Evidence-Based Practice in Sport Science: Integrating Research, Experience, and Athlete Context

Prerequisites: This article assumes familiarity with training load concepts and the sport scientist’s role within a multidisciplinary team. If any of these topics are new to you, start with:

• External and Internal Load: Concepts and Monitoring Fundamentals
• The Sport Scientist as Decision Facilitator in the High-Performance Unit

Learning Objectives

• Define evidence-based practice (EBP) and explain its three pillars: scientific research, coaching expertise, and athlete values.
• Understand the limitations of applying scientific research in the field and evaluate evidence through critical thinking.
• Acknowledge the value of experiential knowledge while recognising the risks of cognitive biases such as confirmation bias and recall bias.
• Understand how to integrate athlete individual context — characteristics, preferences, and goals — into the decision-making process.
• Explain the cyclical relationship between EBP and practice-based evidence (PBE), and describe the organisational frameworks that enable integration.

---

What Is Evidence-Based Practice?

Evidence-based practice (EBP) is the integration of coaching expertise, athlete values, and the best available scientific evidence into the decision-making processes of daily service delivery (French, 2022). The concept originates from evidence-based medicine but has been adapted to the performance sport context, where decisions must account for the complexity and individuality of athletic performance.

EBP rests on three pillars. The first is scientific research: peer-reviewed findings that describe physiological mechanisms, training effects, and injury risk factors. The second is practitioner experiential knowledge: the accumulated insights that coaches, sport scientists, and medical staff develop through years of applied work. The third is athlete values: the individual characteristics, preferences, goals, and circumstances that shape how each athlete responds to and engages with the performance programme. No single pillar is sufficient on its own. A decision based solely on research may ignore athlete individuality. A decision driven only by experience may perpetuate outdated beliefs. A decision shaped only by athlete preference may lack scientific grounding.

This principle becomes concrete in the field through differential diagnosis, a process where data is used to distinguish the surface-level symptom from the actual cause of a performance problem. Consider a scenario where a coach identifies a player as lacking agility. Reactive agility testing and video analysis may reveal that the deficit is not physical — the player reads the game slowly and reacts late to environmental cues (Brewer, 2022). Without integrating objective data and experiential observation, the intervention would target the wrong problem entirely. Agility drills would be prescribed when perceptual-cognitive training was needed. The differential diagnosis process exemplifies what happens when EBP functions well: research provides the testing protocol, experience identifies the initial concern, and athlete-level data reveals the true cause.

EBP is not a fixed formula that produces a single correct answer. It is an iterative process with five stages: identify a relevant performance question, critically appraise the available evidence, develop a strategy to implement the best evidence in the field, evaluate the effectiveness of the new practice, and continuously re-evaluate the evidence base (French, 2022). Each cycle refines both the practitioner’s understanding and the quality of the intervention. The process never truly ends — new evidence, new athletes, and new contexts demand continuous re-evaluation.

Why Research Doesn’t Always Apply

Scientific research provides the foundation of EBP, but applying it in elite sport is not straightforward. Most published studies are conducted with sub-elite populations whose genetic, anatomical, physiological, and biomechanical characteristics differ from those of elite athletes (Brewer, 2022). Research designs also tend to be reductionist — isolating a single variable while controlling others — which poorly reflects the multifactorial nature of sport performance. A periodisation study that manipulates only training volume in recreationally active participants, for example, cannot capture the complexity of managing an elite squad through a 38-week competitive season with congested fixtures, travel, and individual recovery demands.

The motivational and psychosocial differences between elite and sub-elite athletes add a further layer of complexity. Elite athletes operate under unique competitive pressures, contractual obligations, and performance expectations that fundamentally alter how they respond to training interventions. Findings from controlled laboratory settings with university students do not automatically transfer to this environment (Brewer, 2022).

