EnglishSTN Journal of Sports Modelling and Trading – Volume 1(1) January-June 2016
Welcome to advanced online publication of Volume 1 Part 1 of the very first edition of the new Sports Trading Network Journal of Sport Modelling and Trading. In time we will be developing this content into a more formalised academic journal with DOI numbers, but wanted to make this content accessible here as soon as possible.
This very first edition contains a series of three very original articles by Georgi Boshnakov, Tarak Kharrat and Ian McHale that refers to the use of Weibull Count models for predictive models in football. Also included is a very current piece on the decline of home advantage in the English Premier League in 2015/16 and also an extensive account of the fraud detection system used at Sportradar.
Are Goals Poisson Distributed?
Georgi Boshnakov, Tarak Kharrat, Ian McHale
In this article, we provide our own slant on George Box’s famous quote (“All models are wrong but some are useful.”, 1976) and show how we try to solve the issue… Continue reading the full article
Using the Weibull count distribution for predicting the results of football matches?
Georgi Boshnakov, Tarak Kharrat, Ian McHale
In this article, we ‘up-the-ante’ and use the Weibull count distribution as the basis for a model that we use for betting… Continue reading the full article
Modelling the Time to the First Goal in Football
Georgi Boshnakov, Tarak Kharrat, Ian McHale
In this article we look at what is the most important event in determining the outcome of a football match… Continue reading the full article
The Decline in Home Advantage in the English Premier league in 2015/16: Is it Random Fluctuation or A Systematic Change?
Dr Alun Owen
In this article, we discuss the dramatic decline of the ‘Home Advantage’ specifically in the Premier League… Continue reading the full article
AN EVALUATION OF SPORTRADAR’S FRAUD DETECTION SYSTEM
Professor David Forrest from the University of Liverpool, and Professor Ian McHale from the University of Salford, were commissioned in 2015 to evaluate the processes used by Sportradar to monitor football betting markets for “unusual” activity. Their overall conclusions suggest that matches reported as suspicious by Sportradar’s Fraud Detection System are very likely to have indeed been manipulated. UEFA and Sportradar have therefore now decided to make the report public in order to publicise the results as widely as possible, and the STN Journal is delighted to have been asked to disseminate this work.
Continue reading the full article