Journal of Sports Analytics

In order to find the determinants of non-penalty goals scored per match, in association football (soccer), this paper developed a regression model consisting of 8 explanatory variables, based on observations for 98 teams playing in the top tiers of club football in England, Spain, Germany, France and Italy. We started with a framework that considered twenty-one different pitch actions that included both technical and tactical variables. Using data for the 2015-16 football season we narrowed down to the 8 variable model. The paper used a log-linear regression model in order to remove heteroscedasticity. The model estimated the number of non-penalty goals per game with error of less than |0.33| for 93 teams out of 98. For 52 teams the margin of error was less than |0.1|. Shots from penalty box per game, share of shots from goal box in total shots and long pass accuracy were found to have statistically significant positive impact on non-penalty goals scored per game. Share of long passes in total passes and crosses per game have significant negative impact.

In recent years, data analytics has been a growing phenomenon in many fields, including sport, and there has been an increased focus on how technology will impact the work of coaching and performance professionals. This paper provides a reflection of the use of data analytics within the Gaelic Athletic Association’s (GAA) football coaching practice in Ireland, and evaluates for coaches how to enhance kick-out distribution and effectiveness. Specifically, the study aimed to dissect and analyse kick-out strategies to assess from a coaching perspective, the impact on distribution and effectiveness. The research is a result of the modern day professional approach taken by most GAA Inter County teams, and intends to fill the research gap of performance analysis within the context of Gaelic Football. Video analysis of eight Men’s Inter-County Ulster Senior Gaelic Football Championship matches was completed using video analysis software. The key parameter of investigation was kick-outs, with distance and direction being the variables specifically analysed. Three factors were established to measure kick-out effectiveness: retaining possession, attack building, and score/scoring opportunity. A questionnaire was devised for data collection with 75 players and 25 coaches to gather opinions and add further depth to the statistical evidence. Kick-outs to the wings (57.6%) and 65m+ (53.8%) were the most common distribution strategy employed by goalkeepers. Kick-outs to 21–45 m produced notably greater success rates of possession retention (87%) and attack building (55%) than the longer kick-outs. Kick-outs to the wings contributed an 8% greater success rate of creating scoring opportunities as opposed to central kick-outs. Kick-outs to the centre were the only parameter to result in a negative net difference of scores between both teams (–1%). It was established that kick-outs can be successful attacking elements when used efficiently. Short kick-outs to the wings appear to be the most effective strategy to improve the likelihood of possession retention, attack building and creation of scoring opportunities. An implication of this study for sports science is to make data analytics more accessible to GAA coaches, and address the challenges faced by coaches in this area by translating knowledge of performance analysis of kick-outs into working practice on the field.

In the last few decades, the sports world has seen the precipitous rise of data-driven player analysis methods across many professional sports. In light of these advents to the field, the current paper offers a novel application of a multi-criteria classification scheme developed within the scholarly literature, that being: grey relational analysis (GRA), to the milieu of professional sports. Specifically, this technique is utilized in the context of the National Basketball Association (NBA) to rank regular seasons during the Modern Era (i.e., since the 1979-80 season, after the merger between the NBA and ABA) through the use of a constellation of statistics. A sample of 100 seasons initially classified by player efficiency rating (PER) were examined using GRA. Findings from the present study illustrate that Stephen Curry’s recent MVP campaign during the 2015-16 season for the Golden State Warriors was the top ranked overall regular season out of those sampled in the Modern Era. The use of GRA is compared to current popular indicators for player evaluation and efficiency. Future directions and implications for professional sports are discussed.

Each year, 68 teams are invited to participate in the NCAA basketball tournament. Of these 68 teams, 36 are given an at-large bid by a selection committee rather than automatically qualifying. However, despite the process being public, the exact method by which teams are ranked against other teams to determine which teams are given at-large bids is not known. Many factors are known or theorized to be in use, including the RPI, top 50 wins, and strength of schedule. In this paper we attempt to provide a quantitative method to mimic the team made by the selection committee by constructing a decision tree. We then test our method on years 2012–2016, where we find that in each year all but one team selected by our method was invited to the tournament.

As one-day international (ODI) games rise in popularity, it is important to understand the possible predictors that affect the game outcome. The home-field advantage, coin-toss result, bat-first or second, and day vs day-night game format are such popular variables being considered in the cricket literature. This article focuses on a comprehensive study of quantifying the significance of those important predictors via graphical ‘classification and regression tree’ (CART) and the popular logistic regression approaches. This study reveals the importance of the home-field advantage for major cricket playing nations in one-day international games but questions the uniformity of such factors under different playing conditions. Importantly, the home-field advantage is investigated further based on the opponent’s geographical location. Conclusively, the CART approach provides interesting and novel interpretations for popular predictors in ODI games.

Since the introduction of PITCHf/x in 2006, there has been a plethora of data available for anyone who wants to access to the minute details of every baseball pitch thrown over the past nine seasons. Everything from the initial velocity and release point to the break angle and strike zone placement is tracked, recorded, and used to classify the pitch according to an algorithm developed by MLB Advanced Media (MLBAM). Given these classifications, we developed a model that would predict the next type of pitch thrown by a given pitcher, using only data that would be available before he even stepped to the mound. We used data from three recent MLB seasons (2013-2015) to compare individual pitcher predictions based on multi-class linear discriminant analysis, support vector machines, and classification trees to lead to the development of a real-time, live-game predictor. Using training data from the 2013, 2014, and part of the 2015 season, our best method achieved a mean out-of-sample predictive accuracy of 66.62%, and a real-time success rate of over 60%.