Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model

Magno Xavier de Arajo; Anna Karla Barros da Trindade; Francisco de Paula Santos de Araujo Junior

doi:doi:10.11648/j.pamj.20221103.11

| Peer-Reviewed

Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model

Magno Xavier de Arajo, Anna Karla Barros da Trindade, Francisco de Paula Santos de Araujo Junior

Published in Pure and Applied Mathematics Journal (Volume 11, Issue 3)

Received: 26 April 2022 Accepted: 14 May 2022 Published: 27 June 2022

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

The present work aims to approach the Holgate bivariate Poisson model in its application to soccer data, considering the importance that soccer has worldwide and as a very popular sport in Brazil. In a football match, two teams play for 90 minutes, plus extra time defined by the match referee according to the organizations' own criteria, with the objective of scoring as many goals as their opponent. It can be seen in some studies the statement that the final result of a soccer match is a bivariate random vector whose number of goals scored by each team can be represented by its parameters. In view of the above, from Holgate's bivariate Poisson model, the attack and defense capacity of each team will be estimated, based on the average of goals scored and conceded by the home team playing at home, and the visiting team playing away from home. Such a model allows the analysis of prediction of the probability of victory, draw and defeat of the teams in a certain future match, with defined parameters and a model that takes into account the goals scored and suffered by the teams until the previous round to the analysis of the model. With this, the focus is on the application of the model, to improve and perfect the ability to predict results, where the four initial rounds of the championship serve to define the initial parameters for the model. It is worth noting that the methodology presented will be applied with data from the 2018 Campeonato Brasileiro de Futebol da Série A, which is the main national championship in Brazil and involves the top twenty teams based on their rankings in the previous year's championship.

Published in	Pure and Applied Mathematics Journal (Volume 11, Issue 3)
DOI	10.11648/j.pamj.20221103.11
Page(s)	39-46
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2022. Published by Science Publishing Group

Keywords

Football, Forecast, Poisson Distribution, Simulation

References

[1]	ARAUJO, Clodoaldo Tsuyoshi Pereira de; TAVARES, Leandro; ALVARES, Luis Gus- tavo; LOUZADA NETO, Francisco; SUZUKI, Adriano Kamimura. MODELAGEM ES- TATÍSTICA PARA A PREVISA˜ O DE JOGOS DE FUTEBOL: UMA APLICAÇÃO NO CAMPEONATO BRASILEIRO DE FUTEBOL 2014. Revista da Estatística UFOP, Vol IV, 2015, ISSN 2237-8111.
[2]	ARRUDA, Marcelo Leme de. Poisson, Bayes, Futebol e DeFinetti. Dissertação (Mestrado em Estatística) - Instituto de Matemática e Estatística da USP, São Paulo, 2000.
[3]	BARBETTA, Pedro Alberto. Estatística para cursos de engenharia e informática. Atlas, São Paulo, 2004.
[4]	BARREIRA, Júlia. Vantagem de jogar em casa no futebol feminino: uma análise dos três principais campeonatos no Brasil. Revista Brasileira de Ciência e Movimento. 2018; 26 (3): 83-87.
[5]	CEOLATO JUNIOR, Carlos Alberto. CHUTEGOL - CHance de Um Time Em GOLs: Um sistema para progno´sticos de jogos de futebol. Dissertação (Bacharelado em Sistemas de Informação) - Universidade Federal de Santa Catarina, 2007.
[6]	CLARKE, S.; NORMAN, J. M. Home ground advantage of individual clubs in English soccer. The Statistician, London, v. 44, n. 4, p. 509-21, 1995.
[7]	AMORIM, A. M. do N. TRINDADE, A. K. B AND ARAUJO JUNIOR, F. P. S. Cadeias de Markov: Uma introdução apresentável ao ensino médio com aplicação ao soneto “Amor é fogo que arde sem se ver”. Revista Eletrônica da Sociedade Brasileira de Matemática PMO. v. 8, n. 2, 2020: 236-248.
[8]	GRIFFITHS, R. C., MILNE, R. K & WOOD, R. (1979), Aspects of Correlation in Bivariate Poisson Distributions and Processes. Australian Journal of Statistics 21.
[9]	KUK, A. Y. C. Modelling paired comparison data with large numbers of draws and large variability of draw percentages among players. The Statistician, 44, 523 – 228, (1995).
[10]	POLLARD, R. Home advantage in soccer: a retrospective analysis. Journal of Sports Sciences, London, v. 4, n. 3, p. 237-48, 1986.
[11]	SILVA, Cristiano Diniz da; MEDEIROS, N´ısio Cunha; SILVA, Ana Cristina Diniz da. Vantagem em casa no campeonato brasileiro de futebol: efeito do local do jogo e da qualidade dos times. Revista Brasileira de Cineantropometria e Desempenho Humano. 2010; 12 (2): 148-154.
[12]	SILVA, Wesley Bertoli da. DISTRIBUIÇÃO DE POISSON BIVARIADA APLICADA À PREVISÃO DE RESULTADOS ESPORTIVOS. Dissertação (Mestrado em Estatística) - Universidade Federal de São Carlos - DEs/UFSCar, São Carlos, abril de 2014.
[13]	SUZUKI, Adriano Kamimura. Modelagem Estatística para a Determinação de Resultados de Dados Esportivos. Dissertação (Mestrado em Estatística) - Universidade Federal de São Carlos - DEs/UFSCar, São Carlos, junho de 2007.

