Naslov
Euro-BES, A Computerized performance evaluation model to grade basketball player performance in the Euro-League

Autori
Swalgin, Keneth ; Knjaz, Damir

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
World Congress of Performance Analysis of Sport 7: Book of Abstracts / Dancs, Henriette ; Hughes, Mike ; Ekler, Judit H. - Szombathely : Berzsenyi Daniel College, 2006, 54-54

Skup
World Congress of Performance Analysis of Sport 7

Mjesto i datum
Szombathely, Mađarska, 23.08.2006. - 26.08.2006

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
basketball ; evaluation model

Sažetak
The Euro-BES model was developed to grade individual basketball performance in relationship to position-of-play under game conditions. The Basketball Evaluation System (BES) is a computerized model that incorporates three measurement constructs that can be applied to most team sport structures ; a) common set of performance criteria specific to the sport, b) a norm based context to measure the criteria, and c) an accurate functional measurement system inherent to the structure of the sport. Data for the Euro-BES model was collected for all players by position-of-the sport. Data for the Euro-BES model was collected for all players by position-of-play in the 2004-05 Euro-League season of play. The sample represents all professional male players (n=319) for the 24 team league. Position-of-play were designated by coaches from respective team. Performance norms were established for nine positions: point-guard-1, shooting guard-2, small forward-3, power forward-4, center-5 and "swing" positions 1-2, 2-3, 3-4, 4-5. Normative data for 10 performance factors were established: 2FG%, 3FG%, FG%, FT%, REB, PF, AST, TO, BLK ; and STL. The EURO-BES model is based on "normal probably theory" where raw game scores for each performance factor are standardized and scaled to produce a grade ranging from 0.0 to 4.0, with 2.0. representing the mean or average grade. Grades are generated for each performance factor and used to create an overall grade for the game being evaluated. The model also produces grades on season-to-date basis. To strenghten the validity of performance factor scores, a group of expert coaches (n=18) were asked to weigh each performance factor for its importance to the position. The weighted scores established by the coaches were then incorporated to produce a factor-weighted model. To further increase the accuracy of the model, performance factors are not graded until a minimum "time treshold" is passed. This insures that players have sufficient playing time to acquire a raw score on a performance factor. Time tresholds are based on the average number of minutes it takes to produce a raw game score i.e. a rebound, a personal faul, etc. The model also give coach's the ability to eliminate a performance factor if he/she feels the factor is not relevant to the position, i.e. blocked shots for point guards or 3FG% for centers. One of the primary constructs of the model is the generation of grades in relationship to time-played. Six of the 10 performance factors are graded in relationshio to time-played. Six of the 10 performance factors are graded in relationship to "minutes played". For rebounds, personal fouls, assists, turnovers, blocked shots and steals, "minutes played" in the common denominator for these factors and are evaluated on the basis of a "real game" or 40 minutes-of-play. To illustrate this measurement concept, a player who acquired 6 rebounds in 20 minutes-of-play would receive a higher score on the factor than if he/she scored 6 rebounds in 30 minutes-of-play. For 2FG%, 3FG%, FG% and FT%, the relationship between shots made and shots attempted in the common denominator for these factors.

Izvorni jezik
Engleski

Znanstvena područja
Kineziologija

POVEZANOST RADA