<abstract><title style='display:none'>Abstract</title>The increasing availability of data from sports events has led to many new directions of research, and sports analytics can play a role in making better decisions both within a club and at the level of an individual player. The ability to objectively evaluate individual players in team sports is one aspect that may enable better decision making, but such evaluations are not straightforward to obtain. One class of ratings for individual players in team sports, known as plus-minus ratings, attempt to distribute credit for the performance of a team onto the players of that team. Such ratings have a long history, going back at least to the 1950s, but in recent years research on advanced versions of plus-minus ratings has increased noticeably. This paper presents a comprehensive review of contributions to plus-minus ratings in later years, pointing out some key developments and showing the richness of the mathematical models developed. One conclusion is that the literature on plus-minus ratings is quite fragmented, but that awareness of past contributions to the field should allow researchers to focus on some of the many open research questions related to the evaluation of individual players in team sports.</abstract>

A comprehensive review of plus-minus ratings for evaluating individual players in team sports

<abstract><title style='display:none'>Abstract</title>The purpose of the present study was to demonstrate an inductive approach for dynamically modelling sport-related injuries with a probabilistic graphical model. Dynamic Bayesian Network (DBN), a well-known machine learning method, was employed to illustrate how sport practitioners could utilize a simulatory environment to augment the training management process. 23 University of Iowa female student-athletes (from 3 undisclosed teams) were regularly monitored with common athlete monitoring technologies, throughout the 2016 competitive season, as a part of their routine health and well-being surveillance. The presented work investigated the ability of these technologies to model injury occurrences in a dynamic, temporal dimension. To verify validity, DBN model accuracy was compared with the performance of its static counterpart. After 3 rounds of 5-fold cross-validation, resultant DBN mean accuracy surpassed naïve baseline threshold whereas static Bayesian network did not achieve baseline accuracy. Conclusive DBN suggested subjectively-reported stress two days prior, subjective internal perceived exertions one day prior, direct current potential and sympathetic tone the day of, as the most impactful towards injury manifestation.</abstract>

Decision Support System for Mitigating Athletic Injuries

<abstract><title style='display:none'>Abstract</title>The intention of Key Performance Indicators (KPI) is to map complex system-behaviour to single values for scaling, rating and ranking systems or system components. Very often, however, this mapping only reduces important information about tactical behaviour or playing dynamics without replacing it by useful ones. The presented approach tries to bridge the gap between complex dynamics and numerical indicators in the case of offensive effectiveness in soccer in two steps. First, a model is developed which visualises offensive actions in a process-oriented way by using information units to represent offensive performance – i.e. Key Performance Indicators. Second, this model is organised in relation to time intervals, which enables to measure the effectiveness for a whole half-time as well as for arbitrary intervals of any desired lengths.This contribution is meant as an introduction to a new modelling idea, where examples are calculated as case studies to demonstrate how it works. Therefore, only two games have been exemplarily analysed yet: The first one, which is used to demonstrate the method, is an example for similar quantitative indicators but different dynamic behaviour. The last one is used to demonstrate the results in the case of teams with extreme different strengths.</abstract>

A Pilot Study on Offensive Success in Soccer Based on Space and Ball Control – Key Performance Indicators and Key to Understand Game Dynamics

<abstract><title style='display:none'>Abstract</title> AIM: The current investigation aimed to create an objective rating of Gaelic football teams and to examine factors relating to a team's rating. METHOD: A modified version of the Elo Ratings formula (Elo, 1978) was used to rate Gaelic football teams. A total of 1101 competitive senior Inter County matches from 2010-2015 were incorporated into calculations. Factors examined between teams included population, registered player numbers, previous success at adult and underage levels, financial income from the GAA, team expenses and number of clubs in a county. RESULTS: The Elo Ratings formula for Gaelic football was found to have a strong predictive ability, correctly predicting the result in 72.90% of 642 matches over a 6 year period. Strong positive correlations were observed between previous success at senior level, Under 21 level, Under 18 level and current Elo points. Moderate correlations exist between population figures and current Elo points. Moderate correlations are also evident between the number of registered players in a county and the county’s Elo rating points. CONCLUSION: Gaelic football teams can be objectively rated using a modified Elo Ratings formula. In order to develop a successful senior team, counties should focus on the development of underage players, particularly up to U18 and U21 level.</abstract>

