45 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 ASSESSING COMPETITIVE AND POTENTIAL SUCCESSFULNESS OF ADVANCED YOUNG VOLLEYBALL PLAYERS OCENJEVANJE TEKMOVALNE IN POTENCIALNE USPE[NOSTI KAKOVOSTNIH MLADIH ODBOJKA[EV Marko Zadra`nik Abstract The potential successfulness of a sample of twenty- four male volleyball players, between fifteen and six- teen years of age from four selected teams, in morp- hological-motor space was assessed with the expert system NDN. On the basis of the marks for playing successfulness given by volleyball experts to the indi- vidual players we also formed marks for competitive successfulness. Their congruence was assessed by different correlation coefficients. The results show a high and statistically significant correlation of the marks for potential successfulness (generated by the expert-system method) and com- petitive successfulness. This means that we used sui- table variables from the morphologic-motor space to assess potential successfulness and also an appro- priate method. We can therefore successfully assess the potential successfulness of volleyball players in a simple and intelligible way and predict their compe- titive successfulness as well. The obtained informa- tion is also useful in the process of orientation and selection of children into volleyball and in planning and managing the training process. Keywords: volleyball, youth, potential successfulness, competitive successfulness, expert system Izvle~ek Na vzorcu 24 odbojkarjev starih 15 in 16 let, iz {tirih izbranih ekip, smo z ekspertnim sistemom NDN ugotavljali njihovo potencialno uspe{nost v morfo- lo{ko-motori~nem prostoru. Na osnovi ocen odboj- karskih ekspertov o uspe{nosti igranja posamezne- ga igralca smo oblikovali tudi ocene tekmovalne us- pe{nosti. Skladnost obeh ocen smo ugotavljali z raz- li~nimi koeficienti korelacije. Rezultati ka`ejo visoko in statisti~no zna~ilno pove- zanost ocen potencialne (izra~unane z ekspertno metodo) ter tekmovalne uspe{nosti. To pomeni, da smo za oceno potencialne uspe{nosti v morfolo{ko- motori~nem prostoru uporabili ustrezne spremen- ljivke in tudi primerno metodo. Na tak pregleden in enostaven na~in lahko torej uspe{no ocenimo po- tencialno uspe{nost odbojkarjev in na osnovi te oce- ne tudi napovemo njihovo uspe{nost na tekmova- nju. Dobljene informacije so uporabne tudi v pro- cesu usmerjanja otrok v odbojko, selekcioniranja odbojkarjev ter pri na~rtovanju in vodenja procesa treniranja. Klju~ne besede: odbojka - mladi igralci - potencialna uspe{nost - tekmovalna uspe{nost - ekspertni sistem University of Ljubljana – Faculty of Sport, Gortanova 22, SI-1000 Ljub- ljana, Slovenia tel.: +386 61 140-10-77 fax: +386 61 448-148 E-mail: marko.zadraznik@guest.arnes.si Received: 11. 01. 1999 – Accepted: 21. 10. 1999 46 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 Introduction Volleyball belongs, according to its motor structure, to polystructural acyclic sports and by the number of pla- yers to team sports (games). The successfulness of a pla- yer or a team is, because of the complexity of the game, dependent on a multitude of factors and interactions between them. Important advances have been made in studying such problems with the introduction of modern scientific approaches (cybernetics, system theory) and use of computer technology. Because of the large number of factors influencing success, which need to be taken in- to account, various expert system methods are asser- ting themselves (Lesko{ek, 1996). The idea is to join humans and computers in order to solve certain problems. The methods are useful also for research in sport. Some researchers are still studying the possibilities of their use in sport (Blahus, 1990), ot- hers already use them for solving practical problems in recreation (Yie, Tang and Wang, 1998), controlling a match (Vickers and Kingston, 1987), planning training- loads of athletes (Lee and Kim, 1992), track and field (Barac, Situcek and Kodejs, 1992), cross-country ski- ing (Sprigings, Dorotich and McGibney, 1992), swim- ming (Persny, Willems, Van Tilbourgh, Daly and Ver- vaecke 1987) etc. Such methods have found their pla- ce also in researching problems in sports games: ba- sketball (Nissan, Simhon and Zigdon, 1992), rugby (Singer and Villepreux, 1995) etc. In Slovenia