This paper examines three statistical studies related to sports performance, focusing on rugby and basketball. Drawing on the anthology of statistics in sports scholarship, the paper reviews the research methods, independent and dependent variables, and analytical techniques used in each study. Study I investigates differences in game statistics between winning and losing rugby teams in the Six Nations tournament using discriminant analysis. Studies II and III analyze game-related statistics that discriminate winning from losing basketball teams across consecutive games in Spanish youth leagues. The paper also evaluates the quality of each study's research design and offers suggestions for future improvements, highlighting the growing role of statistical analysis in understanding athletic performance.
"Statistics never lie," as sports enthusiasts have long insisted (as cited in Camillo, 2008, ¶1). Each day, a number of Americans, seemingly mesmerized with sports data collections, read newspaper accounts relating raw and summary data about a variety of sports. Donald Guthrie explains in "Introduction to the Miscellaneous Sports Articles" that "statistical studies of sports seem to fall into three categories" (Guthrie as cited in Albert, Bennett, & Cochran, 2005, p. 267). These categories include the analysis of rules and strategies, and the analysis of the extent of participation (Guthrie as cited in Albert, Bennett, & Cochran, 2005, p. 267).
In the book Anthology of Statistics in Sports, Jim Albert, professor of mathematics and statistics at Bowling Green State University; Jay Bennett, Principal Scientist with Telcordia Technologies and active member of the ASA Section of Statistics in Sports; and James J. Cochran (2005), Assistant Professor in the Department of Marketing and Analysis at Louisiana State University, investigate the use of statistics to analyze sports. The authors stress that researchers who collect and interpret data do not always limit their studies to only the major sports such as baseball, basketball, football, and hockey — some have even studied statistics for the game of darts.
This paper examines three studies relating to everyday statistics in sports, noting the method of each study and identifying the independent and dependent variables. It also reflects on the quality of the research processes each study utilized and offers suggestions for improvement. The focus is on statistics relating to rugby (one study) and basketball (two studies).
In the study "Differences in Game Statistics Between Winning and Losing Rugby Teams in the Six Nations Tournament," Enrique Orgega, Diego Villarego, and Jose M. Palao (2009) report results of a study they conducted using observational methods. Orgega, Villarego, and Palao used "univariate (t-test) and multivariate (discriminant) analysis of data… [to analyze] the differences in rugby game statistics between winning and losing teams…[; to evaluate] the data from 58 games of round robin play from the Six Nations tournament from the 2003–2006 seasons" (Abstract Section). The groups of variables included the number of points the teams scored, how the teams scored those points, the way the teams acquired the ball, and how the teams implemented tactical and technical aspects of the game.
Statistical analysis for Study I proceeded as follows:
First, a descriptive analysis of the data was conducted. Second, a Mann-Whitney U (non-parametric) test was carried out with the goal of analyzing the differences between winning and losing teams. Finally, a discriminant analysis was performed to identify those statistical variables that best differentiate winning from losing teams. Structural Coefficients (SC) greater than or equal to |.30| were considered relevant for the interpretation of the linear vectors. All statistical analyses were conducted at a significance level of p < 0.05. (Orgega, Villarego & Palao, 2009, Statistical Analysis Section, ¶1)
The quality of research in Study I appears effective, as a study's goal reportedly best determines the type of statistical analysis the researcher employs. The type of statistical analysis Orgega, Villarego, and Palao (2009) utilized replicated the methods used in the studies they reviewed. A researcher conducting a similar study in the future may benefit from considering particular external factors when determining "values that can be used as normative data to design and evaluate practices and competitions for rugby peak performance teams in a collective way" (Orgega, Villarego & Palao, Conclusion Section, ¶6). Incorporating factors external to team members' deliberate contributions may offer additional insight to further enhance knowledge about rugby — knowledge coaches can apply in their practice design.
