Where do we intervene to optimize sports systems? Leverage Points the way

ABSTRACT

Sport and sports research are inherently complex systems. This appears to be somewhat at odds with the current research paradigm in sport in which interventions are aimed are fixing or solving singular broken components within the system. In any complex system, such as sport, there are places where we can intervene to change behaviour and, ideally, system outcomes. Meadows influential work describes 12 different points with which to intervene in complex systems (termed “Leverage Points”), which are ordered from shallow to deeper based on their potential effectiveness to influence transformational change. Whether research in sport is aimed at shallow or deeper Leverage Points is unknown. This study aimed to assess highly impactful research in sports science, sports nutrition/metabolism, sports medicine, sport and exercise psychology, sports management, motor control, sports biomechanics and sports policy/law through a Leverage Points lens. The 10 most highly cited original-research manuscripts from each journal representing these fields were analysed for the Leverage Point with which the intervention described in the manuscript was focused. The results indicate that highly impactful research in sports science, sports nutrition/metabolism, sports biomechanics and sports medicine is predominantly focused at the shallow end of the Leverage Points hierarchy. Conversely, the interventions drawn from journals representing sports management and sports policy/law were focused on the deeper end. Other journals analysed had a mixed profile. Explanations for these findings include the dual practitioner/academic needing to “think fast” to solve immediate questions in sports science/medicine/nutrition, limited engagement with “working slow” systems and method experts and differences in incremental vs. non-incremental research strategies

KEYWORDS:

• Systems thinking
• meta-science
• human factors

Introduction

The broad aim of sports research is to understand and enhance the performance, health and well-being of coaches, athletes, clubs, leagues and organisations (McLean, Kerhervé, et al., 2021; McLean, Rath, et al., 2021). With the increased participation and interest in organised sport throughout the 21^st century has come a proliferation of academic research attempting to understand every aspect of sport and its broader societal influence. This has led to a myriad of changes and improvements in sport. An example of improvements to athletes is the male marathon world record, which has improved from 2:32:35 in 1920 to 2:01:09 in September of 2022 as the cumulative result of improvements in training, nutrition, shoe technology, running surface technology and athlete psychology (amongst other factors) (Goss et al., 2022; Hoogkamer et al., 2017). For injury prevention, implementing the FIFA 11+ injury prevention programme in soccer (football) has been found to be associated with a ~30% decrease in injury risk (Sadigursky et al., 2017). An example of an organizational improvement is the formulation of the World Anti-Doping Agency (WADA), a hybrid organization of national governments and the Olympic movement which was created with a core mission to harmonize anti-doping policy globally (D. Read et al., 2020).

Sport and sport research are complex systems. This inherent complexity can be elusive and difficult to define. Nonetheless, complex systems typically contain structurally and functionally different components which interact in a non-linear, dynamic manner to produce emergent behaviours (Cilliers, 2002; Salmon & McLean, 2020). Other notable features of complex systems are that they are open, comprise components that are ignorant of the behaviour of the broader system and have a history or path dependence (Dekker, 2016; Salmon & McLean, 2020). Complex systems can exist within sport, depending on where the boundary conditions are drawn, and these can include matches, teams, sporting leagues, major events (e.g., the Olympics) and other entities (e.g., WADA, National Sporting Organisations [NSOs]). Understanding complex systems through a “systems thinking” lens is crucial as it can lead to innovations that are not possible through typical reductionist, compartmentalized scientific approaches (Liu et al., 2015).

