Journal of Quantitative Analysis in Sports The Curse of Scoreless Draws in Soccer : The Relationship with a Team ' s Offensive , Defensive , and Overall Performance

In soccer, a scoreless draw is typically considered as an unwanted result that often jeopardizes the spectacle value of the game. In the present work, we tried to investigate a) how common scoreless draws are, and b) the relationship between scoreless draws and other indices of a football team's performance using machine learning techniques. Using data from 54 competitions around the world, Bayesian networks, least squares support vector machines and Hybrid Monte Carlo multi-layer perceptrons were used to investigate which combination of indices best predicts a team's season proportion of scoreless draws and to see exactly what predictions these indices make. There was ample variability in the proportion of scoreless draws, both when comparing individual teams and countries. For individual teams, the percentage scoreless draws varied between 0 and 30%. On average, nearly 9% of all games ended in 0-0. Not surprisingly, the most important parameter appeared to be the total number of goals per game. More interestingly, the earned points per game were also linked to the proportion of scoreless draws. Games of average and bad-to-average teams more often resulted in a scoreless draw, in particular when the games of these teams saw few goals. Such a team could end up having 20% scoreless draws in one season. A suggestive result that more spectators may be associated with less scoreless draws is also presented.

[1]  Sabine Van Huffel,et al.  Prediction of mental development of preterm newborns at birth time using LS-SVM , 2002, ESANN.

[2]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[3]  Claus Dethlefsen,et al.  Learning Bayesian Networks with R , 2003 .

[4]  Vladimir Vapnik,et al.  The Nature of Statistical Learning , 1995 .

[5]  M. Wright,et al.  Determining the best strategy for changing the configuration of a football team , 2003, J. Oper. Res. Soc..

[6]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[7]  M. Wright,et al.  The professional foul in football: Tactics and deterrents , 2003, J. Oper. Res. Soc..

[8]  Ian T. Nabney,et al.  Netlab: Algorithms for Pattern Recognition , 2002 .

[9]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[10]  Johan A. K. Suykens,et al.  Knowledge discovery in a direct marketing case using least squares support vector machines , 2001, Int. J. Intell. Syst..

[11]  A. Horowitz.  A generalized guided Monte Carlo algorithm , 1991 .

[12]  J. Suykens,et al.  Preoperative diagnosis of ovarian tumors using Bayesian kernel‐based methods , 2007, Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology.

[13]  Chris Visscher,et al.  Kicks from the penalty mark in soccer: The roles of stress,skill, and fatigue for kick outcomes , 2007, Journal of sports sciences.

[14]  Richard Pollard,et al.  Measuring the effectiveness of playing strategies at soccer , 1997 .

[15]  Is soccer dying? A time series approach , 2000 .

[16]  A. W. Simonetti,et al.  The use of multivariate MR imaging intensities versus metabolic data from MR spectroscopic imaging for brain tumour classification. , 2005, Journal of magnetic resonance.

[17]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machines , 2002 .

[18]  Mike Wright,et al.  Using a Markov process model of an association football match to determine the optimal timing of substitution and tactical decisions , 2002, J. Oper. Res. Soc..

[19]  William D. Penny,et al.  An empirical evaluation of Bayesian sampling with hybrid Monte Carlo for training neural network classifiers , 1999, Neural Networks.

[20]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[21]  G. Ridder,et al.  Down to Ten: Estimating the Effect of a Red Card in Soccer , 1994 .

[22]  Johan A. K. Suykens,et al.  Electric Load Forecasting: Using Kernel-Based Modeling for Nonlinear System Identification , 2007 .

[23]  David Maxwell Chickering,et al.  Learning Bayesian Networks: The Combination of Knowledge and Statistical Data , 1994, Machine Learning.

[24]  Glen D Meeden,et al.  An Improved Award System for Soccer , 2003 .

[25]  Dimitris Rizopoulos,et al.  The logistic transform for bounded outcome scores. , 2007, Biostatistics.

[26]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[27]  N. Pochet,et al.  New models to predict depth of infiltration in endometrial carcinoma based on transvaginal sonography , 2006, Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology.

[28]  Johan A. K. Suykens,et al.  Financial time series prediction using least squares support vector machines within the evidence framework , 2001, IEEE Trans. Neural Networks.

[29]  D Timmerman,et al.  Predicting the clinical behavior of ovarian cancer from gene expression profiles , 2005, International Journal of Gynecologic Cancer.

[30]  Håvard Rue,et al.  Prediction and retrospective analysis of soccer matches in a league , 2000 .