In the case study "Effects of Consecutive Basketball Games on the Game-Related Statistics That Discriminate Winning and Losing Teams," Sergio-Jose Ibanez, Javier Garcia, and Sebastian Feu, all with the University of Extremadura; Alberto Lorenzo of the Technical University of Madrid; and Jaime Sampaio (2009) of the University of Trás-os-Montes e Alto Douro at Vila Real, analyzed data obtained from the Spanish Basketball Federation, including games played during the 2005–2006 and 2006–2007 seasons in the Under-20 league. The researchers used the observational method to analyze a total of 223 games.
The variables were "selected according to the FIBA (International Basketball Federation) normative for game analysis" (Ibanez, Garcia, Feu, Lorenzo, & Sampaio, 2009, Sample Section, ¶1) and included:
Two- and three-point field goals (made and missed), free throws (missed and made), offensive and defensive rebounds, assists, steals, turnovers, blocks (made and received), fouls committed, ball possessions, and offensive rating. Ball possessions and offensive rating were calculated using the following equations:
Ball Possessions = (field goals attempted) − (offensive rebounds) + (turnovers) − 0.4 × (free throws attempted)
Offensive Rating = (Points Scored / Ball Possessions) × 100
(Ibanez, Garcia, Feu, Lorenzo, & Sampaio, 2009, Sample Section, ¶1–3)
A 2 × 3 factorial ANOVA (game outcome: winners, losers; game number: first, second, third) was performed to identify univariate within-group main effects and interactions. A descriptive discriminant analysis was then employed to identify a subset of game-related statistics that discriminated between winning and losing teams across each of the three consecutive games played in each three-day competition. This analysis also identified the best mathematical equation so that the group means on the function were as different as possible. In each of the three subsamples (consecutive games), one discriminant function was obtained and interpreted based on examination of structure coefficients (SC) greater than |0.30|. Validation of discriminant models was conducted using leave-one-out classification — that is, each case was classified by applying the classification function computed from all data except the case being classified. Statistical analysis was performed using SPSS software 15.0, and significance was set at p ≤ 0.05. (Ibanez, Garcia, Feu, Lorenzo, & Sampaio, 2009, Data Analysis Section)
Ibanez, Garcia, Feu, Lorenzo, and Sampaio (2009) reported that their research did not confirm the "supposed influence" of accumulated fatigue across three consecutive games. To the contrary, overall team performances proved to be similar throughout. Royal et al. assert that a player's decision-making may actually improve under high-effort conditions. High-level performances reportedly link with:
Game rhythm, because more defensive rebounds implies more fast-break ball possessions; players' somatic characteristics, as taller and stronger players secure more rebounds; technical and tactical preparation, including pivoting, blocking, anticipation, securing and pulling the ball away; and muscular fitness, particularly in stretch-shortening-cycle jumping performances.
"Critical evaluation and recommendations for each study"
Across both rugby and basketball studies, the observational method provided a consistent and replicable research framework. Future studies in this area could benefit from incorporating external variables — such as travel fatigue, player workload metrics, and environmental conditions — that may influence game outcomes beyond the in-game statistics typically captured by federation records. Such additions would strengthen the explanatory power of the statistical models and offer more actionable guidance for coaches and sports scientists alike. An expanded dataset drawn from multiple seasons or international competitions could also improve the generalizability of findings across different levels of athletic competition.
The three studies examined in this paper demonstrate the breadth and rigor with which statistical methods are increasingly being applied to sports performance research. Study I illustrated how discriminant analysis can effectively differentiate winning from losing rugby teams when applied to Six Nations tournament data. Study II showed that consecutive game fatigue may not significantly alter the game-related statistics that separate winners from losers in youth basketball, while also highlighting the complex interplay of physical, tactical, and technical factors that drive performance. Each study employed the observational method and relied on multivariate statistical techniques to extract meaningful patterns from large game datasets. Taken together, they underscore the value of systematic statistical inquiry for coaches, sports scientists, and administrators seeking evidence-based strategies for improving athletic performance.
You’re 92% through this paper. Sign up to read the remaining 1 section.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.