researchers use them for assessing poten- tial successfulness of players in different sports (Bon, 1998; ^oh, ^uk and Bor{tnik, 1993; Dolenc, 1996; Er- ~ulj, 1996; Filip~i~, 1996; Le{nik, 1996; Pustovrh, 1994; [ibila, 1995; Zadra`nik, 1998). Two expert met- hods are mainly used: NDN and NDP (Lesko{ek, 1996). Findings are quite unanimous: • the correlation between marks of potential and competitive successfulness is high, • the knowledge database should be enhanced furt- her with new findings on playing successfulness, • this method is efficient, informative and cheap. The sphere of evaluating successfulness of a player is one of the important areas for scientific and research work also in volleyball. The successfulness of a player can be assessed in two ways. First by assessing (compu- ting) his/her potential, or second, by assessing (measu- ring) the manifestation of this potential in competition. In the first case we talk about potential successfulness - ie. potential - (PS) in the second about competitive suc- cessfulness (CS). Competitive successfulness, on the ot- her hand, is obtained on the basis of marks of success- fulness at matches in a certain period. The individual game is weighed according to its competitive level. The foundation of a player’s successfulness is his/her PS, which must be suitably developed with numerous factors, representing training conditions, competitive circumstances, etc. These depend a lot on the functio- ning of the factors of the wider milieu (role of parents, tradition of volleyball,…). PS of a player is composed of his/her basic, realisational and motivational factors, as well as experience. Aut- hors researching successfulness in volleyball stress the importance of morphologic characteristics and motor abilities. Their affect on successfulness in volleyball is also researched the most. We have therefore decided to base the computation of PS on several indicators of these two sub-spaces. The variables, with which we measured the factors important for success in volley- ball, were chosen on the basis of findings of numerous Slovene and foreign authors. The reasons and sources for this selection are described in more detail in the dis- sertation »Tekmovalna uspe{nost in psihosomati~ni po- tencial kakovostnih mladih odbojkarjev« [Competitive successfulness and psychosomatic potential of advan- ced young volleyball players] (Zadra`nik, 1998). PS was computed with the NDN expert method (met- hod of weighted usefulness) and the obtained marks compared with actual competitive successfulness (CS) at the finals of the 1995 National Cadet Championship of Slovenia. In the case that the PS and the CS marks show adequate congruence, we would have at our dis- posal a useful system for assessing the level of potential successfulness of young volleyball players and a source of additional information for analysing the efficiency of the training process. METHODS Subject sample The sample comprised of twenty-four male players (from fifteen to sixteen years of age) from four teams that competed at the finals of the Slovene cadet championship in 1995. Six players were selected from each team according to their playing time. Each player took part in two games of the finals. The cho- sen players trained 3-4 times/week and were invol- ved in approximately the same number of training and official games up to the finals. The team mem- bers were therefore sufficiently used to playing with each other. All above shows that the conditions allo- wed the players to demonstrate their full potential. The measurements of the morphologic and motor dimensions of the players of the chosen teams were made at the Faculty of Sport in Ljubljana, a week pri- or to the finals. Variable sample Three anthropometric measures and two morpho- logic indices were chosen to represent the morpho- logic space (see Table 1). Conditional capabilities were assessed on the basis of the results of eight tests and specific motorics without the ball with six. The 47 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 end result of the player in a certain variable was the average of three repetitions. Specific motorics with the ball was assessed by precision and stability of tar- get hitting with the spike and overhand pass. Preci- sion of hitting the target is an average