Loading field map
Performance analysis is the systematic observation, measurement, and interpretation of human movement and behavior in sport to inform coaching decisions and improve athletic performance. It sits at the intersection of sports science and coaching practice, translating raw data about what athletes actually do into actionable insights. The field is defined less by a single method than by a shared commitment to evidence-based feedback: rather than relying on memory, intuition, or anecdote, performance analysts document performance in ways that can be reviewed, quantified, and compared.
The central questions of the field are deceptively simple: What happened? Why did it happen? What should happen next? Answering these requires deciding what to measure, how to measure it, and how to present the results so that coaches and athletes can use them. The stakes are practical and immediate—performance analysis does not aim to discover universal laws of human movement but to solve specific problems for specific teams and individuals, often under severe time constraints.
Performance analysis emerged from two distinct practices that developed largely independently before converging in the late twentieth century. The first is notational analysis, which has roots in early attempts to quantify game events. As early as the mid-twentieth century, researchers and coaches began recording discrete actions in match play—passes, shots, tackles, errors—using paper-and-pencil systems. These early systems were labor-intensive but revealed patterns that casual observation missed, such as the relationship between scoring opportunities and specific passages of play. Notational analysis matured into a recognized research area through the 1970s and 1980s, when researchers developed structured coding schemes for invasion sports like soccer, hockey, and basketball. Its core assumption is that a match can be broken down into a finite set of countable events, and that the frequencies, sequences, and contexts of these events carry meaningful information about performance.
The second tradition is biomechanical analysis, which applies the principles of mechanics to human movement. Where notational analysis operates at the level of game events, biomechanical analysis operates at the level of the body itself—joint angles, forces, velocities, and timing. Its methods include high-speed video capture, force plates, and motion capture systems. The goal is typically to identify technically efficient movement patterns, correct faults, or understand the mechanical demands of a skill. This tradition has deep roots in kinesiology and sports medicine, and it developed alongside advances in film and video technology that made detailed frame-by-frame examination possible.
These two traditions differ not only in their unit of analysis but in their fundamental orientation. Notational analysis is primarily concerned with outcomes and decisions—what players chose to do and whether it worked. Biomechanical analysis is concerned with execution—how the body produced the movement. A soccer penalty kick, for example, can be analyzed notationally (where the ball was placed, which corner the goalkeeper dived to, whether the shot scored) or biomechanically (the angle of the run-up, the position of the supporting foot, the velocity of the kicking leg at impact). Both are forms of performance analysis, but they answer different questions and require different expertise.
The widespread adoption of video recording in the 1980s and 1990s transformed both traditions and ultimately brought them together. Video provided a permanent, reviewable record of performance that could be annotated, replayed, and shared. For notational analysts, video made coding more accurate and allowed events to be verified. For biomechanists, video offered a non-invasive way to capture movement in competition settings, not just in laboratory conditions. The result was a gradual convergence around a common workflow: capture performance on video, code or measure it systematically, analyze the resulting data, and present findings to coaches and athletes.
This convergence was also driven by the practical needs of high-performance sport. Professional teams began employing dedicated performance analysts whose job was not to conduct research but to support coaching staff on a daily basis. These practitioners needed to be fluent in both traditions—able to produce a shot map for a match analysis meeting and also to break down a technical fault in an athlete's serve. The field's identity crystallized around this dual competence, and by the early 2000s, performance analysis was recognized as a distinct subdiscipline within sports science, with its own journals, conferences, and academic programs.
Contemporary performance analysis is organized around several distinct approaches that coexist and often overlap. The most established is event-based analysis, the direct descendant of notational analysis. In this approach, an analyst defines a coding scheme—a set of event categories relevant to the sport—and then watches video, logging each occurrence with time stamps and contextual variables. Modern software allows analysts to code in real time or from recorded footage, and to generate reports automatically. The power of event-based analysis lies in its flexibility: coding schemes can be tailored to any sport and any tactical question. Its limitation is that it depends entirely on the quality of the coding scheme. If the categories are poorly defined, or if the analyst misses events, the resulting data will be misleading regardless of how sophisticated the subsequent statistics are.
