Predicting Football Team Performance with Explainable AI: Leveraging SHAP to Identify Key Team-Level Performance Metrics

Moustakidis, Serafeim; Plakias, Spyridon; Kokkotis, Christos; Tsatalas, Themistoklis; Tsaopoulos, Dimitrios

doi:10.3390/fi15050174

Cited by 7 publications

(7 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is noteworthy that some of the above studies [ 16 , 19 ] have incorporated situational variables, as the impact of context on performance is significant [ 22 ]. Additionally, in recent years, artificial intelligence, thanks to the ability to use large volumes of data and a large number of variables, has offered new possibilities in exploring factors that can predict a victorious outcome [ 23 , 24 , 25 ].…”

Section: Introductionmentioning

confidence: 99%

Tactical Situations and Playing Styles as Key Performance Indicators in Soccer

Plakias,

Tsatalas,

Armatas

et al. 2024

JFMK

Self Cite

View full text Add to dashboard Cite

The game of soccer is complex and unpredictable, demanding multifaceted strategies for success. Performance analysis has evolved, focusing on key performance indicators (KPIs) to determine the factors that most significantly influence a team’s success or failure during matches. Traditional performance analysis methods have emphasized quantifiable data like physical exertion and basic play events but often neglected the subtler tactical dimensions that could significantly impact game outcomes. This study aimed to fill the gap in the current literature by creating a comprehensive framework that incorporates tactical situations as KPIs. The objective was to examine whether specific playing styles adopted by teams in various tactical situations and phases of the game could predict the outcome of matches. A dataset comprising all First Division Championship matches from 11 different European countries for the 2021–2022 season was analyzed. Variables representing tactical situations were correlated with match outcomes using a Generalized Estimating Equation framework. The model was specified with a binomial distribution and a logit link function. Statistical significance was determined using Wald χ2 tests with a significance level set at p < 0.05. The study’s findings revealed that possession style, counterattacking during offensive transitions, and a balanced aggressive defensive strategy significantly increase a team’s chances of victory. It also showed that successful teams tend to focus on central attacks, minimize crossing, and execute strategic plays that lead to final attempts on goal with minimal ball possession. The above findings demonstrate that adopting certain tactical approaches significantly influences soccer match outcomes, highlighting the importance of considering tactical aspects as KPIs.

show abstract

Section: Introductionmentioning

confidence: 99%

Tactical Situations and Playing Styles as Key Performance Indicators in Soccer

Plakias,

Tsatalas,

Armatas

et al. 2024

JFMK

Self Cite

View full text Add to dashboard Cite

show abstract

“…Short passing drills can vary in terms of distance, speed and intensity (Arrosyid et al, 2023). Including this variety in practice helps players adapt to the various game situations they may encounter (Moustakidis et al, 2023). Players may perform short passing drills under pressure from teammates or coaches acting as opponents.…”

Section: Introductionmentioning

confidence: 99%

Improved passing accuracy by using pair practice in adolescent soccer players

Marsuna,

Rusli,

Saman

2024

SPORTIF

View full text Add to dashboard Cite

Passing is one of the fundamental techniques in football. Good passing skills enable players to participate in the game effectively and maintain possession of the ball. Paired short passing drills also help improve individual skills. This research aims to find out whether short passing training in pairs can increase the accuracy of soccer passing in adolescent soccer players. The quantitative method used in the research is a pre-experimental design through one group pretest and posttest. The research instrument used a ball passing accuracy test for 30 seconds and used the short passing practice method in pairs. The population in this study consisted of 20 male Rekminers Club soccer players aged 15-17 years. The sample was selected using a total sampling technique, and the entire population sampled was 20 players. The data analysis technique uses the t-test to test the hypothesis to increase soccer passing accuracy in the pretest and posttest treatment using SPSS version 26. Based on the results of hypothesis testing using the t-test in paired short passing exercises with football passing accuracy ability, there is a significant influence based on pretest and posttest data analysis of 0.000 < 0.05. This exercise was successfully carried out to improve the accuracy of football passing. The research results show that short passing training in pairs is an effective method for improving the passing accuracy of football players.

show abstract

“…XAI is a branch of artificial intelligence dedicated to developing models that can offer clear and interpretable explanations for their predictions and decisions [15,16]. Among the most widely used techniques in this field, SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) stand out for human performance modelling; however, the latter only present local (individual) explanations [17]. SHAP values are based on cooperative game theory and offer a measure of feature importance with both local and global accuracy [15,17].…”

Section: Introductionmentioning

confidence: 99%

“…Among the most widely used techniques in this field, SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) stand out for human performance modelling; however, the latter only present local (individual) explanations [17]. SHAP values are based on cooperative game theory and offer a measure of feature importance with both local and global accuracy [15,17]. Other XAI approaches can be found in the literature, with their usefulness dependent on the objective and type of AI used [18].…”

Section: Introductionmentioning

confidence: 99%

“…XAI with SHAP values permits more accurate and interpretable predictive models (presenting lower error than traditional statistical methods), with better generalisation and representation of non-linear system behaviours like human athletic performance [6,11,[19][20][21]. This approach has begun to be applied in team-sports-related research [17,22] but to our knowledge only one technical study has applied the SHAP interpretation method to swimming performance prediction, specifically backstroke-to-breaststroke turn performance by age group [2], while another study used SHAP to predict teamwork effects on individual effort expenditure in swimming relays [21].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Swimming Performance Interpreted through Explainable Artificial Intelligence (XAI)—Practical Tests and Training Variables Modelling

Carvalho,

Goethel,

Silva

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Explainable artificial intelligence (XAI) models with Shapley additive explanation (SHAP) values allows multidimensional representation of movement performance interpreted on both global and local levels in terms understandable to human intuition. We aimed to evaluate the swimming performance (World Aquatics points) predictability of a combination of demographic, training, anthropometric, and biomechanical variables (inputs) through XAI. Forty-seven swimmers (16 males), after completing a training questionnaire (background and duration) and anthropometric assessment, performed, in a randomised order, a 25 m front crawl and three countermovement jumps, at maximal intensity. The predicted World Aquatics points (516 ± 159; mean ± SD) were highly correlated (r2 = 0.93) with the 529 ± 158 actual values. The duration of swimming training was the most important variable (95_SHAP), followed by the countermovement jump impulse (37_SHAP), both with a positive effect on performance. In contrast, a higher percentage of fat mass (21_SHAP) corresponded to lower World Aquatics points. Impulse, when interpreted together with dryland training duration and stroke rate, shows the positive effects of upper and lower limb power on swimming performance. Height should be interpreted together with arm span when exploring positive effects of anthropometric traits on swimming performance. The XAI modelling highlights the usefulness of specific training, technical and physical testing, and anthropometric factors for monitoring swimmers.

show abstract

Predicting Football Team Performance with Explainable AI: Leveraging SHAP to Identify Key Team-Level Performance Metrics

Cited by 7 publications

References 63 publications

Tactical Situations and Playing Styles as Key Performance Indicators in Soccer

Tactical Situations and Playing Styles as Key Performance Indicators in Soccer

Improved passing accuracy by using pair practice in adolescent soccer players

Swimming Performance Interpreted through Explainable Artificial Intelligence (XAI)—Practical Tests and Training Variables Modelling

Contact Info

Product

Resources

About