Tie correction for dynamic modeling of soccer matches

  • Anderson Ribeiro Duarte Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil
  • Helgem de Souza Ribeiro Martins Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil
  • Bruno Fernandes da Silva Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil
Keywords: Football, Poisson truncated, Brazilian football championship, Beta generalized linear model

Abstract

Soccer is among the most difficult predictability sports, in which the occurrence of atypical results, in which inferior teams supplant the better teams becomes almost commonplace. This work presents an already existent mechanism to obtain predictability to outcome of matches result through a truncated Poisson. In addition, it presents a novel proposals for adapting the dynamic threshold to determination ties in the previous model. The simulation model used is described and also the strategies for producing the correct tie threshold and promising results are discussed for the Brazilian Serie A Soccer Championship in 2013 and 2015.

Author Biographies

Anderson Ribeiro Duarte, Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil

Departamento de Estatí­stica

Helgem de Souza Ribeiro Martins, Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil

PPESTBIO - Estatí­stica Aplicada e Biometria

Bruno Fernandes da Silva, Universidade Federal de Ouro Preto (UFOP), Ouro Preto-MG, Brasil

Departamento de Estatí­stica (graduando)

References

-Alves, A. M. e colaboradores. Logit models for the probability of winning football games. Pesquisa Operacional. Vol. 31. p. 459-465. 2011.

-Brillinger, D. An analysis of chinesesuper league partial results. Science in China Series A: Mathematics. Vol. 52. p. 1139-1156. 2009.

-Cribari-Neto, F.; Zeileis, A. Beta regression in R. Journal of Statistical Software. Articles. Vol. 34. Num. 2. p. 1-24. 2010.

-Faria, F. F. Análise e Previsão de Resultados de Partidas de Futebol. Universidade Federal do Rio de Janeiro. Tese de Doutorado. 2005.

-Ferrari, S. L. P.; Cribari-Neto, F. Beta regression for modelling rates and proportions. Journal of Applied Statistics. Vol. 31. Num. y. p. 799-815. 2004.

-Karlis, D.; Ntzoufras, I. On model soccer data. Student. Vol. 3. Num. 4. p. 229-244. 2000.

-Karlis, D.; Ntzoufras, I. Analysis of sports data by using bivariate poissonmodels. Journal of the Royal Statistical Society. Series D (The Statistician). Vol. 52. Num. 3. p. 381-393. 2003.

-Knorr-Held, L. Dynamic rating of sports teams. Journal of the Royal Statistical Society. Series D (The Statistician). Vol. 49. Num. 2. p. 261-276. 2000.

-Liu, F.; Zhang, Z. Predicting soccer league games using multinomial logistic models. Course Project. 2008.

-Macedo, P. A. P.; Silva, C. D. Prediction results in the brazilian championship 2012 series a. The Brazilian Journal of Soccer Science. Vol. 7. Num. 2. p. 35-41. 2014.

-Martins, H. S. R.; Duarte, A. R. Modelagem dinâmica de partidas de futebol. Revista da Estatística da Universidade Federal de Ouro Preto. Vol. 3. Num. 2. p. 157-169. 2014.

-Rotshtein, A. P.; Posner, M.; Rakityanskaya, A. B. Football predictions based on a fuzzy model with genetic and neural tuning. Cybernetics and Systems Analysis. Vol. 41. Num. 4. p. 619-630. 2005.

-Rue, H.; Salvesen, Ø. Prediction and retrospective analysis of soccer matches in a league. Journal of the Royal Statistical Society. Series D (The Statistician). Vol. 49. Num. 3. p. 399-418. 2000.

-Souza Jr., O. G.; Gamerman, D. Previsão de partidas de futebol usando modelos dinâmicos. In: Anais do XXXVI evento da Sociedade Brasileira de Pesquisa Operacional. SBPO. 2004. p. 650-659.

Published
2019-01-20
How to Cite
Duarte, A. R., Martins, H. de S. R., & Silva, B. F. da. (2019). Tie correction for dynamic modeling of soccer matches. RBFF - Brazilian Journal of Futsal and Football, 10(41), 774-784. Retrieved from https://www.rbff.com.br/index.php/rbff/article/view/693
Issue
Vol. 10 No. 41 (2018): Supplement 2
Section
Scientific Articles - Original

Authors who publish in this journal agree to the following terms:

  • Authors retain the copyright and grant the journal the right of first publication, with work simultaneously licensed under the Creative Commons Attribution License BY-NC which allows the sharing of the work with acknowledgment of the authorship of the work and initial publication in this journal.
  • Authors are authorized to enter into additional contracts separately for non-exclusive distribution of the version of the work published in this journal (eg, publishing in institutional repository or book chapter), with acknowledgment of authorship and initial publication in this journal.
  • Authors are allowed and encouraged to post and distribute their work online (eg, in institutional repositories or on their personal page) at any point before or during the editorial process, as this can bring about productive change as well as increase impact and impact. citation of published work (See The Effect of Free Access).