coefficient and root mean squared error. The maximum root mean squared error was observed for the normal axis (20 N). The validation method indicates that the pedal forces sensor demonstrated excellent agreement with the reference sensor for foot pedal reaction forces measurement and could be adapted to other force sensors.
Aerodynamic optimization has gained increasing attention in competitive cycling. It is widely believed that aerodynamics plays a more critical role at higher riding speeds, as the power required to overcome aerodynamic drag is proportional to the cube of the riding speed, making it increasingly dominant at higher riding speeds. This study challenges the applicability of this paradigm for sprint cycling. We demonstrate that neither riding speed nor the percentage of power output used to overcome aerodynamic drag can adequately reflect the importance of aerodynamics in sprint cycling, due to the significant role of transient dynamics in sprint cycling. To address this gap, we present a theoretical framework based on perturbation analysis to examine marginal gains in sprint time resulting from time-dependent parameter variations. Our analysis also explores the trade-off between power output and aerodynamic efficiency. Through several case studies of standing-start sprints, we establish a quantitative relationship between aerodynamic improvements and savings in sprint time. Furthermore, we propose a metric to explicitly quantify the time-dependent importance of aerodynamic drag on sprint performance. Overall, this study advances the understanding of performance optimization in sprint cycling and provides a practical tool for marginal gain analysis.
Sports turf surfaces, including natural turfgrass and synthetic turf, are complex systems with many parameters influencing their performance. This study aims to classify sports turf surfaces using data collected from a bespoke testing device, fLEX, which calculates seven separate metrics related to sports surface performance in both an acceleration format (designed to simulate an athlete accelerating) and deceleration format (designed to simulate an athlete decelerating). Sixty-eight collegiate and professional sports surfaces across the USA and UK were tested, covering a range of climates and field constructions. Surfaces were classified as cool-season, warm-season, or synthetic turf. After data preprocessing, including outlier removal and imputation, two machine learning models, decision tree and random forest, were trained and tested on the dataset. Feature importance was assessed using mutual information, revealing that recoil distance and maximum vertical force were the most critical variables for classification. The decision tree model achieved an accuracy of 84% for acceleration and 79% for deceleration, while the random forest model performed slightly better, with accuracies of 89% and 83%, respectively. Both models demonstrated low overfitting risk, with a minimal difference between training and testing accuracies. Misclassifications were analysed, highlighting the complexity of surface characteristics and potential for improving classification accuracy. The high performing models suggest that the fLEX testing device is an appropriate tool to classify the surfaces, and that unique characteristics exist within each surface category. Collectively, these findings represent a step toward advancing our understanding of the complexity of sports turf surfaces.
Ice hardness and friction influence performance and safety in ice sports, yet the role of water quality remains poorly understood. This study examines how total dissolved solids in rink water affect ice characteristics using nondestructive hardness testing and a skate-to-ice static friction index measured under controlled rink conditions. Results show that temperature influences ice hardness and static friction, with colder ice being harder and exhibiting higher static friction. Water quality also plays a role, as lower total dissolved solid levels reduce static friction, with effects comparable to a 1C increase in ice temperature. However, ultra-pure water produced softer ice, while moderate total dissolved solid levels maintained ice hardness with minor static friction increases. These findings support hockey rink recommendations to maintain total dissolved solids near 80 to 100 ppm to balance ice hardness with friction performance. However, other sports, such as curling, may benefit from lower total dissolved solids for reduced static friction. Understanding these relationships informs best practices in rink maintenance and water treatment.