Cite This Article

Plain Text BibTeX RIS

APA Style

Magno Xavier de Arajo, Anna Karla Barros da Trindade, Francisco de Paula Santos de Araujo Junior. (2022). Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure and Applied Mathematics Journal, 11(3), 39-46. https://doi.org/10.11648/j.pamj.20221103.11

Copy | Download

ACS Style

Magno Xavier de Arajo; Anna Karla Barros da Trindade; Francisco de Paula Santos de Araujo Junior. Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure Appl. Math. J. 2022, 11(3), 39-46. doi: 10.11648/j.pamj.20221103.11

Copy | Download

AMA Style

Magno Xavier de Arajo, Anna Karla Barros da Trindade, Francisco de Paula Santos de Araujo Junior. Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure Appl Math J. 2022;11(3):39-46. doi: 10.11648/j.pamj.20221103.11

Copy | Download

@article{10.11648/j.pamj.20221103.11,
  author = {Magno Xavier de Arajo and Anna Karla Barros da Trindade and Francisco de Paula Santos de Araujo Junior},
  title = {Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model},
  journal = {Pure and Applied Mathematics Journal},
  volume = {11},
  number = {3},
  pages = {39-46},
  doi = {10.11648/j.pamj.20221103.11},
  url = {https://doi.org/10.11648/j.pamj.20221103.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.pamj.20221103.11},
  abstract = {The present work aims to approach the Holgate bivariate Poisson model in its application to soccer data, considering the importance that soccer has worldwide and as a very popular sport in Brazil. In a football match, two teams play for 90 minutes, plus extra time defined by the match referee according to the organizations' own criteria, with the objective of scoring as many goals as their opponent. It can be seen in some studies the statement that the final result of a soccer match is a bivariate random vector whose number of goals scored by each team can be represented by its parameters. In view of the above, from Holgate's bivariate Poisson model, the attack and defense capacity of each team will be estimated, based on the average of goals scored and conceded by the home team playing at home, and the visiting team playing away from home. Such a model allows the analysis of prediction of the probability of victory, draw and defeat of the teams in a certain future match, with defined parameters and a model that takes into account the goals scored and suffered by the teams until the previous round to the analysis of the model. With this, the focus is on the application of the model, to improve and perfect the ability to predict results, where the four initial rounds of the championship serve to define the initial parameters for the model. It is worth noting that the methodology presented will be applied with data from the 2018 Campeonato Brasileiro de Futebol da Série A, which is the main national championship in Brazil and involves the top twenty teams based on their rankings in the previous year's championship.},
 year = {2022}
}

Copy | Download

TY  - JOUR
T1  - Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model
AU  - Magno Xavier de Arajo
AU  - Anna Karla Barros da Trindade
AU  - Francisco de Paula Santos de Araujo Junior
Y1  - 2022/06/27
PY  - 2022
N1  - https://doi.org/10.11648/j.pamj.20221103.11
DO  - 10.11648/j.pamj.20221103.11
T2  - Pure and Applied Mathematics Journal
JF  - Pure and Applied Mathematics Journal
JO  - Pure and Applied Mathematics Journal
SP  - 39
EP  - 46
PB  - Science Publishing Group
SN  - 2326-9812
UR  - https://doi.org/10.11648/j.pamj.20221103.11
AB  - The present work aims to approach the Holgate bivariate Poisson model in its application to soccer data, considering the importance that soccer has worldwide and as a very popular sport in Brazil. In a football match, two teams play for 90 minutes, plus extra time defined by the match referee according to the organizations' own criteria, with the objective of scoring as many goals as their opponent. It can be seen in some studies the statement that the final result of a soccer match is a bivariate random vector whose number of goals scored by each team can be represented by its parameters. In view of the above, from Holgate's bivariate Poisson model, the attack and defense capacity of each team will be estimated, based on the average of goals scored and conceded by the home team playing at home, and the visiting team playing away from home. Such a model allows the analysis of prediction of the probability of victory, draw and defeat of the teams in a certain future match, with defined parameters and a model that takes into account the goals scored and suffered by the teams until the previous round to the analysis of the model. With this, the focus is on the application of the model, to improve and perfect the ability to predict results, where the four initial rounds of the championship serve to define the initial parameters for the model. It is worth noting that the methodology presented will be applied with data from the 2018 Campeonato Brasileiro de Futebol da Série A, which is the main national championship in Brazil and involves the top twenty teams based on their rankings in the previous year's championship.
VL  - 11
IS  - 3
ER  -