A Rating System For Gaelic Football Teams: Factors That Influence Success

<abstract><title style='display:none'>Abstract</title>In this paper, we critically evaluate the performance of nine machine learning classification techniques when applied to the match outcome prediction problem presented by American Football. Specifically, we implement and test nine techniques using real-world datasets of 1280 games over 5 seasons from the National Football League (NFL). We test the nine different classifier techniques using a total of 42 features for each team and we find that the best performing algorithms are able to improve one previous published works. The algoriothms achieve an accuracy of between 44.64% for a Guassian Process classifier to 67.53% with a Naïve Bayes classifer. We also test each classifier on a year by year basis and compare our results to those of the bookmakers and other leading academic papers.</abstract>

A Critical Comparison of Machine Learning Classifiers to Predict Match Outcomes in the NFL

<abstract><title style='display:none'>Abstract</title>In the last thirty years, an increasing interest in sport sciences regarding the analysis of expert athletes’ gaze behavior has become apparent. This narrative review provides an overview of the use of eye tracking systems in high-performance sports from 1987 to 2016. A systematic search of the PubMed, Scopus, SPORTDiscus, and WebofScience databases was conducted. The search was performed using the keywords <italic>eye tracking</italic>, <italic>eye movement</italic>, <italic>gaze behavior/patterns</italic>, and <italic>visual search strategies</italic> in combination with <italic>high-performance sports</italic>, <italic>elite athletes</italic>, <italic>high-class athletes</italic>, <italic>sport experts</italic>, and <italic>top-athletes</italic>. It yielded a total of 86 studies of which almost half were conducted computer-based or in front of a screen. Most studies dealt with the analysis of gaze behavior during dead ball situations while also focusing on differences between expert athletes and novices. More high-quality intervention studies are essential to determine if there are ideal gaze strategies and, if yes, how it is possible to learn/implement these.</abstract>

Eye tracking in high-performance sports: Evaluation of its application in expert athletes

<abstract>
<title style='display:none'>Abstract</title>
Major sports events are watched by millions around the world and the prediction of event outcomes is a subject of interest to many stakeholders which underlines the relevance of continuous development and improvement of prediction models. This study uses a factorial design methodology to develop and test 18 Artificial Neural Network (ANN) models for the prediction of world championship boxing matches. The methodology was applied to evaluate the individual and collaborative effects of feature selection, ANN architecture and training data selection on the prediction performance of ANNs. Feature selection was found to be the most influential factor on prediction performance with a statistically significant Analysis of Variance (ANOVA) between the feature selection levels and the test accuracy (p-value of 0.012). The collaborative effect of training data selection and feature selection on prediction performance was found to be statistically significant with ANOVA p-value of 0.007. The best performing model achieved a test accuracy of 81.53% which is an improvement to current benchmarks for sports prediction. The findings of this study contribute to the development of future machine learning sports prediction models.
</abstract>

Developing a High-Performance Sports Results Prediction Artificial Neural Network: Case Study on World Championship Boxing

<abstract>
<title style='display:none'>Abstract</title>
This study examines the possibilities of sensor-instrumented training (SIT) in mid-distance running training sessions. Within this framework, variations of ground contact time (GCT) between straight and curved running, as well as GCT as a fatigue indicator, are explored. Seven experienced runners, with two elite female athletes, participated in two training protocols: 15 sets of 400 m with 1-minute rest and five sets of 300 m with 3-minute rest. GCT was calculated using two inertial measurement units (IMU) attached to the athletes’ feet. The running speed of all athletes was measured with wearable GPS devices. GCT showed variations between inner and outer feet, notably during curve running (300m: 2.56%; 400m: 2.35%). However, for the 300m runs, statistically insignificant GCT differences were more pronounced in straight runs (3.54%) than in curve runs (2.56%), contrasting with the typical assumption of higher differences in curve running. A fatigue-indicating pattern is visible in GCT, as well as speed curves. Other data of this study are consistent with prior research that has observed differences between the inner and outer foot during curve running, while our understanding of the development throughout the training session is enhanced. Using SIT can be a valuable tool for refining curve running technique. By incorporating novel sensing technology, the possibilities enhance our understanding of running kinematics and offer an excellent application of SIT in sports.
</abstract>

A Pilot Study in Sensor Instrumented Training (SIT) - Ground Contact Time for Monitoring Fatigue and Curve Running Technique

