Progetto MSC

Descrizione del progetto

Il Progetto MSC (Modelli Statistici per il Calcio) nasce nel 2011, con l'obiettivo di utilizzare i dati statistici sulle azioni che si sviluppano sul campo di calcio durante una partita raccolti da Panini Digital di Brescia tramite la tecnica dello scouting. Panini Digital raccoglie 1.500 eventi a partita attraverso un software di riconoscimento vocale che permette di rilevare il nome del giocatore, il gesto tecnico effettuato, l’esatta posizione in campo e quanto tempo rimane in possesso del pallone prima di cederlo o perderlo. Utilizzando questi dati per il campionato italiano di Serie A delle stagioni dalla 2008-2009 alla 2011-2012, sono stati sviluppati e comparati alcuni modelli di data mining per misurare le correlazioni tra le azioni di gioco e i risultati delle partite. Negli anni il progetto di è sviluppato in varie direzioni, sia nell'ambito dei modelli statistici sia nella costruzione di indicatori utili a misurare le performance dei giocatori.

Pubblicazioni

• Cefis M., Carpita M. (2025). Accuracy and explainability of statistical and machine learning xG models in football. Statistics, 59(2), pp. 426-445, DOI: 10.1080/02331888.2024.2445305.

• Cefis M., Carpita M. (2025). A new xG Model for football analytics. Journal of the Operational Research Society, 76(1), 1-13, DOI: 10.1080/01605682.2024.2323669.

• Zanardelli R., Carpita M., Manisera M. (2024). Statistical models for classification by handedness of Olympic Trap shooters in digital training services and remote coaching. Computational Statistics, DOI: 10.1007/s00180-024-01552-8.

• Cefis M., Carpita M. (2024). On the CTA‑PLS test for hierarchical models: an application to the football player’s performance. Computational Statistics, DOI: 10.1007/s00180-024-01566-2. 

• Cefis M., Carpita M. (2024). The higher‐order PLS‐SEM confirmatory approach for composite indicators of football performance quality. Computational Statistics, 39(1), pp. 93-116, DOI: 10.1007/s00180-022-01295-4.

• Cefis M., Carpita M. (2023). An original application to football of PLS-SEM for the xG Model. In Chelli F.M., Ciommi M., Ingrassia S., Mariani F., Recchioni M.C. (eds). Book of Short Papers of the Conference IES 2023 Statistical Methods for Evaluation and Quality: Techniques, Technologies and Trends (T3), pp. 78-83, Edizioni BACME, ISBN 979-12-803-3369-8, DOI 10.60984/978-88-94593-36-5-IES2023.

• Carpita M., Metulini R., Van Eetvelde H. (2023). Statistics for Performance and Match Analysis in Sports - Editorial. Italian Journal of Applied Statistics, 35(1), pp. 7-9, DOI: 10.26398/IJAS.0035-001.

• Carpita M., Pasca P., Arima S., Ciavolino E. (2023). Clustering of variables methods and measurement models for soccer players’ performances. Annals of Operations Research, 325, pp. 37-56, DOI: 10.1007/s10479-023-05185-w.

• Cefis M., Carpita M. (2023). A Higher-Order PLS-SEM approach to evaluate players’ performance. In Salvati N., Perna C., Marchetti S., Chambers R. (eds), Studies in Theoretical and Applied Statistics, pp. 45-56, Springer. ebook ISBN: 978-3-031-16609-9, print ISBN: 978-3-031-16608-2.

• Cefis M., Carpita M. (2022). PLS-SEM with CCA for football goalkeeper’s performance indicators. In Lombardo R., Camminatiello I., Simonacci V. eds. (2022). IES 2022: Innovation and Society 5.0: Statistical and Economic Methodologies for Quality Assessment, Book of Short Papers of the 10th Scientific Conference of the SVQS - Statistics for the Evaluation and Quality in Services Group, pp. 288-293. PKE Press, Milano. DOI: 10.36253/978-88-5518-340-6, ISSN: 978-88-94593-35-8 (print) - ISSN: 978-88-94593-36-5 (online).

• Cefis M., Carpita M. (2021). Football analytics: a Higher-Order PLS-SEM approach to evaluate players’ performance. Perna C., Salvati N., Schirripa Spagnolo F. eds. (2021). Book of Short Papers SIS 2021, pp. 508-513, Pearson, ISBN: 9788891927361.

• Carpita M., Ciavolino E., Pasca P. (2021). Players’ role-based performance composite indicators of soccer teams: A statistical perspective. Social Indicators Research, 156(2-3), pp. 815-830, DOI: 10.1007/s11205-020-02323-w.

• Carpita M., Golia S. (2021). Discovering associations between players' performance indicators and matches' results in the European Soccer Leagues. Journal of Applied Statistics, 48(9), pp. 1696-1711, DOI: 10.1080/02664763.2020.1772210.

• Cefis M., Carpita M. (2020). Football analytics: performance analysis differentiate by role. In Antonucci L., Kostiuk Y. eds. (2020). Book of Abstracts, 3rd international conference on Data Science & Social Research, pp. 22, CIRPAS. ISBN: 978-886629-051-3.

• Carpita M., Ciavolino E, Pasca P. (2020). A comparison of MBC with CLV and PCovR methods for dimensional reduction of the soccer players’ performance attributes. In Pollice A., Salvati N., Schirripa Spagnolo F. eds. (2020). Book of Short Papers SIS 2020, 50th Meeting of the Italian Statistical Society, pp. 360-365, Pearson Ed. ISBN: 978-889191-077-6.