This disconnect is reflected in how practitioners actually source their knowledge. A survey of 206 senior football practitioners across six FIFA confederations found that scientific literature was the least preferred evidence source regardless of department, context, or competition level (Dello Iacono et al., 2025). Practitioners relied more heavily on professional industry communities and internal projects. This finding does not mean research is irrelevant — it means research requires deliberate filtering before it becomes actionable in the field.

Critical thinking provides the framework for this filtering. Three components guide the process (French, 2022). Empiricism demands that decisions rely on repeatable, observable evidence rather than tradition or authority. Rationalism requires logical reasoning rather than emotional conviction or wishful thinking. Scepticism insists on continuously questioning accepted beliefs and conclusions, examining both evidence and arguments before accepting them.

These components also serve as a boundary against pseudoscience — claims that present themselves as scientific but fail to provide supporting evidence. Over-reliance on tradition, “the coaching eye,” anecdotal success stories, and reluctance to question established conventions are patterns that exist even at the highest levels of sport. When unchallenged, these patterns become barriers to effective decision-making and can suppress the adoption of evidence-informed interventions (French, 2022).

The practical principle that emerges is to be data-informed, not data-driven (Walker et al., 2023). Research should inform decisions, not dictate them. The practitioner’s role is to critically evaluate published findings — examining the population, design, and ecological validity of each study — and then filter them through the specific context of their athletes and environment. A finding may be valid in its original context but inapplicable to an elite squad with different physical profiles, competition demands, and recovery constraints. The reductionist approach that isolates single variables in research should not be mirrored by a reductionist approach in application. Performance is multifactorial, and decision-making must reflect that complexity (Walker et al., 2023).

Experience: The Greatest Dataset and the Greatest Bias

The performance environment is populated by practitioners with unique and extensive experiential knowledge. Coaches have watched thousands of training sessions. Sport scientists have tracked load responses across multiple seasons. Medical staff have managed hundreds of injuries. This collective experience constitutes what has been described as the greatest dataset a team possesses (Brewer, 2022). Derived data — GPS metrics, force plate outputs, wellness questionnaires — exists not to replace this experiential knowledge but to remove guesswork and strengthen it with specificity.

The problem is that human observation is far less reliable than practitioners typically assume. When qualified football coaches watched a 45-minute match and were asked to recall critical events, their overall accuracy was 59.2% (Laird & Waters, 2008). Accuracy varied dramatically by category: shooting events were recalled at 95%, but possession — a continuous, less discrete event — was recalled at only 32.5%. Coaches were accurate about what drew their attention and inaccurate about what did not. Individual coach accuracy ranged from 36.7% to 76.7%, meaning some coaches remembered barely a third of what happened.

The limitations extend beyond memory to interpretation. When coaches were presented with identical video footage of a Bundesliga match and asked to identify tactical formations, build-up patterns, and key players, the level of agreement between them was extremely low (Furley et al., 2024). Fleiss-Kappa coefficients ranged from -.036 to .236 — essentially no agreement to weak agreement at best. Different coaches watching the same footage reached different conclusions about the same team’s formation, spatial occupation, and most influential players. Even higher-qualified coaches (UEFA A/B licence holders) achieved only marginally better agreement.

These findings reveal three cognitive biases that operate in experiential decision-making:

Bias	Mechanism	Example
Confirmation bias	Seeking or prioritising information that confirms pre-existing beliefs.	A coach who believes a player is unfit notices every error and overlooks every positive contribution.
Recall bias	Selectively remembering events that are dramatic, recent, or emotionally salient.	Shooting events are recalled at 95% while possession patterns are recalled at 32.5%.
Conformity bias	Adjusting observations to align with the dominant view within the staff group.	A sport scientist modifies their data interpretation to match what the head coach has already stated publicly.