In any complex system, such as sport, there are places where we can intervene in an attempt to change behaviour and, ideally, system outcomes. This is at odds with the current paradigm in sport in which interventions are aimed are fixing or solving singular broken components within the system (e.g., the athlete, the equipment, the resources, etc.). Meadows (1997) seminal work in this area neatly describes 12 different points with which to intervene in complex systems (termed “Leverage Points”). Leverage Points are ordered into a hierarchy based on their potential effectiveness to influence transformational change. Deeper Leverage Points (e.g., the power to change mindsets/mental models and the structure of the system) have a stronger influence on system behaviour and performance, and shallow Leverage Points (e.g., system numbers, parameters, constants) have less impact (Meadows, 1997). Consider a football club attempting to develop a team to become dominant in the league with the overall aim to have sustained success, e.g., to win multiple championships. For this team, the impact of improving shooting percentage or running speed (i.e., parameters) will have a limited ability to influence transformational change towards this goal in comparison with changing the clubs’ policies to recruit and develop youth prospects to have a consistent pipeline of high-quality first-team players, rather than relying on expensive player transfers. Implicit within the Leverage Points hierarchy is the understanding that influencing the deeper points subsequently alters each of the shallower points. Building off of Meadow’s Leverage Points, Abson et al. (2017) suggested these points could be overlayed with four system “Realms of Leverage” (arranged from deeper to shallow): 1) mental models, 2) system design/structure, 3) feedback loops and 4) parameters. Collectively, these represent the hierarchy of Leverage Points. The Leverage Points, the broad Realms of Leverage that they cover, and examples of these from sport are described in Table 1.

Table 1. Different examples from research in sport of interventions and where those examples sit within Meadow’s Leverage Points (Meadows, 1997) and Realms of Leverage (Abson et al., 2017).

Depth	Leverage Points (LP) (Meadows, 1997)	Realms of Leverage (Abson et al., 2017)	Examples from sport
Shallow Leverage Points	LP1 – Constants, parameters, numbers	Parameters – the mechanistic, immediate characteristics of the system.	Distance, mass, time, height, force, power, players, coaches, teams, fans, wins, losses, points, income, expenditure, profit, balance sheets, transfer fees and contracts.
	LP2 – Buffer sizes
	LP3 – Material stocks and flows
	LP4 – Relative delays LP5 – Negative feedback loops LP6 – Positive feedback loops	Feedback loops – the interactions between elements within a system of interest.	Training induced fitness changes, training induced fatigue changes, absorption of nutrients, change in ticket price, introducing new funding models, negotiation of a collective bargaining agreement, coaching of athletes to improve their performance, investment in new infrastructure (e.g., stadia, buildings), change in government, promotion/relegation and awarding the Olympics host city/country.
Deep Leverage Points	LP7 – Information flows LP8 – Rules of the system LP9 – Self-organising system structures	System design/structure – the structures and rules in place that manage the feedback loops and parameters.	Introduction of WADA, creation of a National Institute of Sport (e.g., AIS, EIS), case law precedents, introduction of the offside rule, Bosman ruling, European labour laws, team ownership rules, Fit-and-proper-person-test, introduction of the Court of Arbitration for Sport and development of new tournament series (e.g., LIV golf tour).
	LP10 – Goal of the system LP11 – Mindset/paradigm LP12 – The power to transcend paradigms	Mental models – the underpinning values, goals and world views of actors within the system which shape the emergent direction to which the system is orientated.	Focusing a club on youth development to elicit transfer fees in order to fund club development, investment in sport to improve a national and/or organisational reputation (i.e., “sportswashing”), shifting the focus from high performance sport to improving public health through sport.

AIS – National Institute of Sport, EIS – English Institute of Sport, WADA – World Anti-Doping Agency.

There are diverse paradigms with which interventions can be drawn from for research. For example, experimental interventions are typically aimed at manipulating one variable (i.e., the independent variable) whilst keeping others constant (i.e., the dependent variable[s]) (Schenker & Rumrill, 2004). In this way, causality can be assumed as the intervention is focused on a single variable which is being manipulated, and the difference between the outcome and the comparator (i.e., group, pre-intervention, counterfactual) can be used to examine the existence of an effect and size of the intervention effect. Other research interventions can be assessed by quantitative or qualitative research, or through mixed method qualitative and quantitative evidence for the intervention of interest (termed “mixed methods”) (Arghode, 2012). In sport, interventions can come in different forms, from the application of new technologies, application of existing technologies to new problem spaces, new athlete training interventions, rule changes, changes in funding models and changes in laws (Branch, 2003; Cermak et al., 2012; Gordon, 2009; Lockie et al., 2012; Meir et al., 2001; Stewart, 2017).

