Performance Analysis

Performance analysis is the systematic study of what athletes and teams actually do in competition and training.

7 frameworks, from 1950 to today · Sports Science · Sports & Human Performance

still a working view ended · each leads to its entry below

Overview

Performance analysis is the systematic study of what athletes and teams actually do in competition and training. Its practitioners record behaviour—passes, shots, sprints, positions, errors—turn the record into numbers or pictures, and ask what those patterns mean for performance and for the decisions coaches make. What is at stake is simple to state and hard to deliver: memory and impression are unreliable, so a coach who "remembers" that a team faded in the second half may be wrong, and a team that believes it presses high may not. Performance analysis exists to replace anecdote with evidence about the game itself, as distinct from the athlete's physiology or psychology.

Charting the Game by Hand: Notational Systems Before Computing

The oldest strand is notational analysis: the hand-recorded, systematic tally of events in a contest1. Its logic is as old as organised sport's desire to compare performances. In the mid-twentieth century, analysts worked with paper sheets, pencils, and stopwatches, coding each action against a predefined list and later counting the codes. A typical system divided a pitch or court into zones and recorded, for each entry into a zone, the action taken and its outcome—so that a coach could see, for example, how often attacks begun on one flank produced shots.

The approach's strength was its discipline: because the coding scheme was fixed in advance, two observers could be trained to record the same match and their sheets compared, making reliability a measurable property rather than a hope1. Its weakness was cost. A single match could take hours to code by hand, and the resulting counts were often crude—how many, but rarely how fast or where exactly. Still, the core idea that took hold in this era, and that has never left the field, is that a performance must be defined in operational terms before it can be counted.

Measuring Movement: Time–Motion Analysis

A parallel tradition asked not what players did with the ball but how they moved without it. Time–motion analysis measures the activity profile of an athlete or team: how much distance is covered, at what speeds, in what proportions of walking, jogging, running, and sprinting, and how those change across a match. Early work in the 1970s and 1980s relied on manual observation and film, with observers timing intervals and classifying movement by eye—laborious, and limited by what a human can track.

The method mattered because it exposed the mismatch between what a sport demands and what training supplied. If a midfielder spends a large share of a match at low intensity but repeatedly performs short maximal efforts, then training built only on steady running is misdirected. Time–motion analysis gave coaches a quantitative picture of the "work profile" of a position, and it became a foundation for conditioning and for substitution strategy. Its limitation was resolution: manual coding could not reliably capture very short high-speed efforts, and different observers classified movement differently.

The Video and Computer Revolution

Two technologies changed the field's economics. Video recording let analysts replay and re-examine events, and the personal computer let them store, sort, and query coded data1. From the 1980s onward, computerised video and database analysis systems combined a video timeline with a coded event log, so that a coach could call up every instance of a chosen event—every turnover in a defensive third, say—and watch them in sequence1. Commercial and academic systems of this kind spread through professional football, rugby, and other team sports, and the analyst became a recognised member of the coaching staff.

The gain was not merely speed. A database made comparison possible across matches and seasons, and it made the coding scheme itself an object of design: analysts debated what counted as a "successful" pass or a "tackle," because the answer determined everything downstream. This is where performance analysis began to overlap with match statistics and public data analysis—the use of published box scores, league records, and, later, freely available event data. Public statistics had long existed in sports like baseball and cricket, but their systematic analysis for insight rather than record-keeping grew into its own enterprise, and it fed back into professional practice as clubs hired analysts to mine data that anyone could obtain.

Tracking the Whole Field: Spatiotemporal Data

The next turning point removed the human coder from the loop of position recording. Spatiotemporal tracking systems—using multiple cameras, radio-frequency tags, or GPS units worn by players—record the position of every athlete many times per second, producing a continuous map of the game. Early commercial systems appeared in the 1990s and spread widely in elite football and other team sports through the 2000s.

This changed the questions that could be asked. Instead of counting events, analysts could measure the space between players, the compactness of a defensive block, the speed of a counterattack, or the synchrony of teammates' movements. A concrete consequence: a team's defensive shape could be quantified as an area or a distance between lines, and the effect of a tactical instruction could be tested against that measure. Tracking also sharpened time–motion analysis, since speed and distance could now be derived automatically rather than observed. The new problem was volume—enormous data streams requiring cleaning, modelling, and interpretation—and the risk of measuring what is easy to measure rather than what matters.

