Talent Identification in Soccer
A critical analysis of contemporary psychological research

By Dr. Ian Lawrence
York St. John University

Elite athletes in all sports have to repeatedly perform under high pressure and at a high level. This possibly is best exemplified by the scrutiny applied to the world of the professional association football player. It is therefore not surprising that many researchers argue that psychological characteristics often distinguish those successful at the highest standard from their less successful counterparts (Morris, 2000). Early research already has supported an association between psychological characteristics and sports performance (Morgan and Pollock, 1977; Morgan, 1979; May et al., 1985). Further research has evolved with an emphasis in identifying psychological skills relevant to sport (Meyers et al., 1996). Mahoney et al. (1987) identified potential constructs assessing motivation, confidence, anxiety control, mental preparation, team emphasis and concentration.

For several years, researchers have tried to identify key predictors of talent in many sports. A decade ago, Regnier and colleagues (1993) published a review on talent detection and development in sports with the purpose of providing a better understanding of the process by which one achieves greatness in sports. Until that time, most studies were cross-sectional in nature and measured general characteristics. Literature on talent identification and development has largely emerged during the 1990s. Books that contribute substantially to our basic understanding of expertise are ‘The Road to Excellence: The Acquisition of Expert Performance in the Arts, Sciences, Sports, and Games” by Ericsson (1996) and “Expert Performance in Sports,” edited by Starkes and Ericsson (2003). Some years ago, the Journal of Sports Sciences devoted a special issue to talent identification and development in soccer (Williams and Reilly, 2000 a)

Many methods of talent identification (TI) are in use today. The Southern Hemisphere in particular seems to be interested in the talent identification process (Hoare, 1996; McClymont, 1996). This could be because of the relatively low populations of those nations. Although the United Kingdom is significantly larger than the Southern Hemisphere countries, it still is significantly smaller than nations such as the United States and China. If the United Kingdom is to remain competitive internationally, it cannot continue to simply wait for talent to emerge through its national competitive structure. When talent is allowed to emerge through a national competitive structure, those nations with the largest populations will be dominant. If Olympians or elite athletes were “one in a million,” large nations would have others to replace the lost talent. Conversely, smaller nations would not have the base population sufficient to compensate.

Function of Talent Identification Systems
Talent Identification is perceived by many governments as a means to harness a nation’s sporting talent, to bring about future success in the international arena (Burns, 1996; Prescott, 1996). Talent identification generally takes the form of the selection of a series of tests that are thought to measure key factors for success in a specified sport (Balyi and Hamilton, 1995). These measures then are applied to as large a population as is possible. In this way, talent identification systems act as a filter to remove people who have relatively few perceived important characteristics, leaving people who should have a relatively strong chance of success in that sport. This enables the governing body of that sport to concentrate much of its resources into developing these talented few performers. In this way TI schemes can be used to increase the profile of sports in the community and attract more athletes to a sport. Once an athlete is attracted to a sport, a TI system then can be used to target resources to athletes identified as having the greatest potential. The arguments against TI can be easily recognized. Perhaps the most fundamental argument raises questions about the whole ethos of sport. Is sport something that should be taken so seriously that a TI scheme is necessary, or should talent be allowed to emerge naturally through freedom of choice? In the United Kingdom, there still exists an ethos of non-interference with sports, more a form of ‘talent emergence’ rather than identification. The philosophy is, “If you are good enough, you will make it.” In many well documented cases British athletes succeed despite the system, rather than as a result of it.

Talent identification systems are more premeditated in their approach to uncovering talent. This may be viewed as unsporting or as a form of social control. It could be perceived that sports people are being “shoehorned” into sports to which they are physically or psychologically unsuited. In fact, research has demonstrated that this has not been the case in Western TI systems to date (Grice, 2003). The athletes are merely advised as to their potential for success in certain sports.

A second argument that can be leveled against TI systems is the cost. National TI systems require time and money to carry out effectively. This money has to come from somewhere and will inevitably reduce spending in other areas of a government’s agenda. Within the context of sport, the further dilemma is whether funds devoted toward TI should be allocated to developing and supporting current elite sports people. Additionally, funds directed toward elite sport may result in less grassroots investment in sport. This again may reduce the pool of athletes from which future elite performers are drawn. This could damage the likelihood of the nation achieving international success in chosen sports, unless the TI program is successful, in which case the role models will be available to potentially promote participation. The fiscal argument is not the subject of this article, but is deserving of further consideration elsewhere.

