PerSoccer 2014 Abstracts


Full Papers
Paper Nr: 1
Title:

Do We Need Goal Line Technology in Soccer?

Authors:

Otto Kolbinger, Daniel Linke, Daniel Link and Martin Lames

Abstract: This study examined the frequency of goal-line critical decisions and compared it to other sources of critical goal calls, for investigating the necessity of Goal Line Technology for top level soccer. 1167 games of the 1st and 2nd German Bundesliga were screened for such critical decisions, including penalty calls as well. 16.8 critical goal-line decisions were found per season, of which 76.6 % could directly be resolved by tv-review. On average, 5.0 and 2.8 cases were found respectively in the 1st and 2nd Bundesliga that could justify goal-line technology. Furthermore, just 5.0 % of all critical calls concerned goal-line decisions, while 84.3 % involve offside, representing the main reason for critical goal calls (Handing: 3.3 %; Foul: 7.3 %). Furthermore, 20.6 % of the investigated penalty calls (i.e. one per match day) were rated “doubtful”. The findings of the study show, that in terms of a cost-benefit-relation a TV-review would be more useful than Goal Line Technology.

Short Papers
Paper Nr: 2
Title:

Evaluating and Comparing Football Striker Performance Using the Generalized Linear Mixed Model

Authors:

Harry Louris and Stylianos Kampakis

Abstract: Discovering the best players is a very common problem in football, with teams investing large sums of money in order to scout new recruits. Furthermore, once new potential players have been discovered, evaluating their real monetary value becomes an issue. A statistical methodology could help in this process by offering measures by which to compare players, understand their potential and evaluate them more accurately. Our research focuses on the identification of the most effective Strikers in League 1 and League 2, since this is a potential “cheap” pool of talent for teams from the premier league.

Paper Nr: 4
Title:

The Footbonaut as an Innovative Diagnostic System - Differentiating Response Times in Soccer Players of Different Age-groups

Authors:

Christian Saal, Sven Müller, Harald Fiedler, Jan Mayer and Ralf Lanwehr

Abstract: The Footbonaut is a high-tech instrument for training and diagnostics of agility in soccer. However to our knowledge there existing no scientific studies targeting the application in diagnostics of this measuring instrument. The aim of the cross-sectional cohort study was to test if the response time results from tests in the Footbonaut is a appropriate trait to distinguish between age groups. We assume that professional soccer player had a shorter response time measured by the Footbonaut. The Results of these tests indicated significant differences between the professionals and age groups lower than U17.

Paper Nr: 8
Title:

Analysis of Small Game Areas of 4vs4 and 7vs7 in Spanish Professional Football

Authors:

O. Caro, L. Fradua, A. Zubillaga and J. Fernandez-Navarro

Abstract: The aim of this study was to analyse the reduced space game situations that appear during competition in soccer considering interactions with a small number of players involved. The Amisco system was employed to collect the dimensions of playing areas of different situations from competition in order to extrapolate the results to the appropriate training drills design in reduced space. 8,477 4v4 and 7v7 game situations were collected for the study. Width, length and individual playing area were registered from these situations. The influence of the zone of the pitch in the configuration of these spaces was also considered. The results showed that the dimensions of the areas designed for training drills are larger than those in competition. The influence of the zone of the pitch is also significant, mainly in 4v4 situations. This should be considered to adjust the training drills design depending on the intended tactical aim, directly associated with the zone of the pitch where the training drill is developed.

Paper Nr: 9
Title:

Synthetic Methods to Deal with Big Data in Soccer - The FMH Social Networks Project

Authors:

Ricardo Duarte, Sérgio Tomás and Daniel Baião

Abstract: Big data is a controversial issue in performance analysis. Social networks have been adopted as a synthetic method to uncover the web of interactions captured from long data sets. Here, we describe the research findings of two studies. First, we investigated the influence of the ball possession characteristics in the competitive success of Spanish La Liga teams. We found that competitive performance was influenced by the density and connectivity of the teams, mainly due to the way teams use their possession time to give intensity to his game. In the second study, we developed and validated a multiple context-dependent social networks method for applied performance analysis. Face and quantitative content validity were assessed using panels of subject-matter experts. Sensibility was also measured, suggesting the multiple context-dependent networks are sensitive enough to capture differences in the way players interact with each other in different game contexts.

Posters
Paper Nr: 3
Title:

Chi-Square Testing and Associated Visual Outputs Applied to Performance Analysis in British Professional Soccer

Authors:

John Fraser and Nuran Fraser

Abstract: This study explores the use of analytical methodologies as applied to performance characteristics of professional clubs in the Barclays Premier League for the 2011/ 12 season. This particular study stemmed from an initiative by Manchester City FC and Opta to offer access to a full season’s worth of data and establish an ‘analytics community’. The statistical approach taken was to apply Chi-Square testing in order to establish the significance in proportions between the sets of data provided. The project not only demonstrated the results achievable but it also highlighted the potential for a more rounded, statistically based view of the outputs of games and the value to clubs in adopting a more holistic approach to data analysis. These outputs were also translated into graphical outputs that demonstrated the strength of this approach. Over the past few years, clubs have been swamped by the volume of data available to them so a means of refining a method that club analysts can use simply and with a degree of confidence is hoped as a welcome development. Too often data is treated passively by clubs with basic (and often obvious) conclusions. The aim of this approach is to make analysis more active and meaningful that will illuminate and inform the management and coaching resource within Barclays Premier League clubs.