A second approach is tracking-based analysis, which became possible with the introduction of computer vision systems that automatically track players and the ball throughout a match. Instead of discrete events, tracking data provides continuous positional information for every player at every moment. This allows analysts to measure things that were previously invisible: the distance covered by each player, their speed and acceleration profiles, the distances between players, the size of spaces on the pitch, and the speed of ball movement. Tracking data has enabled a new generation of performance metrics, such as expected goals (a statistical estimate of the probability that a shot will score based on its location and circumstances) and various measures of player load and physical output. The approach is particularly influential in soccer and basketball, where optical tracking systems are installed in major stadiums. Its limitation is that it captures position but not intention or context—knowing where a player ran does not explain why they ran there.
A third approach is qualitative analysis, which resists the quantification impulse of the other two. Qualitative analysts argue that some of the most important aspects of performance—decision-making under pressure, tactical awareness, team cohesion—cannot be reduced to counts or coordinates. Instead, they use structured observation, video review, and interviews to develop rich descriptions of performance. This approach is less common in elite sport, where coaches demand numbers, but it remains influential in coaching education and in research on expertise. Its contribution is to remind the field that performance is not just what can be measured, and that the interpretation of data always requires contextual understanding.
A fourth approach, sometimes called performance analytics, applies advanced statistical and machine-learning methods to performance data. This is a relatively recent development, enabled by the accumulation of large datasets from tracking systems and event coding. Analysts in this tradition build predictive models, identify patterns in large datasets, and develop new metrics that combine multiple variables. Expected goals models are the most famous example, but the approach extends to player recruitment, tactical pattern recognition, and injury risk prediction. Performance analytics differs from earlier approaches in its emphasis on prediction and optimization rather than description and feedback. It is also more likely to be conducted by data scientists than by traditional sports scientists, which has created some tension within the field about what counts as performance analysis.
These approaches are not rival paradigms in the sense of mutually exclusive worldviews. In practice, they form a complementary toolkit, and most professional performance analysis departments use several simultaneously. Event-based analysis provides the tactical narrative of a match—what happened, in what order, with what result. Tracking data provides the physical and spatial context—how much ground was covered, where the spaces were, how the team shape changed over time. Qualitative analysis provides the interpretive layer—what the coach and players were trying to do, what they perceived, how they experienced the match. Performance analytics provides the predictive layer—what the data suggests about likely future outcomes and which patterns are statistically meaningful.
The relationships among these approaches are sometimes complementary and sometimes competitive. Event-based and tracking-based analyses often overlap, since many events can be derived from tracking data, and tracking data can enrich event data with spatial context. Qualitative analysis is sometimes seen as unscientific by quantitatively oriented practitioners, while qualitative researchers criticize the reductionism of purely numerical approaches. Performance analytics has been accused of ignoring the human and contextual factors that qualitative analysts emphasize. These tensions are productive: they force the field to continually question what counts as evidence and what kinds of knowledge are most useful for improving performance.
The contemporary field is characterized by several durable features. First, technology has become the primary driver of methodological change. The shift from manual coding to semi-automated systems, and from video to positional data, has expanded what can be analyzed and who can do it. Wearable sensors, including GPS units and inertial measurement units, have extended analysis beyond competition into training, allowing coaches to monitor workload and fatigue on a daily basis. Artificial intelligence is beginning to automate aspects of both event detection and tactical analysis, though human oversight remains essential for interpretation.
Second, the field has become professionalized. Performance analysis is now a recognized career path, with dedicated roles in professional teams, national governing bodies, and academic institutions. This professionalization has brought standardization—common software platforms, shared terminology, and certification programs—but also a degree of homogenization, as teams increasingly adopt similar methods and metrics.
Third, there is a persistent gap between research and practice. Academic performance analysis tends to be methodologically rigorous but often lags behind the needs of practitioners, who require timely, accessible, and actionable information. Coaches frequently rely on simple metrics and video clips rather than sophisticated statistical models, and many published studies have limited direct application to real coaching contexts. Bridging this gap remains one of the field's central challenges.
Fourth, the field is increasingly data-rich but interpretation-poor. The volume of information generated by modern tracking systems far exceeds what any coaching staff can absorb, and the challenge has shifted from collecting data to deciding what matters. This has elevated the importance of the analyst's interpretive role: the most valuable analysts are not those who can produce the most data but those who can identify the few insights that will change what a team does.
Performance analysis remains a fundamentally applied field, accountable to the practical question of whether it helps athletes and teams perform better. Its methods have become more sophisticated, and its tools more powerful, but its core purpose is unchanged: to replace guesswork with evidence, and to give coaches a clearer view of the game they are trying to win.