Effective and lasting transformation within complex systems, such as sport, requires that interventions be aimed at deeper Leverage Points. Naturally, this will depend on the focus of the intervention and goals of the system writ large. For instance, if the goal is improving the running times of an athlete competing in the 100 m sprint, changing the content of training to improve technique (i.e., a shallow Leverage Point) would naturally be a focus of the intervention. This therefore represents a match between the intervention Leverage Point and the goal of the system. Conversely, recent longitudinal research in sport has identified that hamstring injury rates in elite men’s football have been increasing such that the number of injuries and missed training days have doubled over the course of the last two decades (Ekstrand et al., 2022). This has occurred despite significant efforts and investment aimed at improving rates of injury occurrence and reoccurrence through improved training (e.g., eccentric hamstring strengthening) and technological (e.g., NordBord) advances (Oakley et al., 2018; Opar et al., 2013; Van Dyk et al., 2019). From a systems perspective, this could be occurring as there is a mismatch between the focus of the interventions and the goals of the system such that the other factors which influence system performance and behaviour remain unaccounted for. To positively address these could require a shift to focusing on deeper Leverage Points (i.e., the system design/structure and/or mental models).

The current focus of the interventions used in sport appears to be aimed at fixing isolated components within a system, without consideration for the focus of the Leverage Point within the system. The aim of this study was therefore to examine interventions in the sport literature through the Leverage Points and Realms of Leverage frameworks. Specifically, the aim was to determine what Leverage Points and Realms of Leverage have been targeted by previous interventions described in the highly cited sports science, sport and exercise psychology, sports nutrition, sports medicine, sport management, sports law/policy, sports biomechanics and motor control literature. These were selected as they are broadly cover each aspect of sport within their scope and are recognised fields or sub-fields within the Australian Bureau of Statistics categorization of Fields of Research (FoR) codes (https://www.abs.gov.au/ausstats/[email protected]/0/6bb427ab9696c225ca2574180004463e).

Materials and methods

Journals were selected to represent sports science (Journal of Sports Sciences [JSS]), sports psychology (International Journal of Sport and Exercise Psychology [IJSEP]), sports nutrition (International Journal of Sport Nutrition and Exercise Metabolism [IJSNEM]), sports medicine (British Journal of Sports Medicine [BJSM]), sport management (Journal of Sport Management [JSM]), sports law/policy (International Journal of Sport Policy and Politics [IJSPP]), sports biomechanics (Sports Biomechanics [SB]) and motor control (Motor Control [MC]). These journals are placed either in Q1 or occupy the highest placing in their respective fields based on impact factor. The individual journal aims/scope/mission are outlined in Supplementary File 1.

To assess the focus of interventions in these journals, an assessment of their location on the Leverage Points and Realms of Leverage scale, was undertaken. Briefly, for each intervention, the focus of the intervention and its contents were identified from the full-text and compared against the criteria as outlined originally by Meadows (https://donellameadows.org/archives/leverage-points-places-to-intervene-in-a-system/) to identify the Leverage Point that the intervention was focused at influencing. If, for example, an intervention was aimed at influencing an athlete hitting power by changing the weight of the baseball bat or ball, it is aimed at the constants, parameters and numbers level of the Leverage Points hierarchy which equates to the parameters level of the Realms of Leverage Hierarchy. Conversely, if the intervention was aimed at influencing players beliefs in fair play and sportsmanship, it is aimed at the mindset/paradigm (Leverage Points) and mental model (Realms of Leverage) levels. Each Leverage Point and corresponding Realms of Leverage category are outlined in further detail with examples in Table 1.

Firstly, the top 10 articles based on their all-time citation count were downloaded for each journal listed above. Their citation frequency at the date of access (14 Febuary 2024) was also determined based on the stated citation counts from the Web of Science database. Whilst citation counts are not a proxy for the quality of a scientific article, citation counts over time (and their related metrics) are often used as a measure of scientific impact in a given field (Purkayastha et al., 2019). As the point of intervention was of interest, these articles included only original research, with reviews (including systematic, narrative, and/or scoping reviews) excluded.

