• Tidak ada hasil yang ditemukan

We can also see the agency of numbers at work in the context of physical ability testing. The overall concept of physical ability testing, within the talent identification context, is that an athlete is tested in a variety of ways and then the results of the testing are converted into metrics of some form, whether an absolute score or a ranking, with the metrics then used as a method to determine the talent level of the athlete (Kerr 2018; Lidor et  al. 2009). Sports scientists, coaches or others analysing the testing

The Agency of Numbers: The Role of Metrics in Influencing…

110

results receive quantitative accounts of replicable assessments of skills or abilities, which are used to determine the athlete’s level of talent. Thus, even more than competitive performances which are at least recorded and televised, the results of the tests come to represent sporting talent. In many talent identification scenarios, the quantification is seen as final and objective, and athletes have no opportunity to negotiate the assess- ment of their results. In this context, the metric is expected to act as an immutable mobile (Latour 1987, 1990), by again becoming a way for the details of the movements performed at the test to be distributed in a sta- ble form. It also provides the mechanism to territorialise what can be a highly diverse set of abilities conducted outside the actual practice of a given sport.

There is a large amount of literature critiquing the efficacy of these kinds of quantitative talent identification methods (e.g. see Anshel and Lidor 2012; Collins and Cruickshank 2017; Lidor et al. 2009); but here, we suggest that the overall difficulty lies in the assumption that talent identification can be reduced to a numeric form. Studies have found a very large number of factors that contribute to sporting success, includ- ing physical, physiological, psychological and sociological factors (Lidor et al. 2009). Additionally, these vary over time due to athletes’ matura- tion (Malina et al. 2004) and are complicated by the role of deliberate practice (Ericsson et al. 1993), coaching and other environmental con- cerns. Given this high level of complexity, there are substantial challenges in determining which metrics best represent the relevant abilities and skills, which metrics might sufficiently predict the abilities of an indi- vidual and therefore which metrics merit data collection. This complexity would appear to belie attempts to reduce physical ability to numbers.

However, the necessity of metrics in this context is usually understood, or justified, on the grounds that sports governing bodies have limited funds and so must identify the athletes to support in order to achieve elite sporting success (Grix and Carmichael 2012). However, as Sam and Macris (2014) argued in a wider national policy context, these kinds of funding practices can lead to organisations undertaking behaviours designed to ensure that athletes or organisational bodies are scored with the appropriate metrics, as opposed to producing the best outcome for the country. For example, in the talent identification environment, it

R. Kerr et al.

111

could easily lead to athletes training purely to achieve good scores on the talent identification tests which do not necessarily translate to successful results in elite sport more generally (Muller 2018). Nonetheless, metrics continue to be used to both select and de-select athletes for many elite teams and programmes due to the belief that they are fair, objective and unbiased measures (Collins and Bailey 2013; Kerr 2018).

In the very different, professionalised context of the NFL draft, the process of selecting athletes for teams has become a sporting event of its own. For many observers, a key moment in this process is the League- hosted ‘combine’, where invited athletes are measured across a suite of performance indicators, including speed (e.g. 40 and 100  m sprints), agility (shuttle runs) and strength (bench press, standing jumps) as well as position-specific skills (throwing, catching). The combine also includes the Wonderlic intelligence test to estimate cognitive ability. The value of the combine lies in the ability for evaluators to assess measurable skills in a more objective manner, given that all participants are facing the same conditions and without the influence of varying levels of competition that are characteristic of the collegiate game. Moreover, as in the case of the 100 m sprint, it is possible to set performance criteria that indicate NFL potential relevant to positional skillsets.

Within the context of the NFL, the success (or failure) of individual players can have significant consequences for the team, both in terms of winning games and in relation to the financial implications of a player’s relative location in the draft.3 Furthermore, draft selections can have a substantial impact (positive for good picks, extremely negative for ‘busts’) on a team’s public standing. As a result, each team employs a dedicated scouting staff in addition to the coaches and senior management to con- tribute to the process of selecting future players on potential. This staff follows the on-field performance and training habits of players at the collegiate level developing an assessment informed by attendance at live games, extensive review of tape and interaction with collegiate level coaching staffs. Despite this extensive research, concerns regarding the risks and rewards of player selection reinforce the value of the combine.4 The relative performance of the participants at the event can influence not only draft preferences but also decisions regarding the desirability of current team members and available free agents.

The Agency of Numbers: The Role of Metrics in Influencing…

112

For both the casual and more interested fan, the combine results draw much attention by providing an insight into the evaluation process. This is particularly true given that access to the results for individuals, let alone the actual performance, has recently been publicly accessible. In this con- text, the importance of combine performance to the ranking of players on each team’s draft board is likely overstated. It is possible for the value of an individual athlete to drop as a result of poor performance—for example, the precipitous fall of Orlando Brown, an offensive lineman, due to slow timings in sprint and agility tests as well as a lower than expected number of repetitions in the bench lift.5 Similarly, a less vaunted prospect can overcome relative obscurity through a good performance—

as evident in the increased recognition of NFL potential attributed to Shaquem Griffin whose top of category numbers among linebackers helped overcome concerns about a missing hand.6 Such examples are, however, more the exceptional cases that challenge expectations devel- oped through more traditional scouting or other preconceptions based on physical characteristics such as size and weight. Thus, such situations do not limit discussions in media and among fans that definitively peg the participants’ performance in the combine to their expected NFL potential.

The expectations that potential can be forecast based on the accumu- lated metrics—or that the combine results act as an immutable mobile—

are increasingly questioned (Lyons et  al. 2011). Despite the increased awareness of player capabilities associated with the publication of com- bine results, the draft remains a risky process littered with the proverbial busts and the unrecognised late-round stars. Furthermore, the draft selec- tions of individual teams seldom correspond directly to the ranking of prospects based on the combine results. Rather, increasingly reference is made to such factors as leadership potential, willingness to learn and self- confidence. Assessment of such factors is once again shrouded in the activities and expertise of specialist scouts who are in the employ of spe- cific teams. Thus, metrics may still perform as an immutable mobile, but these are most likely to exert significant influence (beyond the more qual- itative and subjective assessments of players in context) when exposing extremely high or low abilities. In this role, the metrics act to alert

R. Kerr et al.

113

interested teams to additional risk or potential that may be obscured through other assessment methods.