Abstract From the observational methodology approach, this study analyses definitive errors or losing blunders, i.e. errors that result in the loss of the game, in elite players at U8 level. An ad hoc observation instrument has been designed as a combination of field format and category systems, based on a thorough theoretical review of the internal logic of chess. The games were compiled in the ChessBase 17 program and analysed using Stockfish 16 NNUE via https://lichess.org/es. The moment in the game when the error occurs is extracted and recorded and coded using Lince software. The reliability of the records from the observation system developed was guaranteed by interobserver agreement, calculated using Cohen’s Kappa coefficient. This paper’s objective is achieved by means of the decision tree analysis technique, obtained using the CHAID procedure, taking the “impact of the error” as the predicted dimension. The results obtained have allowed us to conclude that the errors that lead to the loss of the game for elite U8 players are related to short-term calculation (tactical motifs, undefended pieces or checkmate) as opposed to long-term strategic errors.

Abstract Wearable sensors combined with machine learning (ML) offer a promising approach for estimating joint kinetics in real-world settings, with potential applications in athlete monitoring and injury prevention. However, the variety of sensor configurations in previous studies complicates comparisons and optimal configuration selection. This study compared different wearable sensor configurations, comprising inertial measurement units (IMUs) and pressure insoles (PIs), to determine their influence on the accuracy of ML – based predictions of 3D knee moments during running. Sensor configurations ranged from one to four IMUs, with and without PIs. The dataset consisted of wearable and ground truth knee moment data from 19 recreational runners during treadmill running. Model performance of the convolutional neural networks was evaluated on an independent test set. Hyperparameter optimization (HPO) was applied to refine model architectures and training parameters. Performance gains by PIs and a greater number of IMUs were small but significant. The results after HPO confirmed similar performances between single- and multi-sensor configurations, suggesting only small benefits from additional sensors. Our findings highlight that both sensor configuration and model optimization play critical roles in achieving optimal performance. We provide practical recommendations for sensor selection, balancing accuracy and feasibility, to enable biomechanical assessments in real-world environments.

Abstract This paper presents a novel approach to analyzing basketball games. It uses image processing techniques to track player movements, evaluate passes and shots, and visualize game dynamics. The system employs player and ball detection methods, leveraging appearance embedding-based particle filters for robust tracking across consecutive frames. We generate trajectory diagrams that provide insights into team strategies and player performance by applying projective transformation to map coordinates from player feet to the basketball court. Key challenges addressed include improving tracking accuracy under dynamic conditions, minimizing over-detections in pass and shot judgment, and refining ball possession calculations. Experimental results show high tracking accuracy for players, but lower performance in ball tracking and shot detection, particularly in high-speed movements or when objects are occluded. The analysis also revealed that player and team behaviors, such as passing success rates and movement patterns, could be effectively visualized through trajectory diagrams. While the current system provides valuable insights into game strategies, further improvements are needed, particularly in enhancing the reliability of tracking, judgment of passes and shots, and clarity of trajectory in dense sequences of plays.

Abstract This paper presents the bivariate Poisson-new XLindley distribution (BPNXLD), which may be used to represent dependent and over-dispersed countdata. Among the characteristics considered are the correlation coefficient, mean, and variance of the distribution. Acomparison with several Bivariate distributions is included. The goodness of fit of this novel model iscompared with the bivariate Poisson, bivariate negative binomial and bivariate Poisson-Lindley distributions using two data sets from a German Bundesliga season.

Abstract We make a unique contribution to momentum research by proposing a way to quantify momentum with performance indicators (i.e., features). We argue that due to measurable randomness in the NHL, sequential outcomes’ dependence or independence may not be the best way to approach momentum. Instead, we quantify momentum using a small sample of a team’s recent games and a linear line of best-fit to determine the trend of a team’s performances before an upcoming game. We show that with the use of SVM and logistic regression these momentum- based features have more predictive power than traditional frequency-based features in a pre-game prediction model which only uses each team’s three most recent games to assess team quality. While a random forest favors the use of both feature sets combined. The predictive power of these momentum-based features suggests that momentum is a real phenomenon in the NHL and may have more effect on the outcome of games than suggested by previous research. In addition, we believe that how our momentum-based features were designed and compared to frequency-based features could form a framework for comparing the short-term effects of momentum on any individual sport or team.