Following the acquisition of the articles, details (e.g., authors, title, year of publication, citations [count]) of each article were recorded into a custom spreadsheet. Each article was then assessed by 1 co-investigator (MN) to determine which Leverage Point and accompanying Realm of Leverage were targeted by the intervention. If an article examined interventions at multiple Leverage Points, it was included for analysis at the deepest Leverage Point that was examined. Concurrently, a second co-investigator (SM) undertook the same process independently on a selected subset of 25% (n = 18) of the articles which were randomly selected. To achieve this, each article was given a number 1–80 and a random number generator (random.org) was used until 18 individual numbers (representing the selected articles) were drawn. Following this, the co-investigator records were compared, and any disagreements were adjudicated by a third co-author (PS). The co-investigator team of analysts includes researchers with who have published over 200 peer reviewed research articles, many of which apply systems thinking and complex systems analysis methods to a variety of different research domains including to sport, road transport safety, autonomous vehicles and healthcare, among others.

Results

Of the articles that were compared between raters (MN and SM), there was ~89% agreement (16/18), with the 2 disagreed to articles decided by adjudication (PS). The articles which were selected and the Realms of Leverage for each article are presented in Supplementary File 1. Table 2 provides examples from the included studies at the different Leverage Points and accompanying Realms of Leverage.

Table 2. Examples of different interventions from the included studies across the Realms of Leverage (Abson et al., 2017) and Leverage Points (Meadows, 1997) frameworks.

Realms of Leverage (Abson et al., 2017)	Leverage Points (Meadows, 1997)	Example (Reference)
Parameters	Constants, parameters, numbers	Supplement use patterns in young German athletes (Braun et al. 2009).
	Buffer sizes	Muscle strength and size as a buffer against stress and strain during contractions (Hortobágyi et al. 1999).
	Material stocks and flows	The (mis)application of economic impact analyses of sports facilities and events (Crompton, 1995).
Feedback Loops	Relative delays	-
	Negative feedback loops	Methods of weight loss to make weight during weight across a range of weight categorized combat sports (Brito et al., 2012).
	Positive feedback loops	Government investment in sport and the “virtuous” cycle wherein investment can lead to positive societal outcomes Grix & (Carmichael, 2012).
System design/structure	Information flows	Measuring the concept of “legacy” with respect to sporting mega events (e.g., the Olympics) (Preuss, 2019).
	Rules of the system	Examining how the “legacy” of mega sporting events influences subsequent national policy (Grix et al. 2017).
	Self-organising system structures	Inclusive youth sport structure using the 3P model (participation, performance and personal development) (Côté & Hancock, 2016).
Mental models	Goal of the system	Why organisation support community sport and the drivers of support (Miragaia et al. 2017).
	Mindset/paradigm	The underpinning values held for brand associations in sport in 3 characteristics: attribute, benefit and attitude (Gladden & Funk, 2002).
	The power to transcend paradigms	-

The articles were published between 1988 and 2021, with the average (± standard deviation [SD]) number of citations for the included articles 199 ± 77 (ISJNEM), 626 ± 199 (BJSM), 232 ± 41 (JSM), 63 ± 46 (IJSPP), 945 ± 547 (JSS), 55 ± 33 (IJSEP), 87 ± 27 (SB) and 144 ± 50 (MC) at the time of analysis. The frequency of articles coded to each of the Leverage Points is presented in Table 3. Overall, across the articles studied, 48% of the interventions were in the Parameters level, while 24%, 8% and 20% were in the Feedback loops, System design/structure and Mental model levels, respectively, when assessing relative to the Realms of Leverage.

Table 3. The frequency of interventions which were coded to each of the Leverage Points (LP) (Meadows, 1997) and Realms of Leverage (Abson et al., 2017) frameworks from the 10 most cited articles in each of the journals investigated. The largest frequency of coded articles for each of the different journals is shaded in grey.

