Table 2 - uploaded by Barış Ergül
Content may be subject to copyright.
Discriminant Function, Structure Coefficients And Tests of Statistical Significance For NBA Teams
Source publication
The National Basketball Association (NBA) is one of the most popular and well-established men's professional basketball leagues in the world. Predicting of NBA playoffs between NBA teams poses a challenging problem of interest to statistical science as well as the general public. We concentrated on modeling and determining the variables from game-r...
Contexts in source publication
Context 1
... the technique of variable selection in discriminant analysis, we determined the variables that should be included in the model. The results are given in Table 2. ...
Context 2
... a large eigenvalue implies that the independent variables explain the dependent variable to a high degree. As given in Table 2, the eigenvalue was found to be 4.991 in the model and thus explained 88.3% of the variance. In addition, the canonical correlation coefficient was found to be 0.913. ...
Similar publications
This chapter proposes a new discriminant approach, called topological discriminant analysis (TDA), which uses a proximity measure in a topological context. The results of any operation of clustering or classification of objects strongly depend on the proximity measure chosen. The concept of topological equivalence uses the basic notion of a local n...
Citations
... After that, playoffs will begin to determine the western and eastern champion teams that will fight for the championship in the NBA Finals. NBA is not just a sports event but also a place for people to get away from stress [3]; the intensity and entertainment of NBA games have attracted numerous fans worldwide. Also, the NBA is a commercial brand. ...
Throughout the history of the NBA, a team’s popularity has been a critical factor in success. Legendary teams tend to have large, loyal, diverse fan bases and considerable revenue. On the contrary, teams with low visibility often face the dilemma of low exposure to the public, low fan participation, and even continuous loss of fans, resulting in low income, bad in-court performance, and even a decline in the team. This research paper will explore the connection between unpopular and popular teams in the NBA and the reasons behind it. This paper aims to reveal the multifaceted factors that influence the popularity of NBA teams and help future teams become popular and avoid failure. Focus on why the Detroit Pistons are unpopular by analyzing team history, geographic location, marketing strategies, team identification, competition with other sports teams, and team strengths and weaknesses. Also, compare the Detroit Pistons with the New York Knicks. While revealing the differences, this paper will also propose strategic recommendations to help unpopular teams overcome adversity and lead to success.
... Quantitative analysis of basketball performance through game-related statistics such as shooting percentage, rebounds, and assists has been widely used to measure player performance and analyze game events [28]. In this sense, which game-related statistics can discriminate winning and losing teams had become a hot topic for team performance studies in basketball [43][44][45][46]. Basketball games are won or lost by the number of points scored. ...
This study aimed to explore the differences in attack-defense performance between the top and bottom teams for starting and bench players during the Tokyo Olympics men’s basketball competition, to determine the relationship between the attack-defense performance of starting and bench players and the final competition rankings, as well as with each performance indicator. The rank-sum ratio (RSR) comprehensive evaluation was employed to describe the attack-defense performance of starting and bench players. Additionally, an independent sample t -test, Spearman Rho Correlation, and Pearson Correlation were conducted to test the differences and relationships between the various variables at a 0.05 level of significance, respectively. The results indicated that the top four teams showed significant differences in the attack-defense performance of their starting players compared to the bottom four teams ( p = 0.021), mainly in terms of 3-point shooting percentage ( p = 0.042) and free throw shooting percentage ( p = 0.044). Besides that, the attack-defense ranks of both starting players ( p = 0.004, r = 0.757) and bench players ( p = 0.020, r = 0.658) had a significant correlation with the final rankings. Points per game, 2-point field goal percentage, and assists had a statistically significant ( p <0.01) and strong positive correlation (0.70< r <0.90) with the attack-defense performance of both starting and bench players. 3-point field goal percentage, offensive rebounds, defensive rebounds, steals, and blocks were the technical indicators that distinguish starting from bench players. In conclusion, one of the common characteristics of the top national basketball teams was the strong attack-defense ability of the starting players. It is recommended that coaches select players with stronger 3-point shooting ability and more accurate free-throw shooting into the team’s starting rotation.