Copy | Download

Author Information

Magno Xavier de Arajo

Mathematics Department, Ceará Regional College, Camocim, Brazil
Anna Karla Barros da Trindade

Mathematics Department, Federal Institute of Piauí, Corrente, Brazil
Francisco de Paula Santos de Araujo Junior

Education Department, Federal University of Piauí, Teresina, Brazil

Download PDF

Sections

Plain Text BibTeX RIS

APA Style

Magno Xavier de Arajo, Anna Karla Barros da Trindade, Francisco de Paula Santos de Araujo Junior. (2022). Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure and Applied Mathematics Journal, 11(3), 39-46. https://doi.org/10.11648/j.pamj.20221103.11

Copy | Download

ACS Style

Magno Xavier de Arajo; Anna Karla Barros da Trindade; Francisco de Paula Santos de Araujo Junior. Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure Appl. Math. J. 2022, 11(3), 39-46. doi: 10.11648/j.pamj.20221103.11

Copy | Download

AMA Style

Magno Xavier de Arajo, Anna Karla Barros da Trindade, Francisco de Paula Santos de Araujo Junior. Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model. Pure Appl Math J. 2022;11(3):39-46. doi: 10.11648/j.pamj.20221103.11

Copy | Download

@article{10.11648/j.pamj.20221103.11,
  author = {Magno Xavier de Arajo and Anna Karla Barros da Trindade and Francisco de Paula Santos de Araujo Junior},
  title = {Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model},
  journal = {Pure and Applied Mathematics Journal},
  volume = {11},
  number = {3},
  pages = {39-46},
  doi = {10.11648/j.pamj.20221103.11},
  url = {https://doi.org/10.11648/j.pamj.20221103.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.pamj.20221103.11},
  abstract = {The present work aims to approach the Holgate bivariate Poisson model in its application to soccer data, considering the importance that soccer has worldwide and as a very popular sport in Brazil. In a football match, two teams play for 90 minutes, plus extra time defined by the match referee according to the organizations' own criteria, with the objective of scoring as many goals as their opponent. It can be seen in some studies the statement that the final result of a soccer match is a bivariate random vector whose number of goals scored by each team can be represented by its parameters. In view of the above, from Holgate's bivariate Poisson model, the attack and defense capacity of each team will be estimated, based on the average of goals scored and conceded by the home team playing at home, and the visiting team playing away from home. Such a model allows the analysis of prediction of the probability of victory, draw and defeat of the teams in a certain future match, with defined parameters and a model that takes into account the goals scored and suffered by the teams until the previous round to the analysis of the model. With this, the focus is on the application of the model, to improve and perfect the ability to predict results, where the four initial rounds of the championship serve to define the initial parameters for the model. It is worth noting that the methodology presented will be applied with data from the 2018 Campeonato Brasileiro de Futebol da Série A, which is the main national championship in Brazil and involves the top twenty teams based on their rankings in the previous year's championship.},
 year = {2022}
}

Copy | Download

TY  - JOUR
T1  - Statistical Modeling for Sports Prediction - an Application in the 2018 Brazilian Football Championship with the Bivariate Poisson Model
AU  - Magno Xavier de Arajo
AU  - Anna Karla Barros da Trindade
AU  - Francisco de Paula Santos de Araujo Junior
Y1  - 2022/06/27
PY  - 2022
N1  - https://doi.org/10.11648/j.pamj.20221103.11
DO  - 10.11648/j.pamj.20221103.11
T2  - Pure and Applied Mathematics Journal
JF  - Pure and Applied Mathematics Journal
JO  - Pure and Applied Mathematics Journal
SP  - 39
EP  - 46
PB  - Science Publishing Group
SN  - 2326-9812
UR  - https://doi.org/10.11648/j.pamj.20221103.11
AB  - The present work aims to approach the Holgate bivariate Poisson model in its application to soccer data, considering the importance that soccer has worldwide and as a very popular sport in Brazil. In a football match, two teams play for 90 minutes, plus extra time defined by the match referee according to the organizations' own criteria, with the objective of scoring as many goals as their opponent. It can be seen in some studies the statement that the final result of a soccer match is a bivariate random vector whose number of goals scored by each team can be represented by its parameters. In view of the above, from Holgate's bivariate Poisson model, the attack and defense capacity of each team will be estimated, based on the average of goals scored and conceded by the home team playing at home, and the visiting team playing away from home. Such a model allows the analysis of prediction of the probability of victory, draw and defeat of the teams in a certain future match, with defined parameters and a model that takes into account the goals scored and suffered by the teams until the previous round to the analysis of the model. With this, the focus is on the application of the model, to improve and perfect the ability to predict results, where the four initial rounds of the championship serve to define the initial parameters for the model. It is worth noting that the methodology presented will be applied with data from the 2018 Campeonato Brasileiro de Futebol da Série A, which is the main national championship in Brazil and involves the top twenty teams based on their rankings in the previous year's championship.
VL  - 11
IS  - 3
ER  -

Copy | Download