	Parameters			Feedback loops			System design/structure			Mental models
	LP 1	LP 2	LP 3	LP 4	LP 5	LP 6	LP 7	LP 8	LP 9	LP 10	LP 11	LP 12
British Journal of Sports Medicine (BJSM)	9	–	–	–	–	1	–	–	–	–	–	–
International Journal of Sport and Exercise Psychology (IJSEP)	3	–	–	–	–	–	3	1	–	2	1	–
International Journal of Sport Nutrition and Exercise Metabolism (IJSNEM)	6	–	–	–	1	3	–	–	–	–	–	–
International Journal of Sport Policy and Politics (IJSPP)	–	–	–	–	–	3	1	2	1	3	–	–
Journal of Sport Management (JSM)	–	–	1	–	–	2	–	–	–	–	7	–
Journal of Sports Sciences (JSS)	8	–	–	–	–	2	–	–	–	–	–	–
Motor Control (MC)	4	1	–	–	–	1	3	–	1	–	–	–
Sports Biomechanics (SB)	10	–	–	–	–	–	–	–	–	–	–	–

Discussion

Give me a lever long enough … and I shall move the world – Archimedes

The aim of this investigation was to examine interventions designed to optimise aspects of sport and identify the Leverage Points that have been targeted to date. The findings indicate that highly cited research in journals SM, JSS, SB and IJSNEM predominantly include articles with interventions which are focused at the shallow end of the Leverage Points/Realms of Leverage hierarchy (i.e., parameters and feedback loops). This, therefore, indicates that the majority of research in sports science, sports medicine, and sport and exercise nutrition are focused on the Leverage Points which are likely to be the least effective in influencing broad system behaviour. In these fields, there is clearly room to design and deploy more powerful interventions which focus on deeper Leverage Points when the aim is to facilitate systemic change. Conversely, the articles which were published in JSM and IJSPP were predominantly focused on interventions at deeper Leverage Points (i.e., system design/structure and mental models), while MC and IJSEP had a more mixed profile. Potential explanations for these outcomes are discussed below.

Our analysis identified that journals which focus on sports management (JSM), and sports law/policy (IJSPP) examined interventions primarily at the deeper Leverage Points of the system. Examples of these included the “legacy” of major sporting events (Grix et al., 2017), the policies and practice of athletes in the university sector (Aquilina & Henry, 2010) and relationships between brand image and fan loyalty (Bauer et al., 2008). Conceptually, this is likely to be indicative that in these domains, the focus of interventions is inherently at influencing the system structure and collective mental models (i.e., the deeper levels of the system). As noted by Meadows (1997), to influence effective and lasting change, it is necessary to ensure that interventions are aimed at the deeper Leverage Points of the system. Therefore, when designing interventions with the aim of stimulating large-scale change, it is pertinent to include experts who have knowledge across various fields of research (i.e., relevant subject matter experts). Further, inclusion of systems thinking experts, and collaboration between these experts and subject matter experts in the various research fields in sport is necessary if research translation is to lead to implementation “at the coal face”. This collaboration and the use of systems thinking should assist to reducing the research to practice translation gap (McLean, Kerhervé, et al., 2021).

Interventions analysed from BJSM, JSS, SB and IJSNEM articles were primarily focused at the shallower end of the leverage hierarchy (Table 1). Examples of these articles include nitrate supplementation for exercise performance (Cermak et al., 2012), weight loss in combat sports (Brito et al., 2012) and running performance in professional soccer (Bradley et al., 2009). This highlights that, for these journals which focus on sports medicine (BJSM), sports science (JSS), sports biomechanics (SB) and sports nutrition/metabolism (IJSNEM), the intervention focus is typically shallow and fixing of individual system components with a consequent limited ability to effect lasting change on the broader system. This may be indicative of an inability to design deeper and more impactful interventions at the higher Leverage Points in these fields of research, which are prone to historical biases in the values, theories and approaches that have been adopted (McLean, Kerhervé, et al., 2021). An alternative explanation for this finding is that researchers who typically publish in these journals are collecting data locally with the sporting team/environment or working directly with athletes/players who become the unit of analysis. Indeed, often individuals will fill roles as both servicing practitioner (e.g., sports scientist, sports dietician, high-performance manager) and academic (i.e., the “pracademic”) to facilitate research (Collins & Collins, 2019). In this way, pracademics use research to answer questions which are perceived to have direct importance to the athlete(s), coach(s) and/or team, they are working with. Conceptually, this requires that they “work fast” to solve questions which have a direct and more immediate application (Coutts, 2016). Alternatively, interventions which focus at deeper Leverage Points may require a “working slow” approach which is more thoughtful, considered and indirect (Coutts, 2016). When designing interventions, integrating methods experts who have the experience and time to “work slow” as well as to bring new theories and perspectives which may provide practitioners with a deeper ability to explore larger, richer and more complex, systems-based questions.