Association Football
In Britain, there are about 10,000 registered players in 146 Centres of Excellence (Pickering, 1996), representing what is arguably the United Kingdom’s largest TI and development program. In Britain, football has a long established TI process, although this process reaches only the section of the population that is involved in regular competition at school or in recreational leagues.

“Since 1994, the talent identification process has moved on a broader basis. With increasingly strong competition, major clubs have been more proactive in their approach, using the new F.A. regulations as an opportunity to attract and develop boys from ages 9 to 16 years.” (Burns, 1996, p.9)

The Bosman ruling has encouraged clubs to develop talent “in house” and try to instill a sense of loyalty in a player. Later, no doubt, the club will look toward securing a long playing contract with the player, so that the club can make a profit if the player decides subsequently to leave.

The TI system used by professional football clubs involves the use of scouts who work for or are attached to the clubs monitoring competitive matches. The criterion typically used assesses players on their techniques, balance, personality and speed. Once players have been identified, they are invited to play for one of the club’s youth teams and to attend their “school of excellence.” “All clubs have differing programs for developing talent; however, general themes, according to Burns (1996), are: “physical conditioning... balance, flexibility, speed, change of pace (and) aerobic endurance.” (p.10). In this way, professional clubs strive to further develop the talent first identified by the scouts.

The United Kingdom already has a well developed, if somewhat subjective, TI scheme. Williams et al. (1998) suggest an alternative, more objective, TI framework, which is summarized in Figure 1. Williams et al. (1998) propose a TI framework by highlighting characteristics that are identified as contributing to success in association football. The above framework is multi-factorial and interdisciplinary, and would involve a comprehensive and time-consuming testing system to operate effectively.

Criticism of the Above Proposed TI Scheme for Association Football in the UK
Although Williams et al. (1998) suggest a comprehensive framework for TI in football, they have not reported suitable tests to examine some of the factors they suggest. This is particularly the case with regard to perceptual-cognitive skills; universally recognized, valid and reliable tests are not identified in existing literature. It also is not clear how the tests are to be scored or prioritized. One potential option is a scoring system similar to the Multi-factorial Inter-disciplinary Talent Identification Scheme (MFIDTIS). Tests in this protocol are instigated as filters, dependent on their ability to discriminate between successful and unsuccessful players. Williams et al. (1998) also do not indicate if there should be any weighting to the characteristics.

This author believes that all the variables are equally influential in the success of players. Therefore, certain characteristics are more important and should theoretically take precedence. This would be reflected in an order or hierarchy of tests. William et al. (1998) do not indicate the perceived relative importance of the characteristics identified. In addition, the author questions whether the tests on “game intelligence,” decision making and anticipation could differentiate between “raw” un-coached talented players, or if less talented, but better coached players would score higher. This would reduce the likelihood of the more talented players being identified through this scheme. If this were to be the case, then this scheme would be unable to identify talented players who do not already compete in football. If so, then this more complex system could be little better than the current, more subjective system. The question of the validity of introducing a more complex, expensive and time consuming program needs further critical analysis.

A Proposed Multi-Disciplinary Talent Identification Scheme for Soccer
In recognition of the need for a multidisciplinary approach to talent identification, Reilly et al. (2000) proposed a multi-factorial interdisciplinary TI scheme for football. The test battery was piloted on 31 boys, about 16 years old. The testing protocol comprised 15 anthropometric measurements, 8 physiological measurements, 3 psychological tests and 2 skill tests. The anthropometric measurements taken were from 13 locations on the body. These measurements established percentage body fat and somatotype. The physiological measurements were estimated max VO2, speed over 5m, 15m, 25m and 30m, agility over a 40m sprint, a speed-endurance and vertica-ljump test. The psychological tests included the Task and Ego Orientation in Sport questionnaire (Duda, 1989), the CSAI-2 (Jones and Swain, 1995) and an anticipation test (Williams and Davids, 1998). Finally, the skill tests were a shooting and slalom test developed by Reilly and Holmes (1983). Statistical analysis of the data from the battery of tests indicated that a combination of four variables successfully discriminated between elite and sub-elite players. These variables were agility, 30m sprint, ego orientation and anticipation. The 30m sprint speed was negatively associated with elite players, making slower sprinters more likely to be elite players than faster sprinters.