of three repe- titions, while we computed stability with the follo- wing formula: NAPS (PODS) = where X represents the average score of three repetitions and X the sum of all hits in the individual repetition. The criterion variable CS is represented by the ave- rage mark of successfulness of the individual player at two final matches of the cadet championship. The players were assessed by three volleyball experts ac- cording to set criteria with marks from 1 to 5, preci- se to one decimal (Zadra`nik, 1998). Model of potential successfulness in volleyball, after Zadra`nik (1998): SUCCESSFULNESS OF A VOLLEYBALL PLAYER —POTENTIAL OF WIDER MILIEAU —POTENTIAL OF TRAINING CONDITIONS, COMPETITION, QUALITY OF OPPONENT —potenTial OF PLAYER —Basic dimensions —morphologic characteristics —functional capabilities —motor abilities and technical skills (special motorics with the ball) —basic motor abilities —informational component of movement —energy component of movement —special motorics without the ball —informational component of movement —energy component of movement —special motorics with the ball —Realisational and motivational dimensions —personality characteristics —socio-psychologic characteristics —structural characteristics —competitive characteristics —abilities —psychomotor abilities —perceptive abilities —mental abilities —tactical skills —motivation and ambition —strength —incentive motivation —competitive motivation —professional ambitions —sports ambitions —Experience —general —in current competition 48 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 Data analysis methods The programme PETA was used to test for linearity of correlation between the predictor variables and the cri- terion. Blackman’s test of linearity (Ferguson, 1966) was used. All variables where the relation was found to be non-linear were linearised by a logarithmic trans- formation. The results in the chosen morphologic and motor va- riables were treated with the computer programme KISS 1.2 (De`man and Lesko{ek, 1990). The base of this programme is a knowledge database for a decision system for potential successfulness. The shell consists of a decision mechanism and a communication inter- face. A morphologic-motor decision tree for PU was constructed (see the first column of Table 1). Then we defined the weights on all the levels of the decision tree. A greater weight was given to those factors that have a greater influence on competitive successfulness in volleyball (see second column of Table 1). We also constructed normalisers in centile values, which the WEIGHT PLAYER20 PLAYER2 CHARACTERISTIC, ABILITY, SKILL CS 4.8 2.5 assessment of competitive successfulness PS 100.0 4.1 2.2 assessment of potential successfulness – MORPHOLOGY 35.0 4.6 1.8 morphologic characteristics – LONGITUDINAL 26.6 5.4 1.3 longitudinal dimensionality – AV 13.3 5.3 1.3 body height – ADL 13.3 5.5 1.4 length of palm – VOLUMINOSITY 5.6 0.7 2.8 body volume – ITV 4.2 0.7 2.8 ratio (weight/height) x 100 – AMAP 1.4 0.9 2.7 percent of body fat according to Matiegka – SPECIAL MORPHOLOGY 2.8 4.4 4.6 special anthropometry – AKOM 2.8 4.4 4.6 distance between the elbows – MOTORICS 65.0 3.9 2.4 motorics – CONDITION 13.0 3.6 3.2 conditional abilities – ENERGY 6.5 4.3 2.2 energy component of movement – STRENGTH 5.2 4.6 2.0 strength – EKS 3.1 4.7 2.5 explosive power – ENOG 1.7 4.4 2.0 explosive power of legs – SER 0.9 3.9 2.5 height of take-off after Sargent – SP5 0.9 4.9 1.4 sprint 5 m – EROK 1.4 5.1 3.1 explosive power of arms – MED 1.4 5.1 3.1 throwing the medicine ball while sitting – ELAST 2.1 4.5 1.2 elastic power – SP20 2.1 4.5 1.2 sprint 20 m – ENDURANCE 1.3 3.1 3.3 endurance – T2KM 1.3 3.1 3.3 2 km run – INFORMATION 6.5 2.8 4.1 informational component of movement – FLEXIBILITY 0.6 4.3 3.5 flexibility – PRK 0.6 4.3 3.5 forward bend and touch on a bench – COORDINATION 4.2 1.6 4.6 co-ordination – POL 4.2 1.6 4.6 polygon backwards – HITALTGIB 1.6 5.3 3.2 speed of alternate movements – TAPR 1.6 5.3 3.2 arm plate-tapping – SPECBRZOGA 19.5 3.1 1.2 special motorics without a ball – SMOC 7.8 5.4 0.7 special power – BLOK 3.9 5.4 0.7 achieved blocking height – VOZ 3.9 5.3 0.7 achieved height from a rush – SPEC. ENDURANCE 3.1 3.1 2.3 special endurance – SVZDREKSM 2.2 2.6 1.8 special endurance in explosive