Stochastic Modelling/Game Theory Analysis of Scoreline


Prediction of a football match result arouses interest from different points of view; for different people, Hence the need for this work which aims at analysing the scores of the four top English clubs to enable prediction of future outcome of matches to be made in a more scientific manner. From the analysis of the scoreline of the top four EPL clubs; Manchester United (M.U), Chelsea (C), Arsenal (A), and Manchester City (M.C) from (2002-2015) using Game theory and Stochastic modelling, Chelsea emerged the best team with a selection probability of  0.41  while Manchester United also emerged second best with a selection probability of 0.37. From the steady state transition probability matrix, for all the six possible pairs of the four clubs shows that the probability of M.U wining C at home is 0.44 while C wins M.U at home with probability  0.67 depicting C as the stronger club. Similarly M.U is stronger than A, with a 0.71 winning probability as against 0.25 winning probability for A, while M.U and M.C appears to be equally matched with 0.48 and 0.49 probability of winning. C against A reveals a probability of 0.58 and 0.25 for A vs C. while C vs M.C showed C to have an upper hand with a 0.71 probability of winning and 0.44 for M.C vs C. Finally A vs M.C gives the two teams 0.53 and 0.42 winning probabilities. Thus, the two most viable clubs out of the four clubs are Manchester United and Chelsea. Using the four step TPM we also predicted the 2015/2016 matches to obtain their various probabilities given the previous game.

Keywords: English Premier League, Football, Game theory, Operations Research, Prediction, Stochastic modelling

Article Review Status: Published

Pages: 20-34 (Download PDF)

Creative Commons Licence
This work by European American Journals is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License