Criticisms of the Proposed Multidisciplinary TI Scheme for Soccer
Reilly et al. (2000) do not indicate how they controlled, eliminated or selected the variables for inclusion in the stepwise analysis. In the absence of an explanation or comment regarding the selection process, the possibility that all the variables were entered cannot be discounted. If all the variables were entered into the analysis, the resulting low power ratio would call into question the results of the study. Entry of all the variables would mean that a total of 30 variables were mapped against 31 players. This would give an approximate power ratio of 1:1. This would certainly cast doubt on the validity of the study’s results. This lack of validity may be apparent in the study results, suggesting that slower sprinters are more likely to be elite players than faster sprinters.

Existing literature suggests that the reverse is in fact the case. Faster players are more likely to be elite players than slower ones (Panfil et al., 1997; Janssens et al., 1998).

Critical Review of an Interdisciplinary TI System
Many TI programs are interdisciplinary in nature. They draw mainly from anthropometric data, such as weight and height, and physiological data such as sprint speed. It is the author’s opinion that no talent search program can be effective without a series of psychological tests to ensure that a rounded profile of the player is developed. Previous research has shown the comparability of elite athlete’s psychological profiles.

“Research so far indicates that elite performers are more alike in terms of psychological profile than they are dissimilar.” (Fisher, 1990)

Physical talent may not be enough to ensure success in the international arena. Research as early as Morgan and Johnson (1978) suggests that a combination of approaches should be used to differentiate athletes. Many psychological tests are available. However, as with the physical and anthropometric tests, key factors in their inclusion in any talent identification scheme would be their reliability, validity and ease of testing.

A Review of Talent Identification in Soccer
Williams et al. (2000) reviewed and integrated the main research findings concerned with TI and football. The review highlights anthropometric factors that may help to predict success in football. The authors, however, concur with the statement made by Fisher and Borms (1990):

“Many of the physical qualities that distinguish elite and sub-elite players may not be apparent until late adolescence, confounding the early selection of performers” (Fisher and Borms, 1990 p.27)

In regard to the physiological factors that affect performance, Williams et al. (2000) found that a number of characteristics have been highlighted in the literature as being able to discriminate between successful and non-successful athletes. This may be in part because of a more systematic approach to training before their induction into a specialized under-age squad. The main physiological characteristic identified by Williams et al. (2000) is that of VO2, although they acknowledge that there is a concern in the literature regarding the extent to which this factor “tracks” from childhood to maturity. Williams additionally highlights the higher tempo of the modern game and postulates that this will continue to increase in the future, and higher levels of VO2 will be required from players.

Psychologically, Williams et al. (2000), identify commitment, self-confidence, less anxiety, the ability to use psychological coping strategies and better concentration as predictors of success. The study also reveals that anticipation, decision-making skills and social background as important characteristics of successful players. Typically, players from a white, middle-class and nuclear family have a greater chance of success.

The talent identification review conducted by Williams et al. (2000) is comprehensive. It is not, however, a TI scheme; none of the characteristics identified has a corresponding test associated with them. Ethically, the question must be raised whether it is appropriate to select athletes based upon their social background. In addition, uncertainty remains as to how much weighting should be attributed to social background as criteria for the basis of analysis.

The problems created by adopting an interdisciplinary approach to talent identification are significant. Professional clubs must take into account the problems of analyzing and compensating for the differing maturation rates of their young players. Indeed, it is highly possible that a number of “late developers” could be completely disregarded, when they may well have benefited from a more complete TI process. Hebbelinck (1988) postulates that late maturers may well have an advantage over their early maturing contemporaries, in that they may well work on the skill aspects of their sport in order to compensate for any identified anthropometric or physiological “disadvantage.” Several researchers (Salmela, 1996; Woodman, 1994) also suggest that the identification process should be continued for several years because the “normal” physical and motor development of young people does not necessarily follow a strict chronological pattern, and the accuracy of such identification improves with age. Children not initially identified as being talented would need to be encouraged to stay involved in sport, to allow them to be regularly retested. This would go some way to counteracting the problems associated with not identifying late maturers as talented athletes.