Abstract In this paper, we proposed a new method of evaluating horse ability and jockey skills in horse racing. In the proposed method, we aimed to estimate unobservable individual effects of horses and jockeys simultaneously with regression coefficients for explanatory variables such as horse age and racetrack conditions and other parameters in the regression model. The data used in this paper are records on 1800­m races (excluding steeplechases) held by the Japan Racing Association from 2016 to 2018, including 4063 horses and 143 jockeys. We applied the hierarchical Bayesian model to stably estimate such a large amount of individual effects. We used the Markov chain Monte Carlo (MCMC) method coupled with Ancillarity- Sufficiency Interweaving Strategy for Bayesian estimation of the model and choose the best model with Widely Applicable Information Criterion as a model selection criterion. As a result, we found a large difference in the ability among horses and jockeys. Additionally, we observed a strong relationship between the individual effects and the race records for both horses and jockeys.

Abstract The Gothenburg Half Marathon is one of the world’s largest half marathon races with over 40 000 participants each year. In order to reduce the number of runners risking over-straining, injury, or collapse, we would like to provide runners with advice to appropriately plan their pacing. Many participants are older or without extensive training experience and may particularly benefit from such pacing assistance. Our aim is to provide this with the help of machine learning. We first analyze a large publicly available dataset of results from the years 2010 - 2019 (n = 423 496) to identify pacing patterns related to age, sex, ability, and temperature of the race day. These features are then used to train machine learning models for predicting runner’s finish time and to identify which runners are at risk of making severe pacing errors and which ones seem set to pace well. We find that prediction of finish time improves over the current baseline, while identification of pacing patterns correctly identifies over 70% of runners at risk of severe slowdowns, albeit with many false positives.

Abstract Simulating and predicting tournament outcomes has become an increasingly popular research topic. The outcomes can be influenced by several factors, such as attack, defence and home advantage strength values, as well as tournament structures. However, the claim that different structures, such as knockout (KO), round-robin (RR) and hybrid structures, have their own time restraints and requirements has limited the evaluation of the best structure for a particular type of sports tournament using quantitative approaches. To address this issue, this study develops a decision support system (DSS) using Microsoft Visual Basic, based on the object-oriented programming approach, to simulate and forecast the impact of the various tournament structures on soccer tournament outcomes. The DSS utilized the attack, defence and home advantage values of the teams involved in the Malaysia Super League 2018 to make better prediction. The rankings produced by the DSS were then compared to the actual rankings using Spearman correlation to reveal the simulated accuracy level. The results indicate that a double RR produces a higher correlation value than a single RR, indicating that more matches played provide more data to create better predictions. Additionally, a random KO predicts better than a ranking KO, suggesting that pre-ranking teams before a tournament starts does not significantly impact the prediction. The findings of this study can help tournament organizers plan forthcoming games by simulating various tournament structures to determine the most suitable one for their needs.

Abstract Typical performance indicators in professional quantitative soccer analysis simplify complex matters, resulting in loss of information. Hence, a novel approach to characterize the performance of soccer teams was investigated: Success-Scores, combining space control with ball control and the correlation between the two. Success-Score Profiles were calculated for 14 games from the German Bundesliga. The dataset was split into two groups: all data points above resp. below the 80 th percentile of Success-Scores. Subsequently, the relative goalscoring frequency in those two groups was compared. All data points were sorted according to their Success-Score and split into equally sized eighths. These groups were tested for a rank order correlation with the number of scored goals. Finally, the Success-Scores of two teams with different success levels as well as their opponents’ Success-Scores were compared. Results indicated significantly higher goalscoring frequencies above the 80th percentile for Success-Scores and a statistically significant rank order correlation between the Success-Scores and the number of scored goals, r s (6) = 0.73, p = .04. The more successful team showed significantly higher Success-Scores. This novel performance indicator shows significant connections to success defined as scoring goals and final ranking in elite soccer and therefore shows potential in reconizing underlying performance.

Abstract In team sport Human Activity Recognition (HAR) using inertial measurement units (IMUs) has been limited to athletes performing a set routine in a controlled environment, or identifying a high intensity event within periods of relatively low work load. The purpose of this study was to automatically classify locomotion in an elite sports match where subjects perform rapid changes in movement type, direction, and intensity. Using netball as a test case, six athletes wore a tri-axial accelerometer and gyroscope. Feature extraction of player acceleration and rotation rates was conducted on the time and frequency domain over a 1s sliding window. Applying several machine learning algorithms Support Vector Machines (SVM) was found to have the highest classification accuracy (92.0%, Cohen’s kappa Ƙ = 0.88). Highest accuracy was achieved using both accelerometer and gyroscope features mapped to the time and frequency domain. Time and frequency domain data sets achieved identical classification accuracy (91%). Model accuracy was greatest when excluding windows with two or more classes, however detecting the athlete transitioning between locomotion classes was successful (69%). The proposed method demonstrated HAR of locomotion is possible in elite sport, and a far more efficient process than traditional video coding methods.