Volume 17 (2018): Issue 2 (December 2018)

Human pose detection systems based on state-of-the-art DNNs are about to be extended, adapted and re-trained to fit the application domain of specific sports. Therefore, plenty of noisy pose data will soon be available from videos recorded at a regular and frequent basis. This work is among the first to develop mining algorithms that can mine the expected abundance of noisy and annotation-free pose data from video recordings in individual sports. Using swimming as an example of a sport with dominant cyclic motion, we show how to determine unsupervised time-continuous cycle speeds and temporally striking poses as well as measure unsupervised cycle stability over time. The average error in cycle length estimation across all strokes is 0.43 frames at 50 fps compared to manual annotations. Additionally, we use long jump as an example of a sport with a rigid phase-based motion to present a technique to automatically partition the temporally estimated pose sequences into their respective phases with a mAP of 0.89. This enables the extraction of performance relevant, pose-based metrics currently used by national professional sports associations. Experimental results prove the effectiveness of our mining algorithms, which can also be applied to other cycle-based or phase-based types of sport.

Generalized knockout tournament seedings for an arbitrary number of participants in one match are designed. Several properties of knockout tournament seedings are investigated. Enumeration results for knockout tournament seedings with different properties are obtained. Several new generalized knockout tournaments seedings are proposed and justified by a set of properties.

This study analyzed the effect of a disproportionate amount of time between rounds of a playoff series—known as the “rest vs. rust” debate in the popular sports media—on the likelihood of winning each game of the subsequent round of the tournament. We utilized NBA Finals data from 1984 – 2018, and analyzed this phenomenon using ordered logistic regression with a categorical dependent variable representing the margin of victory of each game. In addition to several control variables, variables reflecting the difference in the time between series for the two teams were used to measure this effect. The results indicate that having additional time between rounds of the series provides a statistically significant advantage; interestingly, though, it has more of an impact on the second game of the subsequent round than it does for the first game. Teams may utilize the results of this study when deciding on how to schedule and intensify their practice sessions, by providing appropriate rest and maintaining rhythm to increase the likelihood of winning each game.

Tennis matches are hierarchies made up of sets containing games which, in turn, contain points. Traditional tennis games and tiebreakers could theoretically be infinite in length because a player needs to be at least 2 points ahead of the opponent to win. Fast4 tennis is a newer format of tennis that has been used at a number of levels of the sport including professional tennis where it has been used in Next Generation Finals events since 2017. The purpose of the current investigation is to compare the traditional tennis format to Fast4 tennis in terms of the probability of different players winning matches and the duration of matches. Probabilistic models of Fast4 tennis games and tiebreakers were developed. These models allowed the probability of winning games and tiebreakers to be compared between the two formats of tennis for a range of probabilities of players winning points. The models were then used within a series of simulations to determine the probability of winning sets and matches as well as the durations of games, tiebreakers, sets and matches in the two formats. Each component of the two formats of tennis was simulated 100,000 times revealing a reduced impact of serve, greater chance of upsets and shorter matches in Fast 4 tennis than in traditional tennis. The probability of players of differing abilities winning matches as well as the duration of tennis matches should be considered by those making decisions on the format of matches to be applied in tennis tournaments as well as by those preparing to compete in such tournaments.

The main objective of this article was to analyse whether the Interactive Video Test (IVT) is a useful tool for the practical off-field training of soccer assistant referees, and if its use could reduce erroneous on-field decisions when applying Law 11 of the Laws of the Game. Assistant referees were taken from the Spanish 2nd “B” and 3rd Divisions, and were divided into two groups, the Experimental Group (EG) and the Control Group (CG). The referees in the EG were trained with the Interactive Video Test by analysing 720 off-side decisions. Subsequently, both groups were assessed in on-field tests involving the simulation of 326 possible off-side situations. When the results of both groups were compared, there was a continuous improvement over time in the EG associated to the use of the IVT, significantly better than the improvement of the CG. Moreover, the IVT proved to be a good diagnostic tool to assess the skills of assistant referees in perceiving and evaluating off-side situations.

A geometric simplification of a recent proposal for a new geometry of the penalty area for football (soccer) is presented. A simplified line is necessary because the fully mathematical curve previously proposed can be difficult to implement in the real world of football, which has thousands of tournaments at all –economic and generational–levels of the sport. The idea behind the proposal is that the game and its fairness can be improved if the penalty area is drawn according to mathematics or a measure of actual scoring chance.

In the last thirty years, an increasing interest in sport sciences regarding the analysis of expert athletes’ gaze behavior has become apparent. This narrative review provides an overview of the use of eye tracking systems in high-performance sports from 1987 to 2016. A systematic search of the PubMed, Scopus, SPORTDiscus, and WebofScience databases was conducted. The search was performed using the keywords eye tracking, eye movement, gaze behavior/patterns, and visual search strategies in combination with high-performance sports, elite athletes, high-class athletes, sport experts, and top-athletes. It yielded a total of 86 studies of which almost half were conducted computer-based or in front of a screen. Most studies dealt with the analysis of gaze behavior during dead ball situations while also focusing on differences between expert athletes and novices. More high-quality intervention studies are essential to determine if there are ideal gaze strategies and, if yes, how it is possible to learn/implement these.

The players’ positions of tactical groups in soccer can be mapped to formation-patterns by means of artificial neural networks (Kohonen, 1995). This way, the hundreds of positional situations of one half of a match can be reduced to about 20 to 30 types of formations (Grunz, Perl & Memmert, 2012; Perl, 2015), the coincidences of which can be used for describing and simulating tactical processes of the teams (Memmert, Lemmink & Sampaio, 2017): Developing and changing formations in the interaction with the opponent activities can be understood as a tactical game in the success context of ball control, space control and finally generating dangerous situations. As such it can be simulated using mathematical approaches like Monte Carlo-simulation and game theory in order to generate optimal strategic patterns. However, in accordance with results from game theory it turns out that in most cases the one optimal strategy does not exist (e.g. see Durlauf & Blume, 2010). Instead, a variety of partial strategies with different frequencies were necessary – an approach that is mathematically interesting but has nothing to do with soccer reality. An alternative approach, which is developed in the following, is to interrupt the strictness of a single strategic concept by creative elements, which improves flexible response to opponent activities as well as prevents from being analyzed by the opponent team.

The results of respective simulation reach from improving strategic behaviour to recognizing strategic patterns and in particular to analyzing role and meaning of creative elements.