... In recent years statistical data collection during basketball games has been frequently used, and researchers use advanced statistical methods with an analysis of the performance of the games and the recognition of trends and models that are found to be dominant in the game (Sampaio et al., 2006;Sampaio et al., 2010a;Ergül, 2014;Čaušević, 2015;Doğan et al., 2016). With the utilization of notational analysis, basketball has become one of the most analysed sports in the world (Lorenzo et al., 2010) and the investigation on this subject has focused on discriminately identifying the winning and losing teams during the regular season and during the playoff games (Dogan et al., 2016;García et al., 2013;Ibánez et al., 2008;Ittenbach and Esters, 1995;Sampaio et al., 2010b;Taxildaris et al., 2001). ...
... Th e other variables contributing to the qualifying of the teams to the top sixteen were fouls received and blocks in favour during the game. Despite this, in some studies, there were no diff erences observed between the top half teams and the bottom half teams in terms of blocks (Doğan et al., 2016;Gómez et al., 2008;Ergül, 2014). In many studies, there were also no signifi cant results related to fouls received. ...
... In recent years statistical data collection during basketball games has been frequently used, and researchers use advanced statistical methods with an analysis of the performance of the games and the recognition of trends and models that are found to be dominant in the game (Sampaio et al., 2006;Sampaio et al., 2010a;Ergül, 2014;Čaušević, 2015;Doğan et al., 2016). With the utilization of notational analysis, basketball has become one of the most analysed sports in the world (Lorenzo et al., 2010) and the investigation on this subject has focused on discriminately identifying the winning and losing teams during the regular season and during the playoff games (Dogan et al., 2016;García et al., 2013;Ibánez et al., 2008;Ittenbach and Esters, 1995;Sampaio et al., 2010b;Taxildaris et al., 2001). ...
... Th e other variables contributing to the qualifying of the teams to the top sixteen were fouls received and blocks in favour during the game. Despite this, in some studies, there were no diff erences observed between the top half teams and the bottom half teams in terms of blocks (Doğan et al., 2016;Gómez et al., 2008;Ergül, 2014). In many studies, there were also no signifi cant results related to fouls received. ...
Euroleague is one of the most popular professional indoor sports leagues in the world. It is
globally ranked as the fi ft h-highest professional indoor sports league and the second-highest professional
basketball league, just trailing behind the National Basketball Association (NBA). Th e objective of this study
was to determine which game-related statistics can assist in predicting the team that will qualify for the next
rounds of the Euroleague. Th e data used in the study were obtained from each team’s offi cial average box
score on the Euroleague website for 2010-2017. Th e datasets were arranged into two groups depending on
the qualifi cation of the teams into the subsequent round. Discriminant analysis was applied to fi nd the gamerelated statistics that better contribute to qualifying for the next round. A three-point fi eld-goal percentage
was considered to be an essential variable in every round. However, it was also observed that, contrary to
expectations, off ensive rebounds had a negative eff ect in the fi nal four rounds. It is recommended these
results be used to plan the team strategies and the player strategies accordingly in a long-term and demanding
tournament like Euroleague.
... These results, however, were restricted to a descriptive level. The use of more advanced statistics to evaluate or forecast outcomes has been embraced by a majority of professional sports (Ergül, Yavuz, & Yavuz, 2014;Li, De Bosscher, Pion, Weissensteiner, & Vertonghen, 2018;Sgro, Barresi, & Lipoma, 2015). Inspired by the research of Reid and Morris (2013), which suggested that the ages associated with ranking milestones may have some forecasting potential in player's career peak ranking, the current study utilised regression and discriminant analyses to further explore the value of the age at different career milestones in predicting career peak ranking. ...