power – VEM 2.2 2.6 1.8 endurance in jumps – SVZDRHITR 0.9 4.3 3.4 special endurance in speed – JEL 0.9 4.3 3.4 »fir tree test« (from Russian) – SCOORD 8.6 1.0 1.2 special co-ordination – PIR 5.6 0.6 1.2 pirouette – AGILITY 3.0 1.8 1.1 agility – TPI 3.0 1.8 1.1 running from line to line – SPECZZOGA 32.5 4.5 2.8 special motorics with the ball – ATTACK 16.2 4.6 2.9 spike – NAPX 8.9 5.4 1.8 average result in hitting target with spike – NAPS 7.3 3.6 4.4 stability of hitting target with spike – ZGORNJIODB 16.2 4.5 2.7 overhand pass – PODX 8.9 5.5 1.5 average result in hitting target with overhand pass – PODS 7.3 3.2 4.1 stability of hitting target with overhand pass Table 1: Decision tree with weights and examples 49 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 computer uses to transform the results of the players in the chosen tests into marks between one and five. The normalisers are shown in Table 2. Table 2: Marks in centile and z values The reliability of the judges (experts) was assessed by the coefficient of concordance. The congruence of the marks was found on the basis of the percent of com- mon variance of measurement (first principal compo- nent). The level of association between potential suc- cessfulness (PS) and the competitive successfulness (CS) was measured by the Pearson correlation coeffi- cient and Spearman’s rank correlation. Results The tests of linearity of the association between the pre- dictors and the criterion showed that only the variables MED (throwing the medicine ball) and NAPX (hitting the target with the spike) have a non-linear relation with the criterion. The relation was linearised by trans- forming the results into logarithms. A statistically signi- ficant correlation with the criterion was obtained for the following variables: AV , ADL, SP5, MED, PIR, VOZ, BLOK, NAPX, PODX. The basic statistical parameters of the criterion variab- le (CS) are presented in Table 3. The marks of the jud- ges were sufficiently reliable. The coefficient of con- cordance (W) is statistically significant (p = 0.000). The accord of the judges was also satisfactory, the common variance of the given marks was 77.4% (see Table 3). The marks of the players at the competition are there- fore suitable to form the criterion variable CS. Table 3: Reliability of the criterion variable The marks of potential successfulness in the morp- hological-motor space and the marks of competitive successfulness of all the players are presented in Graph 1. Graph 1: Players’ marks of competitive and potential successfulness Table 4 shows the basic statistical parameters of the marks of potential successfulness computed on the basis of the variables of the morphologic-motor spa- ce and the marks of competitive successfulness. Table 4: Basic statistical parameters of potential and competitive successfulness In Table 1 we give as an example the results and marks of two players, who were judged at the com- petition with the highest (PLAYER20) and the lowest (PLAYER2) mark. It is at the same time a coinciden- ce and also a manifestation of reality that they obtai- ned the highest and the lowest computed mark of potential successfulness as well. The congruence between the PS marks in the morphologic-motor space (computed with the NDN expert method) and the CS marks is high. Both correlation coefficients are statistically significant already at the 1% error le- vel (Table 5). We were able to explain 52% of the va- riance of the criterion with such a system of variables of the morphologic-motor space. Table 5: Correlation between the marks of potential and competitive successfulness Discussion This study had several goals. We verified the method and adequacy of assessing CS (competitive successful- ness at a match), that was used for the first time in this study. At the same time, we also verified the usefulness of artificial intelligence software (in our case the expert method NDN) in computing the PS (potential success- fulness). The main goal of the study was the construc- tion of the knowledge database on the basis of expert knowledge. The results of the study show that we used an appro- priate method in forming the criterion variable (CS). We can also say that the selection of variables used to compute the PS marks in the morphologic-motor spa- ce was