The Game as a Complex System: Ecological Dynamics

Running alongside these measurement traditions was a theoretical shift in how performance was understood. Ecological dynamics treats the athlete and the environment as a single system, and performance as the emergence of behaviour from the interaction between a player's capacities and the opportunities the game affords. Its roots lie in the perception–action psychology of James J. Gibson and in dynamical-systems thinking about coordination, and it entered sports science through work on interceptive tasks and team games from the 1990s onward2.

For performance analysis, the implication was that a match is not a list of independent events but a pattern of interacting constraints—pitch dimensions, opponent pressure, teammate positioning—and that analysis should describe those patterns rather than isolate actions. In practice this encouraged analysts to study coordination between players, the stability and breakdown of tactical formations, and the way small changes in constraints shift behaviour. It also connected analysis to coaching: if skill is emergent and situation-dependent, then practice design and match analysis are two sides of the same problem.

Machines That See: Automated Detection and Pose Estimation

The most recent turning point automates the coding itself. Automated event detection uses computer vision and machine learning to identify events—passes, shots, tackles—directly from video, while pose estimation infers the positions of an athlete's joints from images, making it possible to analyse technique without markers or laboratory equipment3. These methods matured through the 2010s as deep learning improved object and human detection, and they are now embedded in commercial tracking and analysis platforms.

The promise is scale: a whole league's matches can be processed without an army of coders, and movement can be analysed at the level of limbs rather than positions on a pitch. The open questions are the familiar ones in new form—accuracy under occlusion and crowd clutter, the gap between a detected event and a meaningful one, and whether automated metrics reproduce or correct the biases of the coding schemes they replace.

Where the Field Stands

Today performance analysis is a hybrid practice. Hand-built notational schemes still underpin much coaching work, because a well-designed code captures what a coach cares about and a generic algorithm may not1. Time–motion and tracking data are routine in elite team sports, feeding conditioning and tactical review. Ecological dynamics supplies a theoretical language for treating performance as emergent pattern, and computer vision is steadily absorbing the labour of event coding. The live questions are about validity and integration: which metrics actually relate to winning, how to combine event, position, and video data into a coherent picture, and how analysis can inform decisions in real time rather than after the fact.

References

  1. Notational analysis (introduction, History of data gathering in sports), Wikipedia
  2. Ecological psychology, Wikipedia
  3. Computer vision, Wikipedia

The frameworks of Performance Analysis

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1950-Present

    Notational analysis

    How can we reliably count and compare events in a game?

    Fly to its tower in Nousfarer

  2. 1970-Present

    Time–motion analysis

    How much and how fast do athletes move during play?

    Fly to its tower in Nousfarer

  3. 1980-Present

    Computerised video and database analysis

    How can video and databases make game analysis faster and more comparable?

    Fly to its tower in Nousfarer

  4. 1980-Present

    Match statistics and public data analysis

    What can public match statistics reveal about performance?

    Fly to its tower in Nousfarer

  5. 1990-Present

    Ecological dynamics

    How does behaviour emerge from athlete–environment interaction?

    Fly to its tower in Nousfarer

  6. 1990-Present

    Spatiotemporal tracking analysis

    How can we measure the positions and movements of all players continuously?

    Fly to its tower in Nousfarer

  7. 2010-Present

    Automated detection and pose estimation

    Can machines automatically detect events and body poses from video?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Notational analysisTime–motion analysisComputerised video and database analy…Match statistics and public data anal…Ecological dynamicsSpatiotemporal tracking analysisAutomated detection and pose estimati…

Beyond the field

Where Performance Analysis's ideas came from in other fields, and what they led to there: the strongest 12 of 28 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Performance Analysis is a planet in Nousfarer

Its city has a tower for each of these 7 frameworks. Land, walk in, and climb one concept at a time.

Land on Performance Analysis

Related fields

Fields its ideas are linked to

More in Sports Science

See where Performance Analysis sits on the Map of Human Knowledge