Learning from the career trajectories of the most successful elite players is central to informing effective strategies and long-term career planning to maximise player development and performance. This article examined the junior competition results and the performing age at major career milestones of top-level professional tennis players, utilising this information to forecast a player’s career peak ranking. Thirty years of longitudinal data which included 82 top 10 professional players between 2007 and 2017, were analysed. Gender and generational differences were compared. The results revealed that good performances at the highest junior level of competition was shown to be a critical precursor to eventual top-level professional success. It was revealed, however, that top 10 professional tennis players spent nearly 10 years from starting age to reaching an international junior level and another 10 years on average to achieve career peak ranking. Additionally, age at major career milestones was shown to be moderately correlated with a player’s career peak ranking, with 61% of the top one players correctly “predicted” to be top one players. The practical implications arising from these findings, specific to informing the career planning, prediction of professional success, monitoring and assessment of emerging tennis players, is discussed.
... For example, Sgrò, Barresi, Lipoma (2015) applied DA to identify indicators related to team match performances that could effectively predict winning or losing teams in a Football Championship. Similarly in basketball, DA has been heavily utilised to predict NBA playoff results (Ergül, Yavuz, & Yavuz, 2014). Specific to this study, regarding the usage of DA, the independent variable was "junior success" measured by the competition results which were represented by six sub-variables (i.e. total number of gold/silver/bronze medals won, colour of medals won, whether or not won medals and total number of medals won). ...
Currently in the literature, there is a dearth of empirical research that confirms whether international junior success is a reliable predictor for future international senior success. Despite the uncertainty of the junior–senior relationship, federations and coaches still tend to use junior success as a predictor for long-term senior success. A range of former investigations utilising a retrospective lens has merely focused on success that athletes attained at junior level competitions. Success that was achieved at senior-level competitions but at a junior age was relatively ignored. This study explored to what extent international senior success can be predicted based on success that athletes achieved in either international junior level competitions (i.e. junior medalists) or senior competitions at a junior age (i.e. early achievers). The sample contains 4011 international male and female athletes from three combat sports (taekwondo, wrestling and boxing), who were born between 1974 and 1990 and participated in both international junior and senior-level competitions between 1990 and 2016. Gender and sport differences were compared. The results revealed that 61.4% of the junior medalists and 90.4% of the early achievers went on to win international medals at a senior age. Among the early achievers, 92.2% of the taekwondo athletes, 68.4% of the wrestling athletes and 37.9% of the boxing athletes could be reliably “predicted” to win international senior medals. The findings demonstrate that specific to the three combat sports examined, international junior success appears to be an important predictor to long-term international senior success.
... While using advanced statistical data coaches could create and adapt more effective game strategies, also managers could have an opportunity to take a closer look at players' individual abilities and their value in the market. Research about basketball players and team performance analyses using comprehensive statistical data is new so far (Ergül, Yavuz & Yavuz, 2014). Basketball analytics is a growing area in Europe, so only few (http://www.inthegame. ...
Background. Many industries use a variety of statistical models for a decision-making, but no industry relies on analytical data as a professional sport (Davenport, 2014). Sports organization management and decision-making have a direct relationship with the sports teams and this relationship is called a comprehensive statistical analysis (Radovanović, Radojičić, Jeremić & Savić, 2013). Alamar (2013) argues that a detailed statistical analysis of the game activities is an important aspect in order to more accurately assess the player’s market value. Methods. The sample consisted of 30 respondents (10 managers, 10 coaches, 10 coach assistants) of the 10 men’s basketball teams. Managers and coaches of Lithuanian Basketball League teams had to fill in online questionnaire, the main focus of which was to identify their opinion about advanced data analytics. The questionnaire was designed based on scientific studies (Alamar, 2013; Martinez & Martinez, 2011). Questionnaire scales were tested using SPSS 20.0 statistical analysis program. Results. Statistical analysis showed that questionnaire reliability was average (Cronbach α = .549–.558). The survey results showed that the teams of Lithuanian Basketball League there employ professionals whose main goal is to analyse performance indicators, statistical data of opponents and new incoming players. Majority of managers and coaches believe that new information technologies of advanced basketball data could help improve team performance results and draw more attention to basketball from fans’ perspective. It was found that managers and coaches thought that offensive strategy depends on the players of the team. Coaches and managers had a positive approach to basketball analytics and believed that it had a bright future in basketball industry. Conclusions. The correct use of limitless data would definitely help improve team performance and effectively use their financial resources recruiting the most efficient players.