This analysis examined interventions through a leverage hierarchy paradigm which was predicated on examining the direct influence of the applied intervention. An alternative perspective is that for fields whereby interventions were focused at the shallow Leverage Points (noted above), change occurs in response to the incremental accumulation of evidence over time. In these fields, it might be necessary for evidence accumulation at shallow Leverage Points to reach a critical threshold before it influences the deeper Leverage Points of the system structure and shared mental models. The mixed Leverage Point profile of MC (motor control) and IJSEP (sports and exercise psychology) may be evidence of a shift akin to this having occurred. Importantly, incremental advances in science allow for continued scientific progress whilst minimizing the potential downside risks of non-incremental strategies (e.g., “moonshots”) that can be prone to failure (Foster et al., 2015). A tangible example from the sport science research literature is the introduction of a novel ergogenic supplement in nitrate-rich beetroot juice. Following early animal model research, the supplement was studied in humans which identified, over successive studies, the there was evidence to suggest it improves endurance performance (i.e., power, time trial time [the parameters level]) (Bailey et al., 2009; Wylie et al., 2013). After sufficient studies suggest a benefit to performance and the circumstances in which that benefit exists, the system information flows and rules (i.e., the system structure) shift and the shared mental models around the supplement also change. More broadly, this threshold is likely to be fuzzy and dependent on factors such as the strength of the evidence, the research designs employed, the context of the research and the feasibility of application (Close et al., 2019).

The above example is illustrative of the potential for interventions to influence outcomes in areas other than those originally intended. Crucially, this highlights that second- and third-order effects can occur elsewhere within the system from wherein the Leverage Points hierarchy the original intervention is focused. Such observations emphasise the non-linear aspects of sport as a complex system in which behaviour is unpredictable and migratory over time (Cilliers, 2002). The analysis of the findings of the present study could suggest that there is a latent journal archetype present. Accordingly, journals are either focused at 1) directly influencing the system through interventions which are initially targeted at deeper Leverage points or 2) on the accumulation of evidence over time with research interventions which are targeted at more shallow Leverage points. Whether this is specific to research in sport or is more broadly attributable is unclear. We believe this is a novel observation which warrants further analysis and discussion in the literature. Examining larger datasets encompassing a broader range of journals and larger number of studies would be necessary to examine this. For example, an analysis approach such as natural language processing (Dessi et al., 2021) could be used to automate the extraction and coding of the necessary information regarding the focus of the interventions to the different Leverage Points or Realms of Leverage. This would enable the identification of the presence of this latent journal archetype, as well as changes across time and across a larger sample of journals representing certain fields (e.g., sports science, sports nutrition, sports medicine).

Often overlooked considerations with the design of interventions in sport research is incorporating the multiple stakeholders throughout a system and the common mistakes which can lead to negative unintended consequences. Within a sporting team, for example, there are stakeholders at various levels within the system hierarchy (e.g., players, coaches, staff, management, board of directors). These actors complete various tasks or functions as part of their contribution to achieving common goals (e.g., winning games/championships, sustainably improving team performance) (McLean, Rath, et al., 2021). Whilst these stakeholder contributions to these overall goals can be more (e.g., players and team performance, board of directors making funding decisions) or less direct (e.g., cleaning staff improving the training environment), they all contribute as performance and functioning in a complex system is an emergent property of the interactions between the systems components and actors (McLean, Rath, et al., 2021; Salmon & McLean, 2020). Therefore, it is necessary to include actors across the system when designing and implementing interventions in sport research. Achieving this at deep Leverage Points requires both sophisticated analysis (e.g., Causal Loop Diagrams) (McLean, Kerhervé, et al., 2021) and intervention design processes (e.g., the Sociotechnical Systems Design Toolkit [STS-DT]) (G. J. Read et al., 2018).