The solution is not to dismiss experience. It is to complement experience with systematic data. When a coach observes that a player “looked tired” in training, GPS load data and wellness scores can confirm or challenge that observation with specificity. When video analysis staff identify a tactical pattern, event data can quantify its frequency and effectiveness across multiple matches. The integration of subjective and objective inputs produces a more complete and accurate picture than either can achieve alone.

The quality of communication between staff members is itself a contributing factor. Research has linked the quality of internal communication within the performance team to injury burden and player availability (Pillitteri et al., 2024). Data that is collected but not shared, shared but not understood, or understood but not acted upon fails to serve its purpose. The flow of information — from collection through interpretation to decision — requires deliberate organisational design, not just individual competence.

Placing the Athlete at the Centre of Decision-Making

The third pillar of EBP requires that athlete individual context informs every decision. The same training session produces different responses in different athletes. Genetic predisposition, training history, injury profile, psychological state, sleep quality, and positional demands all modulate the relationship between the stimulus imposed and the adaptation achieved. An intervention that benefits one player may be neutral or harmful for another.

Monitoring data serves this individualisation, but it must be understood as a decision-support tool that complements rather than replaces professional judgment (Rebelo et al., 2026). A monitoring system that tracks readiness through countermovement jump performance, subjective wellness, and heart rate variability provides daily signals about each athlete’s status. These signals do not prescribe the decision. They inform the conversation between the sport scientist, the coach, and the athlete that leads to the decision.

The balance between sensitivity and specificity matters. Setting a narrow threshold (e.g., ±1 standard deviation from the individual baseline) detects more signals but also produces more false positives. Setting a wider threshold (e.g., ±2 SD) reduces false alarms but risks missing early signs of maladaptation. In high-performance environments, missing an early warning sign of overreaching or injury (a Type II error) is typically more costly than investigating a false alarm (a Type I error), so practitioners tend to favour more sensitive thresholds (Rebelo et al., 2026). The decision about where to set this threshold is itself a judgment call that depends on context, season phase, and the individual athlete’s history.

The distinction between “knowing more” and “knowing better” captures a deeper challenge of athlete-centred practice (O’Sullivan et al., 2023). Traditional sport science has operated primarily through analysis: breaking complex phenomena into smaller parts, quantifying each in isolation, and accumulating more data points. This approach produces more knowledge but does not necessarily produce better understanding. Synthesis — understanding how parts relate within the whole — is required to convert data into insight. Analysis tells you that a player covered 300 metres less high-speed running than their season average. Synthesis tells you whether that decrement reflects tactical instruction, fatigue, opposition quality, or reduced motivation — and what, if anything, to do about it.

O’Sullivan et al. (2023) propose the concept of corresponsive sport science, in which the analyst is repositioned from an external observer — extracting data from the system — to a participant embedded within the system. Knowledge is not imposed on athletes and coaches from above. It grows through dialogue, shared inquiry, and mutual exploration. The athlete is not a passive subject of monitoring but an active contributor to the decision-making process. Their perceptions, preferences, and feedback shape both what is measured and how it is interpreted.

Leaders in elite sport environments navigate this integration through a combination of experience, intuition, and data analysis (King et al., 2026). Over-reliance on data introduces the risk of false certainty: a dashboard that shows green does not guarantee readiness, and a dashboard that shows red does not mandate rest. Effective leaders tend to satisfice rather than optimise — making a decision that is good enough given the available information and time constraints, rather than pursuing a theoretically optimal solution that may not exist in a complex, uncertain environment. The judgment of when to act on data, when to seek more context, and when to rely on professional instinct is what separates effective from ineffective decision-making.

Structures That Enable Integration

Integrating the three pillars of EBP cannot depend on individual effort alone. It requires organisational structures that facilitate cross-disciplinary collaboration, shared understanding, and continuous knowledge exchange.

The Department of Methodology (DoM) is a structural framework designed to coordinate transdisciplinary sport science support (Rothwell et al., 2020). Rather than each discipline — strength and conditioning, nutrition, sports medicine, performance analysis — operating in a silo with its own objectives and language, the DoM unites them under shared principles and a common methodology. The theoretical foundation draws from ecological dynamics, which views athlete preparation as a whole-system problem rather than a collection of independent components.