... Under this purpose, discriminant analysis was applied to the data and results obtained were evaluated. Ergül et al. (2014) reported that basketball game is a defensive game even though it is known for offense. In this context, when our findings are considered as a whole, the importance of defensive side of basketball is observed. ...
... In this study, it was determined that there was no difference between the top half teams and the bottom half teams in terms of successful 2 and 3-point field goals and successful free-throws. Ibáñez et al. (2008) for the successful 2 and 3-point field goals; Sampaio and Janeira (2003) for the successful 2-point field goals; Ergül et al. (2014) for the successful 3-point field goals and successful free-throws; Gómez et al. (2008) for the successful free-throws show similarities with our study. In addition; it was determined that there was no difference between the top half teams and the bottom half teams in terms of unsuccessful 2 and 3-point shots and unsuccessful free-throws. ...
... Furthermore, in our study, it was determined that there was no difference between the top half teams and the bottom half teams in terms of blocks. Gomez et al. (2008) and Ergül et al. (2014) obtained similar results with our study. ...
The aim of this study was to investigate the game-related statistics that discriminate between teams according to their league position in the Turkish Basketball League. Data were collected from the official box scores for 16 participated teams during the 2014-2015 regular season of the Turkish Basketball League. Teams were divided into two groups according to their league position; top 8 teams were called the top half teams and other 8 teams were called the bottom half teams. Players who played for less than one-quarter of the season or less than half a game period were omitted from the sample. The final data set obtained from 198 basketball players. The recorded game-related statistics were standardized according to a player's time on court. Independent samples t-tests were used to compare descriptive results from the game-related statistics. Also, discriminant analysis was applied to discover the statistical team variables that better discriminated between the two types of team. The mean differences between teams that were found to be statistically significant were assists, steals, defensive rebounds, turnovers and offensive rebounds. In the 2014-2015 regular season of Turkish Basketball League assists, steals, defensive rebounds, turnovers and offensive rebounds were the variables that had a major contribution to a team success. As a result, it was found that offensive rebounds, defensive rebounds, assists, steals and turnovers were more important game-related statistics than others. These results suggest that defense is more effective for team success than offense.
Bu çalışmanın amacı, 2020-2021 Basketbol Süper Ligi normal sezonunda oynanan müsabakalardaki oyun tipi istatistiklerini ve playoff’a kalan 8 takım ile kalamayan 8 takım arasındaki farklı incelemektir. Normal sezonda 16 takım arasında oynanan 480 müsabaka sistematik gözlem yolu ile incelenmiş ve 12 oyun tipinde sayı girişimi ve kazanılan sayı olarak toplamda 5760 istatistik kaydedilmiştir. Playoff’a kalan ve playoff dışı kalan gruba ait oyunla tipi istatistikleri karşılaştırmak için Bağımsız örnekler t testleri kullanılmıştır. Bu iki grubu ayırt etmeye katkıda bulunan oyun tipi istatistikleri değişkenlerini keşfetmek için ayırıcı fonksiyon analizi kullanılmıştır. Bulgular playoff’a kalan takımların sayı girişimlerinde catch and shoot, isolation ve transition sayı girişimi ortalamalarının kalamayan takımlardan anlamlı olarak daha yüksek olduğunu göstermiştir. Playoff’a kalan takımların en çok sayı girişimi Isolation ve Transition oyun tiplerindenden geldiği ayrıcı fonksiyon analizini bulgularında gözlenmiştir. Playoff’a kalan takımların Isolation ve Pick and Roll Roller oyun tipinden kazandıkları sayı ortalamalarının kalamayan takımlara göre istatiksel olarak anlamlı olduğu gözlenmiştir. Ayrıcı fonksiyon analizi bulguları isolation ve pick and roll handler oyun tipinin takım başarısına en büyük katkıyı yapan değişkenler olduğu göstermiştir. Bu sonuçlar Türkiye Basketbol Süper Ligi’nin yüksek tempoda, şut ağırlıklı, dış oyuncuların pick and roll oyunuyla ve ayrıca pivot oyuncuların sırtı dönük oyunuyla çembere atış yaptığı bir lig olduğunu göstermektedir.
