Band 5, Heft 1

The concept of teammates whose skills `fit' together is explored through frescoball, the beach paddleball game with two people hitting a ball back and forth trying to keep it in the air. This simple game can be analyzed through Markov Chains to obtain the expected number of hits between the two players as a function of two skills, which we label as `athleticism' and `consistency.' Using this theoretical model and conceptual values of parameters, we examine various combinations where the complementarity of skills between teammates enhances the performance of the team. Given the conceptual model of team performance commonly used in sports analysis – that the sum of player ability equals team performance – we look at how such a conceptual model mismatches team performance, leaving the difference as `fit' of teammates. We choose three examples to illustrate characteristics of fit, particularly where player marginal value varies depending on who they are paired with. Further, simulating 50-player leagues where player movement is simulated by keeping players together if the team is successful and moving them if not, we show how estimates of `fit' and player ability can be confounded. Ultimately, we seek to frame the discussion on teammate `fit,' for which there is no attempt to quantify in sports literature.

The purpose of this study is to identify a function for the team performance for professional ice hockey teams in Sweden. In order to understand how team performance relates to key variables such as coaching ability and coaching experience and succession, the OLS (Ordinary Least Squares) and the more robust quantile regression techniques are used to estimate team performance for the ice hockey teams. Quarterly data for the period 1975-2006 is used for this purpose. The results have shown that coaching ability has a rather significant positive effect on team performance, and that managerial succession during the season is found to have a rather significant negative effect on team performance. The results also indicate a strong correlation between coaching ability and team performance. Moreover, the quantile regression approach provided a better understanding regarding the dynamics of the factors that affect performance and provided more interesting results than the OLS normally does.

This paper aims to analyse the efficiency of the teams that have participated in the last eight seasons in the Spanish football (soccer) league (LFP). Efficiency is based on technical aspects of the sport rather than variables of an economic nature due to how difficult it is to obtain such information from the clubs that make up the Spanish league. These measures are obtained for attack and defense, as both are essentials parts of football as a sport. To measure the efficiency we used a DEA model. Once the two facets mentioned above are defined, it is important to find out whether they are more directly related to the achievement of the points at stake in each match, as the final goal is to gain as many points as possible in order to win the competition or even to avoid are being relegated to the second division. Finally, we measure how the eight teams that have played in the first division for the last few seasons have performed. These teams are primarily the ones that have finished those seasons up the table. In order to achieve this, a DEA-Window analysis is carried out with two windows, as this makes it possible to compare two consecutive seasons. Most teams change either players or their manager from one season to another, which means there is little sense football-wise to analyse windows of more than three seasons.

The rank of an object is its relative importance to the other objects in the set. Often a rank is an integer assigned from the set 1,...,n. A ranking model is a method of determining a way in which the ranks are assigned. Usually a ranking model uses information available on the objects to determine their respective ratings. The most recognized application of ranking is the competitive sports. Numerous ranking models have been created over the years to compute the team ratings for various sports. In this paper we propose a flexible, easily coded, fast, iterative approach we call the Offense-Defense Model (ODM), to generating team ratings. The convergence of the ODM is grounded in the theory of matrix balancing.

Through various media channels we often hear about the value of sport as entertainment. Part of this entertainment stems from observing closely fought competition. In this paper, we ascertain the most competitive mainstream domestic competition using a generic rating engine which creates output known as "Team Lodeings." Team Lodeings measure the relative performance of sports teams.The motivation for the development of this ratings engine was to predict international rugby results. More specifically, to determine how Team A would fare against Team C given the performance of Team A against Team B and how well Team B went against Team C. Standardisation, converting the match result to a score ratio and an adaptation of the chi-square test of independence, make it possible to apply this methodology to different sports and compare results meaningfully.The method is validated internally and externally using established measures of team performance with data from 23 domestic competitions, from a variety of sports across several countries. As this method enables direct comparison of teams from different competitions it allows examination of issues such as the impact of the salary cap on competition balance.