<abstract>
<title style='display:none'>Abstract</title>
There is an ever-present need to objectively measure and analyze sports motion for the determination of correct patterns of motion for skill execution. Developments in performance analysis technologies such as inertial measuring units (IMUs) have resulted in enormous data generation. However, these advances present challenges in analysis, interpretation, and transformation of data into useful information. Artificial intelligence (AI) systems can process and analyze large amounts of data quickly and efficiently through classification techniques. This study aimed to systematically review the literature on Machine Learning (ML) and Deep Learning (DL) methods applied to IMU data inputs for evaluating techniques or skills in individual swing and team sports. Electronic database searches (IEEE Xplore, PubMed, Scopus, and Google Scholar) were conducted and aligned with the PRISMA statement and guidelines. A total of 26 articles were included in the review. The Support Vector Machine (SVM) was identified as the most utilized model, as per 7 studies. A deep learning approach was reported in 6 studies, in the form of a Convolutional Neural Network (CNN) architecture. The in-depth analysis highlighted varying methodologies across all sports inclusive of device specifications, data preprocessing techniques and model performance evaluation. This review highlights that each step of the ML modeling process is iterative and should be based on the specific characteristics of the movement being analyzed.
</abstract>

The application of Machine and Deep Learning for technique and skill analysis in swing and team sport-specific movement: A systematic review

<abstract>
<title style='display:none'>Abstract</title>
We evaluated differences between a six-week exergame-training and a conventional balance training program on the balance and gait speed of older adults’ (&gt;65 years). Forty-two healthy participants were recruited from independent living and community centers and randomized to one of three groups: exergaming balance training (EBT), conventional balance training (CBT), or control (no training). The participants completed two balance measurements (Fullerton Advanced Balance Scale (FAB) and center of pressure (COP) excursion), and gait speed at pre, post-intervention, and after a three-week follow-up. Both EBT and CBT groups improved their scores on the FAB, COP displacement, and gait speed post-intervention (p&lt;0.05) and these changes were maintained and did not return to pre-training values after three weeks of detraining. The control group scores for FAB and gait velocity values declined (p&lt;0.001) but not COP excursions during the study. This six-week exergame training program improved balance control and gait speed in community-dwelling seniors in a similar fashion to conventional training. Participants’ physical abilities scores improved and were maintained following three weeks of detraining. Exergame-based training therefore may be considered as an intervention that can address balance control and gait speed in older adults. As well improved scores can be maintained with transient or sporadic activity.
</abstract>

Comparison between six-week exergaming, conventional balance and no exercise training program on older adults’ balance and gait speed

<abstract>
<title style='display:none'>Abstract</title>
In the growing field of data analysis in soccer tracking data is analyzed utilizing increasingly complex methods to account for the dynamic, multifactorial nature of the game. One promising approach is the Success-Score combining ball control and space control. The resulting metric is hypothesized to indicate performance levels and to distinguish performance from playing style. Position datasets from one season of the German Bundesliga were analyzed by calculating Success-Scores based on different interval lengths for two different areas. The relative goalscoring frequency above resp. below the 80th percentile and the rank order correlation between goals and Success-Scores was used to assess the relevance of the Success- Score for goalscoring. The influence of the Success-Score on match outcome, accounting for possession and opponent quality was analyzed via mixed linear models. Results indicated a relation between goalscoring and the Success-Scores, as well as a considerable influence of the Success-Scores on match outcome. The mixed linear models allowed to conclude that Success-Scores capture performance rather than just playing style. The results highlight the potential of the general concept of the Success-Score, combining space and ball control. However, the practical value of the Success-Score in its current implementation appears limited and requires further development.
</abstract>

The Success-Score in Professional Football: a metric of playing style or a metric of match outcome?

<abstract>
<title style='display:none'>Abstract</title>
Automatic fault detection is a major challenge in many sports. In race walking, judges visually detect faults according to the rules. Hence, automatic fault detection systems will help a training of race walking without experts’ visual judgement. Some studies have attempted to use sensors and machine learning to automatically detect faults. However, there are problems associated with sensor attachments and equipment such as a high-speed camera, which conflict with the visual judgement of judges, and the interpretability of the fault detection models. In this study, we proposed an automatic fault detection system for non-contact measurement. We used pose estimation and machine learning models trained based on the judgements of multiple qualified judges to realize fair fault judgement. We verified them using smartphone videos of normal race walking and walking with intentional faults in several athletes including the medalist of the Tokyo Olympics. The results show that the proposed system detected faults with an average accuracy of over 90%. We also revealed that the machine learning model detects faults according to the rules. In addition, the intentional faulty walking movement of the medalist was different from that of other walkers. This finding informs realization of a more general fault detection model.
</abstract>

Automatic Detection of Faults in Simulated Race Walking from a Fixed Smartphone Camera