adequate. The good and the bad points of a players potential can be quickly determined from the results and the com- puted marks on the different levels of the decision tree (see the two examples in Table 1). We can also find out in the same way how the player compensates his cha- racteristics and abilities (De`man, 1996). MARK 1 2 3 4 5 CENTILE < 16 >= 16 >= 31 >= 69 >=86 z – value < –1.0 z >= –1.0 z <= –0.5 z <= 0.5 z <= 1.0 z Common variance W p (W) Hi 2 CS 77.4 % 0.810 0.000 55.87 X ó MIN. MAX. PS (morphological-motor 2.96 0.47 2.2 4.1 space – NDN) CS 3.17 0.69 2.2 4.8 Pearson correlation coefficient r = 0.72 r2 = 0.52 p = 0.000 Spearman correlation coefficient rho = 0.58 p = 0.003 50 Marko Zadra`nik (1999). Assessing competitive and potential successfulness of advanced young… KinSI, 5(1–2) : 45–50 Information presented in such a way allows the coach a clinical evaluation of each player. Further study is needed to be able to use such a system in praxis, to see if similar results can be reached with less variables. It must be stressed that the knowledge database used in this case is suitable only for assessing PS of 15-16 year old players. For other age categories the knowledge da- tabase should be suitably modified and supplemented. The influence and contribution of some variables (body height, endurance in strength, endurance in speed...) to successfulness in volleyball can be, in our opinion, quite different in other age groups than the one we analysed. The marks of competitive successfulness in the morp- hologic-motor space show that the players mostly pla- yed the finals of the championship in accordance with their potential. The marks of potential successfulness confirm the findings of other authors (De`man, 1996; Er~ulj, 1996; Filip~i~, 1996; Pustovrh, 1994; Zadra`- nik, 1998). These authors used the described method to compute the players’ PS in other sports and found that the best players achieve a high congruence of both marks. Such a tendency is also evident for players with a lower manifestation of their potential. The players which are placed between the two extreme groups ac- cording to their PS oscillate much more in their CS marks. The results of the study showed that we managed to construct an informative and simple system for monito- ring the morphological, motor and playing status of a young volleyball player, which can be an additional in- formation source to the coach, while monitoring and analysing the efficiency of the training process. And from the viewpoint of praxis, the most important fin- ding is that such an expert system is able to differentia- te between better volleyball players and worse ones. This shows that the findings of this study can be useful also in some other areas (ex.: orientation of children in volleyball, selection of players, etc). References 1. Barac, F., Situcek, M., & Kodejs, M. (1992). Expert system 110 m hurdles. In Proceedings of the International Conference on Compu- ter Applications in sport and Physical Education (pp. 188-191). Ne- tanya: The Zinman College. 2. Blahus, P. (1990). Pouziti statistickych metod a moxnosti expertnich systemu. [Application of statistical methods and possibilities of ex- pert systems]. Gymnica, 26 (1), 95-104. 3. Bon, M. (1998). Povezanost izbranih morfolo{kih in motori~nih raz- se`nosti mladih rokometa{ev z uspe{nostjo v rokometni igri [Corre- lation of chosen morphologic and motor dimensions of young hand- ball players with their success in handball]. Master’s thesis, Ljublja- na: Fakulteta za {port. 4. ^oh, M., ^uk, I., & Bor{tnik, I. (1993). Kinemati~ni model skoka v vi{ino, ovrednoten na podlagi ekspertnega modeliranja [Kinematic model of high jump, evaluated with expert modelling]. [port, 41 (1- 2), 41-46. 5. De`man, B. (1996). Diagnosticiranje morfolo{kega, motori~nega in igralnega stanja ter razvoja mladih ko{arkarjev [Diagnosing morpho- logic, motor, playing status and development of young basketball players]. In Zbornik radova 3. konferencije o sportu Alpe – Jadran (pp. 72-77). Zagreb: Fakultet za fizi~ku kulturu. 