The deep integration of experiential knowledge and empirical knowledge is what enables innovative coaching and sport science models to emerge. When practitioners from different disciplines collaborate around shared principles, they can design training environments that simultaneously address physical, tactical, perceptual, and psychological demands — rather than treating each as a separate optimisation problem (Rothwell et al., 2020). The traditional silo model, where a strength and conditioning coach works independently from the technical coaching staff, risks duplicating effort, creating conflicting messages, and missing the interactions between different aspects of athlete development.

A Delphi study with 80 high-performance sport professionals identified five pillars of implementation for the DoM framework (Hydes et al., 2026):

Pillar	Core actions
Shared language	Co-create a working glossary, align terminology with the organisation’s vision and methodology.
Common principles	Define roles and responsibilities clearly, establish psychological safety, respect diverse perspectives.
Collaborative work	Understand coaching intent and performance objectives, communicate across disciplinary boundaries.
Continuous knowledge exchange	Hold regular meetings pre/post-training and pre/post-match, use training sessions as learning opportunities.
Collaborative practice design	Involve all DoM members in session design, include athletes in planning, use performance analysis to inform representative task design.

Consensus rates across these pillars ranged from 82% to 100%, indicating strong professional agreement on what integration should look like in practice. The barriers identified — power dynamics, interpersonal tensions, entrenched disciplinary cultures — are not technical but human. They require deliberate cultural work, not just structural reorganisation.

This structural integration also clarifies the cyclical relationship between EBP and practice-based evidence (PBE). Research guides practice, but the application of research in the field generates new data and new questions. These field-generated insights — practice-based evidence — can then be investigated through formal research, feeding back into the evidence base (French, 2022; Brewer, 2022). A training intervention adopted based on published research may produce unexpected outcomes in a specific squad. Documenting and analysing those outcomes creates new evidence that informs future decisions, completing the cycle.

This cycle carries a risk: practice-based evidence can lead to significant time investment in interventions that are later shown to be ineffective. The safeguard is to maintain scientific rigour in both directions — applying critical thinking when interpreting research and applying systematic documentation when evaluating field innovations. Innovation without evaluation is speculation. Evaluation without innovation is stagnation.

The organisational models that support this integration extend beyond the DoM. Holacracy, for example, distributes authority across the interdisciplinary team (IDT) rather than concentrating it in a top-down hierarchy (French, 2022). In a holacratic structure, situational leadership allows the relevant discipline to lead when its expertise is most needed — a medical professional during an injury crisis, a sport scientist during a load management decision — and then returns to a flat structure once the situation resolves. This prevents any single discipline from dominating the decision-making process and ensures that the most relevant expertise drives each specific decision.

Within this structure, the sport scientist functions as a decision facilitator — not the sole decision-maker, but the person who synthesises information from multiple sources, translates it into accessible formats for coaches and athletes, and facilitates the evidence-based discussions that lead to better decisions (Brewer, 2022). The value lies not in producing more data but in producing better understanding. The shift from data producer to decision facilitator redefines the sport scientist’s contribution: it is measured not by the volume of reports generated but by the quality of decisions supported.

Key Takeaways

• Evidence-based practice (EBP) is a decision-making process that integrates three pillars: scientific research, coaching expertise, and athlete values.
• Scientific literature cannot always be directly applied to elite settings due to participant (sub-elite) and design (reductionist) limitations, and must be evaluated through critical thinking — empiricism, rationalism, and scepticism.
• Experiential knowledge is the greatest dataset a team possesses, yet coaches recall only ~59% of match events accurately and show very low inter-rater agreement on video analysis, necessitating systematic data to complement observation.
• Monitoring data should serve as a decision-support tool that complements rather than replaces professional judgment — the principle of being data-informed, not data-driven.
• Athletes should be active participants in the decision-making process, and the goal should shift from “knowing more” (analysis) to “knowing better” (synthesis) through corresponsive practice.
• EBP and practice-based evidence (PBE) form a cyclical structure: research guides practice, and field application generates new evidence that feeds back into research.
• Organisational structures — shared language, common principles, collaborative practice design — enable the integration of all three pillars, and sport scientists function as decision facilitators within this framework.