• Carpita M., Ciavolino E., Pasca P. (2019). Exploring the statistical structure of soccer team performance variables using the PCovR (short paper). In Carpita M., Fabbris L. (Eds.) Book of Short Papers of the ASA 2019 Conference "Statistics for Health and Well-being", CLEUP, pp. 49-52. ISBN: 978-88-5495-135-8.

• Carpita M., Ciavolino E., Pasca P. (2019). Composite indicators of the Soccer Players' Performance Indices (abstract). In Mariani P. (Ed.), Data Science & Social Research Book of Abstracts, PKE Group, p. 40. 2nd International Conference - University IULM, Milano. ISBN: 9788894312096. 

• Golia S., Carpita M. (2019). On the improvement of soccer match result predictions (abstract). In Mariani P. (Ed.), Data Science & Social Research Book of Abstracts, PKE Group, p. 72. 2nd International Conference - University IULM, Milano. ISBN: 9788894312096.

• Carpita M., Ciavolino E., Pasca P. (2019). Exploring and modelling team performances of the Kaggle European Soccer Database. Statistical Modelling, 19(1), pp. 1-29, DOI: 10.1177/1471082X18810971.

• Golia S., Carpita M. (2018). On classifiers to predict soccer match results. In Capecchi S., Di Iorio F., Simone R. (Eds.) ASMOD 2018: Proceedings of the Advanced Statistical Modelling for Ordinal Data Conference, pp. 125-132, FedOAPress, Napoli. ISSN: 978-88-6887-042-3, DOI: 10.6093/978-88-6887-042-3.

• Carpita M., Golia S. (2018). Exploring the Kaggle European Soccer database with Bayesian Networks: the case of the Italian League Serie A. In Abbruzzo A., Brentari E., Chiodi M., Piacentino D. (Eds.), Book of short Papers SIS 2018. Pearson Publ. ISBN: 9788891910233.

• Carpita M., Sandri M., Simonetto A., Zuccolotto P. (2015). Discovering the Drivers of FootballMatch Outcomes with Data Mining. QualityTechnology & Quantitative Management, 12(4), pp. 537-553. DOI: 10.1080/16843703.2015.11673436.

• Carpita M., Simonetto A., Sandri M., Zuccolotto P. (2014). Football Mining with R. InData Mining Applications with R, Zhao Y., Cen Y. (Eds.), Chapter 14, Academic Press, Elsevier, ISBN: 978-0-12-411511-8.

Workshop e convegni

• On CTA-PLS corrections applied on sports performance. A new dataset for exploring actions of Italian football Matches. IES 2025 Innovation & Society: Statistical Methods for Evaluation and Quality. 12th Scientific Meeting of the Group SIS-SVQS, Università di Padova, Bressanone, 25 - 27 Giugno 2025 (BoSP 1, BoSP 2).

• Predictive modeling of match outcomes in the big five european football leagues. XV Congreso CIED 15, Sociedad Espanola de Economia del Deporte (SEED), Universidad de Oviedo, Gijon, Spagna, 12 Giugno 2025 (locandina).

• Statistical Analysis of Action Player Contribution in Soccer. 11th MathSport International Conference 2025, Université du Luxembourg, Lussemburgo, 4 Giugno 2025 (abstract). 

• An Original Application to Football of PLS-SEM for the xG Model. IES 2023 Statistical Methods for Evaluation and Quality: Techniques, Technologies and Trends (T3). 11th Scientific Meeting of the Group SIS-SVQS, University of Chieti-Pescara “G. d’Annunzio”, Pescara, 30 Agosto - 1 Settembre 2023.

• Higher-order PLS-SEM for football analytics- ECDA 2022 European Conference on Data Analysis, Università degli Studi di Napoli “Federico II”, 15 Settembre 2022. 

• An innovative xG model for football analytics. WCPASS & IACSS 2022 13th World Congress of Performance Analysis of Sport & International Symposium on Computer Science in Sport, University of Vienna, Austria, 12 Settembre 2022.

• The higher-order PLS-SEM confirmatory approach for composite indicators applied on football. AUEB SAW 6th AUEB Sports Analytics Workshop, Atene, 27 Maggio 2022.

• Football analytics: a Higher-Order PLS-SEM approach to evaluate players’ performance. 50th Scientific Meeting of the Italian Statistical Society, Pisa 22 e 24 giugno 2021. 

• Clustering methods for dimensionality reduction of football performance: considerations and future perspectives. 4th AFU Virtual International Conference - Task in Crisis Times, Al Falah University, Dubai, online, 19 e 20 maggio 2021.

• Football analytics: performance analysis differentiate by role. Third international conference on Data Science & Social Research, Università di Foggia e CIRPAS, online, 10 e 11 dicembre 2020. 

• Classification rules for polytomous variables: an application to the soccer matches' results. DIME 2020 International Conference on Distributions and Inequalities Measures in Economics, Università degli Studi di Milano-Bicocca.

• Exploring the statistical structure of soccer team performance variables using the PCovR. ASA 2019 Conference on Statistics for Health and Well-being, Università degli Studi di Brescia.

• Composite indicators of the Soccer Players' Performance Indices. On the improvement of soccer match result predictions. 2nd International Conference on Data Science & Social Research, Universities of Milano Bicocca and IULM, Milano.

• On classifiers to predict soccer match results. ASMOD 2018: Advanced Statistical Modelling for Ordinal Data Workshop, Università di Napoli Federico II, Napoli 

• Exploring the Kaggle European Soccer Database with Bayesian Networks: The Case of the Italian League Serie A. SIS 2018 – 49th Scientific Meeting, Università degli Studi di Palermo, Palermo.

Rassegna stampa

Giornaledi Brescia 14.05.2012: Brescia, La discesa dalla A? Ingiusta lo dice l'algoritmo