This then leads to the need to consider or imagine what interventions at deeper Leverage Points might encompass. An example which focuses on shifting the goals of the system and the paradigm out of which the system arises can be found in anti-doping. Current strategies for anti-doping are typically focused on the individual engaging in doping under the “strict liability” legal standard, and on engineering of new tests (i.e., detection), anti-doping education of athletes and coaches (i.e., deterrence) and stricter sanctions for those detected (i.e., punishment) (Dimeo & Møller, 2018). Importantly, such interventions are at the shallow end of the Leverage Point hierarchy. In anti-doping, previous research has suggested a move from the current detection-deterrence-punishment paradigm, which resembles the broader societal discourse around the “war on drugs” to one which focuses on athlete health and harm reduction is warranted (Kayser & Broers, 2012). This is due to the lack of effectiveness of the current paradigm (as evidenced by continually high rates of admitted doping), the encroachment on athlete human rights and public health principles, and the acceptance of many prohibited substances in broader society (Kayser & Broers, 2012; Kayser & Smith, 2008). One suggestion has been to shift to a harm reduction (or minimization) model which may allow for the supervised use of performance enhancing drugs provided the health risk criterion in the current anti-doping framework is not breached (Kayser, 2018). Such a paradigmatic change would need to be accepted by the raft of stakeholders (including athletes, coaches, administrators and organisations) and may not be permissible to those in positions of accountability and influence within sport. This likely resistance to change makes such Leverage Points difficult to access.

In sport, one of the more common mistakes that is observed is the “Fixes That Fail” scenario where an intervention leads to a negative unintended outcome. An example of this is the removal of the coach when the team experiences a negative change in their results or worsening team performance (McLean et al., 2019). By removing the coach, there is perceived to be a temporary performance improvement which reinforces that the correct decision has been made and the issue has been alleviated. However, if improved team performance occurs it often regresses towards the prior mean performance, with research indicated that changing coaches mid-season often leads to either no positive performance improvement or to negative performance change across a season (de Dios Tena & Forrest, 2007; Flores et al., 2012). This is due to the underlying system structure and the mental models of the organisation including the board members, fans, sponsors and other stakeholders remaining the same throughout. Breaking this cycle of “Fixes That Fail” requires acknowledging that change is likely to result in only short-term improvement as this is at a shallow Leverage Point (McLean et al., 2019).

This study is not without limitations. Firstly, the most cited articles in each journal may not be representative of shifts in research designs to studying deeper Leverage Points that may have occurred in more recent times. Whilst it is true that systems thinking and systems analysis is gaining in traction within research in sport (McLean, Kerhervé, et al., 2021; Salmon & McLean, 2020), there is currently no evidence to suggest that this work is being cited more often to other more reductionist approaches to research in sport, which are common. Nonetheless, our analysis did not account for this in the search strategy. Secondly, while citation counts (and associated metrics) can provide insight into the impact of a paper in a research field, they are imprecise and using these as a surrogate for quality is potentially problematic. They also provide little evidence on how research is being used or translated into practice which is, arguably, the end goal when assessing the impact of research. Finally, as noted above, the included studies were assessed for the direct focus of their interventions. Therefore, the second- and third-order effects of those interventions at other levels of the system were not considered in the analysis. This may have resulted in any assessment of such interventions to be shortsighted, and it is worthwhile to consider these effects in future analyses. Future research can incorporate understanding of the Leverage Points and the Realms of Leverage hierarchy when designing interventions to maximise impact in sport.

We believe this analysis is the first to introduce the concepts of Leverage Points and Realms of Leverage to examine research interventions across different fields of research in sport. The results of the analysis of the most highly cited research in sport indicate that journals which focus on sports medicine, sports science, sports biomechanics and sports nutrition/exercise metabolism predominantly rely on interventions which are at the shallow end of the Leverage Points/Realms of Leverage hierarchy (e.g., parameters, feedback loops). Conversely, journals which focus on sports management and sports law/policy were associated with interventions at deeper Leverage Points (e.g., system design/structure, mental models). Journals representing sport and exercise psychology and motor control had a mixed Leverage Point profile. Overall, to achieve transformational change and maximal impact, interventions should be designed to address deeper points within the system. Integrating system thinking experts into the design of interventions and collaborations between these experts and subject matter experts in sport is recommended to achieve robust system behaviour change.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online https://doi.org/10.1080/02640414.2024.2352681

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.