References

1. Brewer, C. (2022). Performance interventions and operationalizing data. In D. N. French & L. Torres Ronda (Eds.), NSCA’s Essentials of Sport Science. Human Kinetics.
2. Dello Iacono, A., Datson, N., Clubb, J., Lacome, M., Sullivan, A., & Shushan, T. (2025). Data analytics practices and reporting strategies in senior football: insights into athlete health and performance from over 200 practitioners worldwide. Science and Medicine in Football, 10(1), 80-95. https://doi.org/10.1080/24733938.2025.2476478
3. French, D. N. (2022). Interdisciplinary support. In D. N. French & L. Torres Ronda (Eds.), NSCA’s Essentials of Sport Science. Human Kinetics.
4. Furley, P., Mehta, S., Raabe, D., & Memmert, D. (2024). Objectivity of match analysis in football: Testing the level of agreement between coaches’ interpretations of video data. International Journal of Sports Science & Coaching, 20(1), 45-55. https://doi.org/10.1177/17479541241278603
5. Hydes, S., Strafford, B. W., Rothwell, M., Stone, J., Davids, K., & Otte, F. (2026). A Department of Methodology: A feasible framework to integrate the applied practice of multidisciplinary support teams. International Journal of Sports Science & Coaching, 17479541251409729. https://doi.org/10.1177/17479541251409729
6. King, R., Yiannaki, C., Rhodes, D., & Alexander, J. (2026). From chaos to clarity how leaders leverage the value and impact of the multidisciplinary team in elite sport. Managing Sport and Leisure. https://doi.org/10.1080/23750472.2026.2632253
7. Laird, P. & Waters, L. (2008). Eyewitness Recollection of Sport Coaches. International Journal of Performance Analysis in Sport, 8(1), 76-84. https://doi.org/10.1080/24748668.2008.11868424
8. Pillitteri, G., Clemente, F. M., Sarmento, H., Figuereido, A., Rossi, A., Bongiovanni, T., Puleo, G., Petrucci, M., Foster, C., Battaglia, G., & Bianco, A. (2024). Translating player monitoring into training prescriptions: Real world soccer scenario and practical proposals. International Journal of Sports Science & Coaching, 20(1), 388-406. https://doi.org/10.1177/17479541241289080
9. Rebelo, A., Bishop, C., Thorpe, R. T., Turner, A. N., & Gabbett, T. J. (2026). Monitoring training effects in athletes: A multidimensional framework for decision-making. Sports Medicine. Advance online publication. https://doi.org/10.1007/s40279-026-02417-4
10. Rothwell, M., Davids, K., Stone, J. A., O’Sullivan, M., Vaughan, J., Newcombe, D. J., & Shuttleworth, R. (2020). A Department of Methodology Can Coordinate Transdisciplinary Sport Science Support. Journal of Expertise.
11. Sullivan, M. O., Vaughan, J., & Woods, C. T. (2023). Not just to know more, but to also know better: How data analysis-synthesis can be woven into sport science practiced as an art of inquiry. Sport, Education and Society, 29(9), 1114-1132. https://doi.org/10.1080/13573322.2023.2261970
12. Walker, G., Morgan, O., Matinlauri, A., Narcisi, A., Calder, A., & Davidson, C. (2023). Role of the practitioner. In A. Calder & A. Centofanti (Eds.), Peak performance for soccer: The elite coaching and training manual. Routledge.