Premier League Season Research Articles

Profiting from an inefficient association football gambling market: Prediction, risk and uncertainty using Bayesian networks

We present a Bayesian network (BN) model for forecasting Association Football match outcomes. Both objective and subjective information are considered for prediction, and we demonstrate how probabilities transform at each level of model component, whereby predictive distributions follow hierarchical levels of Bayesian inference. The model was used to generate forecasts for each match of the 2011/2012 English Premier League (EPL) season, and forecasts were published online prior to the start of each match. Profitability, risk and uncertainty are evaluated by considering various unit-based betting procedures against published market odds. Compared to a previously published successful BN model, the model presented in this paper is less complex and is able to generate even more profitable returns.

Determining the level of ability of football teams by dynamic ratings based on the relative discrepancies in scores between adversaries

AbstractA rating system provides relative measures of superiority between adversaries. We propose a novel and simple approach, which we call pi-rating, for dynamically rating Association Football teams solely on the basis of the relative discrepancies in scores through relevant match instances. The pi-rating system is applicable to any other sport where the score is considered as a good indicator for prediction purposes, as well as determining the relative performances between adversaries. In an attempt to examine how well the ratings capture a team’s performance, we have a) assessed them against two recently proposed football ELO rating variants and b) used them as the basis of a football betting strategy against published market odds. The results show that the pi-ratings outperform considerably the widely accepted ELO ratings and, perhaps more importantly, demonstrate profitability over a period of five English Premier League seasons (2007/2008–2011/2012), even allowing for the bookmakers’ built-in profit margin. This is the first academic study to demonstrate profitability against market odds using such a relatively simple technique, and the resulting pi-ratings can be incorporated as parameters into other more sophisticated models in an attempt to further enhance forecasting capability.

A COMPARISON OF SIMULTANEOUS KELLY BETTING STRATEGIES

In this paper we consider the problem of Kelly betting on simultaneous games, and the relative performance of betting strategies that use multibets compared to those that do not. We develop a simulation model based on the Dirichlet distribution to test the performance of three Kelly betting strategies using the empirical odds distribution from the 2007-08 English Premier League Season. This model allows us to control for the size of a bettor’s edge and the noise of this edge. The simulation results suggest that the Kelly using multibets of all levels outperforms the portfolio optimisation approach of betting on single game outcomes only. We also provide a discussion regarding the practicalities of implementing the strategies.

Factors Associated with Goals and Goal Scoring Opportunities in Professional Soccer

The aim of the present study was to investigate the factors associated with goals and goal scoring opportunities in soccer taking into consideration a broader range of performance indicators. The study was part exploratory but, based on the current literature, we also developed some a priori predictions. In particular, it was predicted that (a) most goals would be scored within the penalty area (> 70%); (b) approximately 30% of goals would be scored from set plays and (c) the majority (> 70%) of goals would be scored from a relatively short (< 4) passing interchange. Data was collected from 1788 attempts and 169 goals for an English FA Premier League season. The Web-soft snapper performance analysis tool was used to time code when attempts on goal were made and the associated behaviours relating to the attempt on goal. All a priori predictions were supported. The binary logistic regression identified 3 covariates which had a significant (P < 0.05) impact on goals scored. This included position of attempt; goal keepers’ positions and type of shoot. Transitions in play accounted for 63% of all goals scored and well over half of all attempts on goal. Although similarities were evident between this and previous literature, this investigation also highlighted the importance of other key variables associated with goals and goal scoring opportunities. The high contribution of factors associated with transitions in play helped to uncover the importance of tracking goals and goal scoring opportunities back to their point of origin.

Intensities of exercise during match-play in FA Premier League referees and players

The aim of the present study was to examine the relationship between intensities of exercise during match-play of elite-standard soccer referees with those of the players from the same match. Match analysis data were collected (Prozone® Leeds, UK) for 18 elite-standard soccer referees (age 26–49 years) on FA Premier League matches during the 2008/09 English FA Premier League season (236 observations). Running categories for referees and players were as follows: total distance covered (m); high-speed running distance (speed >19.8 km · h−1); and sprinting distance (speed >25.2 km · h−1). Analysis of the distance–time regression coefficients revealed no differences between the referees' and players' within-match rates of change for total distance covered (−0.594 ± 0.394 vs. −0.713 ± 0.269 m · min−1; P = 0.104), high-speed running (−0.039 ± 0.077 vs. −0.059 ± 0.030 m · min−1; P = 0.199), and sprinting (−0.003 ± 0.039 vs. −0.021 ± 0.017 m · min−1; P = 0.114). In addition, there were no differences between across-season rates of change for total distance (–26.756 ± 40.434 vs. −20.031 ± 25.502 m per match day; P = 0.439) and sprinting (–9.662 ± 7.564 vs. −8.589 ± 4.351 m per match day; P = 0.542). These results show that elite-standard soccer referees' intensities of exercise during match-play are interrelated with those of the players and thus demonstrate that referees are able to keep pace with the players during FA Premier League matches.

THE RELATIONSHIP BETWEEN PRESEASON RANGE OF MOTION AND MUSCLE STRAIN INJURY IN ELITE SOCCER PLAYERS

The purpose of the study was to determine the influence of preseason lower-extremity range of motion (ROM) on the risk of muscle strain injury during a competitive season for elite soccer players. Thirty-six elite male soccer players (age, 25.6 = 4.7 years) had maximum static ROM for 6 movements of the lower extremity measured prior to the 2003–2004 English Premier League season. Player age, body size, dominant limb, and playing position were documented also. All lower-extremity muscle strain injuries were recorded prospectively during the competitive season, as was the total amount of time spent in training and games for each player. Soccer players sustaining a muscle strain injury in the hip flexors or knee flexors had lower pre- season ROM (p ≤ 0.05) in these muscle groups compared with uninjured players. Similar trends were observed for the remaining muscle groups, but all failed to reach statistical significance (p ≥ 0.05). Most significantly, soccer players with lower preseason ROM in the hip flexors and knee flexors had a statistically higher risk for a muscle strain injury to these muscle groups during a competitive season. Age, body size, limb dominance, and playing position were not significant intrinsic risk factors for the development of muscle strain injuries. Screening of flexibility for soccer players should be conducted during preseason, and flexibility training should be prescribed to players with reduced ROM to lower the risk of developing a muscle strain injury.

The Relationship Between Preseason Range of Motion and Muscle Strain Injury in Elite Soccer Players

The purpose of the study was to determine the influence of preseason lower-extremity range of motion (ROM) on the risk of muscle strain injury during a competitive season for elite soccer players. Thirty-six elite male soccer players (age, 25.6 +/- 4.7 years) had maximum static ROM for 6 movements of the lower extremity measured prior to the 2003-2004 English Premier League season. Player age, body size, dominant limb, and playing position were documented also. All lower-extremity muscle strain injuries were recorded prospectively during the competitive season, as was the total amount of time spent in training and games for each player. Soccer players sustaining a muscle strain injury in the hip flexors or knee flexors had lower preseason ROM (p < 0.05) in these muscle groups compared with uninjured players. Similar trends were observed for the remaining muscle groups, but all failed to reach statistical significance (p > 0.05). Most significantly, soccer players with lower preseason ROM in the hip flexors and knee flexors had a statistically higher risk for a muscle strain injury to these muscle groups during a competitive season. Age, body size, limb dominance, and playing position were not significant intrinsic risk factors for the development of muscle strain injuries. Screening of flexibility for soccer players should be conducted during preseason, and flexibility training should be prescribed to players with reduced ROM to lower the risk of developing a muscle strain injury.