<abstract><title style='display:none'>Abstract</title> To investigate whether training load monitoring data could be used to predict injuries in elite Australian football players, data were collected from athletes over 3 seasons at an Australian football club. Loads were quantified using GPS devices, accelerometers and player perceived exertion ratings. Absolute and relative training load metrics were calculated for each player each day. Injury prediction models (regularised logistic regression, generalised estimating equations, random forests and support vector machines) were built for non-contact, non-contact time-loss and hamstring specific injuries using the first two seasons of data. Injury predictions were then generated for the third season and evaluated using the area under the receiver operator characteristic (AUC). Predictive performance was only marginally better than chance for models of non-contact and non-contact time-loss injuries (AUC&lt;0.65). The best performing model was a multivariate logistic regression for hamstring injuries (best AUC=0.76). Injury prediction models built using training load data from a single club showed poor ability to predict injuries when tested on previously unseen data, suggesting limited application as a daily decision tool for practitioners. Focusing the modelling approach on specific injury types and increasing the amount of training observations may improve predictive models for injury prevention</abstract>

Predictive Modelling of Training Loads and Injury in Australian Football

<abstract><title style='display:none'>Abstract</title> Baseball is a statistically filled sport, and predicting the winner of a particular Major League Baseball (MLB) game is an interesting and challenging task. Up to now, there is no definitive formula for determining what factors will conduct a team to victory, but through the analysis of many years of historical records many trends could emerge. Recent studies concentrated on using and generating new statistics called sabermetrics in order to rank teams and players according to their perceived strengths and consequently applying these rankings to forecast specific games. In this paper, we employ sabermetrics statistics with the purpose of assessing the predictive capabilities of four data mining methods (classification and regression based) for predicting outcomes (win or loss) in MLB regular season games. Our model approach uses only past data when making a prediction, corresponding to ten years of publicly available data. We create a dataset with accumulative sabermetrics statistics for each MLB team during this period for which data contamination is not possible. The inherent difficulties of attempting this specific sports prediction are confirmed using two geometry or topology based measures of data complexity. Results reveal that the classification predictive scheme forecasts game outcomes better than regression scheme, and of the four data mining methods used, SVMs produce the best predictive results with a mean of nearly 60% prediction accuracy for each team. The evaluation of our model is performed using stratified 10-fold cross-validation.</abstract>

Predicting Win-Loss outcomes in MLB regular season games – A comparative study using data mining methods

<abstract><title style='display:none'>Abstract</title>Decision making in sport involves forecasting and selecting choices from different options of action, care, or management. These processes are conditioned by the available information (sometimes limited, fallible, or excessive), the cognitive limitations of the decision-maker (heuristics and biases), the finite amount of available time to make the decision, and the levels of risk and reward. Decision support systems have become increasingly common in sporting contexts such as scheduling optimization, skills evaluation and classification, decision-making assessment, talent identification and team selection, or injury risk assessment. However no specific, formalised framework exists to help guide either the development or evaluation of these systems. Drawing on a variety of literature, this paper proposes a decision support system development framework for specific use in high-performance sport. It proposes three separate criteria for this purpose: 1) Context Satisfaction, 2) Output Quality, and 3) Process Efficiency. Underpinning these criteria there are six specific components: Feasibility, Delivered knowledge, Decisional guidance, Data quality, System error, and System complexity. The proposed framework offers a systematic approach for users to ensure that each of the six components are considered and optimised before, during, and after developing the system. A DSS development framework for high-performance sport should help to improve both short and long term decision-making in a variety of sporting contexts.</abstract>

A development framework for decision support systems in high-performance sport

Article Publishing Charges

Open Access License

Editorial Policy

Template

Instructions for Authors

Guidelines for Authors

International Journal of Computer Science in Sport

 The International Journal of Computer Science in Sport (IJCSS) is published in association with the International Association of Computer Science in Sport (IACSS).  It is a...

Cite Score

SNIP

MNiSW

 The International Journal of Computer Science in Sport (IJCSS) is published in association with the International Association of Computer...

AHEAD OF PRINT

Volume 23 (2024): Issue 2 (June 2024)

Volume 23 (2024): Issue 1 (February 2024)

Volume 22 (2023): Issue 2 (August 2023)

Volume 22 (2023): Issue 1 (March 2023)

Volume 21 (2022): Issue 2 (December 2022)

Volume 21 (2022): Issue 1 (March 2022)

Volume 20 (2021): Issue 2 (December 2021)

Volume 20 (2021): Issue 1 (July 2021)

Volume 19 (2020): Issue 2 (December 2020)

Volume 19 (2020): Issue 1 (July 2020)

Volume 18 (2019): Issue 3 (December 2019)

Volume 18 (2019): Issue 2 (September 2019) Special Issue: Selected papers presented at the 12th Symposium of the Section Computer Science in Sport of the German Association of Sport Science (September 4.-7., 2018)

International Journal of Computer Science in Sport

Tous les volumes et éditions dans cette revue