This paper describes an objective way of predicting the likelihood of major league baseball players being elected to the Hall of Fame by members of the Baseball Writers’ Association of America. A radial basis function (RBF) network is used to build separate machine learning models for pitchers and non-pitchers. These models use simple player statistics such as number of wins and earned run average for pitchers and number of hits and home runs for non-pitchers. The models are trained on data representing players who played for at least 10 years and who retired between the years 1950 and 2002. In cross-validation tests, the models correctly identified 21 of 24 Hall of Fame pitchers, with 3 false positives, and 38 of 45 non-pitchers, with 7 false positives. When run over data representing active and recently retired players who played for at least 10 years, the models rate 12 pitchers and 22 non-pitchers to have approximately a 40% or better chance of election to the Hall of Fame.

In this paper we examine the use of neural networks as a tool for predicting the success of basketball teams in the National Basketball Association (NBA). Statistics for 620 NBA games were collected and used to train a variety of neural networks such as feed-forward, radial basis, probabilistic and generalized regression neural networks. Fusion of the neural networks is also examined using Bayes belief networks and probabilistic neural network fusion. Further, we investigate which subset of features input to the neural nets are the most salient features for prediction. We explored subsets obtained from signal-to-noise ratios and expert opinions to identify a subset of features input to the neural nets. Results obtained from these networks were compared to predictions made by numerous experts in the field of basketball. The best networks were able to correctly predict the winning team 74.33 percent of the time (on average) as compared to the experts who were correct 68.67 percent of the time.

We study the effect of the red card in a soccer game. A red card is given by a referee to signify that a player has been sent off following serious misconduct. The player who has been sent off must leave the game immediately and cannot be replaced during the game. His team must continue the game with one player fewer. We estimate the effect of the red card from betting data on the FIFA World Cup 2006 and Euro 2008, showing that the scoring intensity of the penalized team drops significantly, while the scoring intensity of the opposing team increases slightly. We show that a red card typically leads to a smaller number of goals scored during the game when a stronger team is penalized, but it can lead to an increased number of goals when a weaker team is punished. We also show when it is better to commit a red card offense in exchange for the prevention of a goal opportunity.

It is often postulated that an increased hip to shoulder differential angle (`X-Factor') during the early downswing better utilises the stretch-shorten cycle and improves golf performance. The current study aims to examine the potential relationship between the X-Factor and performance during the tee-shot. Seven golfers with handicaps between 0 and 10 strokes comprised the low-handicap group, whilst the high-handicap group consisted of eight golfers with handicaps between 11 and 20 strokes. The golfers performed 20 drives and three-dimensional kinematic data were used to quantify hip and shoulder rotation and the subsequent X-Factor. Compared with the low-handicap group, the high-handicap golfers tended to demonstrate greater hip rotation at the top of the backswing and recorded reduced maximum X-Factor values. The inconsistencies evident in the literature may suggest that a universal method of measuring rotational angles during the golf swing would be beneficial for future studies, particularly when considering potential injury.

The placing of NHL players into categories such as stars, average journeyman forwards, defensive forward, etc., is a popular pastime of fans and professional sports commentators. This is usually done in an ad hoc manner, sometimes contentiously, and without reference to any detailed analysis of empirical data. This paper structures the debate and provides a statistical basis for such categorization. This is achieved through an application of cluster analysis to formally allocate players to these categories. Using variables from the earnings determination literature, we are able to define the style of play of both forwards and defensemen. In addition, there appears to be a link to external outcome measures.

This paper uses K-means cluster analysis and multinomial mixture models to categorize professional golfer performance for the period 1980 to 2006. We collected, cleaned and analyzed final annual money list standings for PGA Tour players from www.pgatour.com. Correlation patterns between other measures suggested that career performance was well described by the proportion of years a player finished in specific meaningful money list categories such as Top 10 or outside the top 125. Using both clustering methods, we found that players divide into five natural and interpretable groupings, one being a small `Elite' group and four others which we refer to as `Distinguished,' `Established,' `Journeymen' and `Grinders.' We used analysis of variance to compare groups on the basis of other career performance measures including consistency, streakiness, longevity, and others as well as to investigate differences in the clusters produced by the two methods. A summary of these measures for the entire set gives a bleak outlook for any prospective professional golfer and singles out the greatness of the elite few, epitomized by Tiger Woods. This methodology extends to any sport or endeavor in which performance is measured in terms of end of year rankings or money lists.