Balyi, I. and Hamilton, A. (1995) “The concept of long term athlete development.” Strength and Conditioning Coach Vol. 3, No. 2.
Ericsson, K.A. (1996). The acquisition of expert performance: An introduction to some of the issues. In The road to excellence: The acquisition of expert performance in the arts and sciences, sports and games (edited by A.Starkes and K.A. Ericsson), pp. 1-50. Mahwah, NJ: Lawrence Erlbaum Associates
Fisher, R.J. and Borms, J. (1990) The Search for Sporting Excellence. Sports Studies 3. International Council of Sports Science and Physical Education. Germany: Karl Hoffman.
Grossarth-Maticek, R., Eysenck, H.J., Rieder, H., and Rakic, L. (1990). Psychological factors as determinants of success in football and boxing: the effects of behaviour therapy. International Journal of Sport Psychology, 21, 237-255.
Janssens, M., Van Renterghem, B., Bourgois, J. and Vrijens, J. (1998) Physical fitness and specific motor performance of young soccer players aged 11-12 years. Communication to the European College of Sport Science. Journal of Sports Sciences, 16, 434-435.
Mahoney, M.J., Gabriel, T.J., & Perkins, T.S. (1987). Psychological skills and exceptional athletic performance. The Sport Psychologist, 1, 181-199.
May, J.R., Veach, T.L., Reed, M.W., and Griffey, M.S. (1985). A psychological study of health, injury and performance in athletes on the US alpine ski team. Physician and Sportsmedicine, 13, 111-115.
Meyers, M.C., LeUnes, A., and Bourgeois, A.E. (1996). Psychological skills assessment and athletic performance in collegiate rodeo athletes. Journal of Sport Behavior, 19, 132-145.
Morgan, W.P. (1979). Prediction of performance in athletics. In Coach, Athlete and the Sport Psychologist (edited by P. Klavora and J.V. Daniel), pp. 172-186. Champaign, IL: Human Kinetics.
Morgan, W.P. and Pollock, M.L. (1977). Psychologic characterization of the elite distance runner. Annals of the New York Academy Sciences, 301, 382-403.
Morris, T. (2000). Psychological characteristics and talent identification in soccer. Journal of Sports Sciences, 18, 715-726.
Panfil, R., Naglak, Z., Bober, T. and Zaton, E.W.M. (1997). Searching and developing talents in soccer: A year of experience. In Proceedings of the 2nd Annual Congress of the European College of Sports Science (edited by J.Bangsbo, B.Saltin, H.Bonde, Y.Hellsten, B.Ibsen, M.Kjaer and G.Sjogaard), pp. 649-650. Copenhagen: HO+Storm.
Prescott, J. (1996) Talent identification and development in female gymnastics.Coaching Focus, 31, 12-13
Régnier, G., Salmela, J.H., and Russell, S.J. (1993). Talent detection and development in sport. In A Handbook of Research on Sports Psychology (edited by R. Singer, M. Murphey, and L.K. Tennant), pp. 290-313. New York: Macmillan.
Reilly, T., Williams, A.M., Nevill, A., and Franks, A. (2000). A multidisciplinary approach to talent identification in soccer. Journal of Sports Sciences, 18, 695-702.
Salmela, J.H. (1996) Expert coaches’ strategies for the development of expert athletes. In Current Research in Sports Sciences (edited by V.A.Rogozkin and R.Maughan), pp.5-19. New York: Plenum Press
Williams, A.M. and Reilly, T. (2000a). Searching for the Stars. Special issue of the Journal of Sports Sciences, 18, 655-775.


    Title Talent Identification in Soccer.
    Source Soccer Journal
    Vol (iss) Vol. 55 Issue 1
    Date Jan/Feb2010
    Page p.24-28


This material has been copied under license from the Publisher. Any resale for profit or further copying is strictly prohibited.