6. De`man, B., & Lesko{ek, B. (1990). Ra~unalni{ko podprt informa- cijski sistem za ugotavljanje in spremljanje telesnega, motori~nega in igralnega razvoja mladih ko{arkarjev in ko{arkaric [Computer as- sisted informational system for assessing and monitoring physical, motor and playing development of young male and female basket- ball players]. In I. International symposium »Sport of the Young«, Bled, Slovenia (pp. 617-622). Ljubljana, Fakulteta za telesno kultu- ro. 7. Dolenc, M. (1996). Vrednotenje modela uspe{nosti mlaj{ih deklic v alpskem smu~anju [Evaluation of a successfulness model of female juniors in alpine skiing]. Master’s thesis, Ljubljana: Fakulteta za {port. 8. Er~ulj, F. (1996). Ovrednotenje modela ekspertnega sistema poten- cialne in tekmovalne uspe{nosti mladih ko{arkaric [Evaluating an ex- pert system model of potential and competitive successfulness of young female basketball players]. Master’s thesis, Ljubljana: Fakul- teta za {port. 9. Ferguson, G.A. (1966). Statistical analysis in psychology and educa- tion. London: Mc Graw – Hill. 10. Filip~i~, A. (1996). Evalvacija tekmovalne in potencialne uspe{nosti mladih teni{kih igralcev [Evaluation of competitive and potential suc- cessfulness of young tennis players]. Doctoral dissertation, Ljublja- na: Fakulteta za {port. 11. Lee, J., & Kim, K.J. (1992). Development of expert system on the diagnosis and training prescription of cardio-respiratory function in athletes. In Proceedings of the International Conference on Compu- ter Applications in sport and Psychical Education (pp. 277-284). Ne- tanya: The Zinman College. 12. Lesko{ek, B. (1996). Komparativna analiza ekspertnih metod z vi- dika njihove uporabnosti za za~etni izbor in usmerjanje otrok v raz- li~ne {portne panoge [Comparative analysis of expert methods from the viewpoint of their usefulness for initial selection and orientation of children in different sports disciplines]. Doctoral dissertation, Ljubljana: Fakulteta za {port. 13. Le{nik, B. (1996). Vrednotenje modela uspe{nosti mlaj{ih de~kov v alpskem smu~anju [Evaluation of a successfulness model of young male alpine skiers]. Master’s thesis, Ljubljana: Fakulteta za {port. 14. Nissan, E., Simhon, D., & Zigdon, N. (1992). Resource evaluation and counter-planning with multiple-layer rule sets, in the BASKET- BALL expert system. In Proceedings of the International Conferen- ce on Computer Applications in sport and Psychical Education (pp. 60-80). Netanya: The Zinman College. 15. Persny, U., Willems, E., Van Tilbourgh, L., Daly, D., & Vervaecke, H. (1987). An expert system on personal computer for diagnosis and advice of elite swimmers. Sport international, June (71), 32-37. 16. Pustovrh, J. (1994). Evalvacija tekmovalne uspe{nosti mladih smu- ~arjev teka~ev na osnovi stopenjske regresijske analize in ekspertne- ga modeliranja [Evaluation of competitive successfulness of young ski runners on the basis of hierarchical regression analysis and expert modelling]. Doctoral dissertation, Ljubljana: Fakulteta za {port. 17. Singer, B., & Villepreux, P. (1995). Internal and external validation of a tactical decision-making expert system in team gams: the rug- by case-study. In Windows to the future: bridging the gaps between disciplines, curriculum and instruction (Book of abstracts). Netan- ya: The Zinman College. 18. Sprigings, E., Dorotich, P.D., & McGibney, S. (1992). Development of a knowledge based expert system for classic techniques in cross- country skiing. Saskatoon: University of Saskatchewan. 19. [ibila, M. (1995). Oblikovanje in ovrednotenje informacijskega siste- ma za iskanje nadarjenih rokometa{ev in spremljanje njihovega raz- voja [Construction and validation of an information system for fin- ding talented handball players and monitoring their development]. Doctoral dissertation, Ljubljana: Fakulteta za {port. 20. Vickers, J., & Kingston, G.E. (1987). Modelling the master coach: building an expert system for coaching. In Proceedings of the Inter- national Conference on Computer Assisted Learning in Post-Secon- dary Education (pp. 207-212). Calgary: University of Calgary. 21. Yie, G., Tang, H., & Wang, G. (1998). Chinese specialised PC expert system for aged people’s exercise prescription. Journal of Wuhan Institute of Physical Education 6 (1), 68-73. 22. Zadra`nik, M. (1998). Tekmovalna uspe{nost in psihosomati~ni po- tencial kakovostnih mladih odbojkarjev [Competitive successful- ness and psychosomatic potential of talented young volleyball pla- yers]. Doctoral dissertation, Ljubljana: Fakulteta za {port.