Fundamentals of Wearable Technology

Wearable monitoring has roots in clinical instrumentation rather than sport. The Holter monitor, a portable ECG recorder developed in the 1950s, established the principle that physiological signals could be captured during free-living activity. Mechanical pedometers, popular by the 1960s, did the same for ambulatory activity counts. Heart-rate chest straps reached the consumer sport market with Polar’s launch of the Sport Tester in the early 1980s, opening structured heart-rate-based endurance training to recreational and competitive athletes. GPS-based athlete tracking emerged in elite team sport in the early 2000s, initially as backpack-worn units in rugby and Australian football. Inertial measurement units (small packages combining accelerometer, gyroscope, and magnetometer) have followed a similar trajectory from research instrumentation to mainstream coaching tools over the past decade. The next wave includes smart textiles (sensors woven into garments), continuous biochemical monitoring (notably continuous glucose monitors transitioning from clinical to performance contexts), and increasingly sophisticated optical sensing in consumer-grade watches and rings.

The pattern across this 70-year arc is not “technology pushed to coaches”; it is “coaches and researchers asked questions that needed instrumentation to answer”. The implication for the practising coach is simple: when you encounter a new wearable, ask first what coaching question it is trying to answer. If you cannot articulate that question for your own sport and athletes, the device is unlikely to change your practice.

Coaches often hear sensors named and conclude they understand them. The deeper competence is knowing what each sensor physically measures, and what the device then derives from that measurement. The two are often confused, leading to overconfidence in derived metrics whose underlying signal is noisy.

Accelerometers measure linear acceleration along one, two, or three axes. From accelerometer data alone, devices estimate step counts, impacts, and certain composite “load” metrics. The signal is robust but ambiguous: vibration from a vehicle and a footstrike can look similar at low sampling rates.

Gyroscopes measure angular velocity: the rate of rotation about an axis. Combined with accelerometers in a tri-axial IMU, they enable orientation estimation and motion-pattern recognition (jumping, change of direction, gait phases).

Magnetometers measure the local magnetic field, used as a compass reference and to correct drift in orientation estimates. They are sensitive to nearby ferrous metal and electronic noise, which limits their reliability in indoor sport environments.

Optical photoplethysmography (PPG) sensors shine green or infrared light into tissue and measure the variation in reflected light caused by pulsatile blood flow. Wrist-based watches and arm-band heart-rate monitors use PPG. The signal is sensitive to motion artefacts, skin tone, peripheral vasoconstriction, and the tightness of the device against the skin.

Electrocardiogram (ECG) electrodes measure the electrical activity of the heart directly. Chest-strap heart-rate monitors typically use a one-lead ECG approach. The signal is generally more accurate than PPG, especially during dynamic exercise, but requires direct skin contact and tolerance of the strap.

GPS/GNSS receivers triangulate position from satellites. Accuracy depends on the number and geometry of visible satellites, the receiver’s chipset quality, atmospheric conditions, and the local environment (open field versus urban canyon versus indoors). GPS-derived metrics such as total distance and high-speed running distance are robust in open outdoor sport; they degrade or fail indoors.

Surface electromyography (sEMG) sensors measure the electrical activity of muscle through the skin. Used to estimate muscle activation and fatigue, though signal quality is highly sensitive to electrode placement, skin preparation, and motion.

Temperature sensors and galvanic skin response sensors capture skin temperature and electrodermal activity, used in some recovery- and stress-monitoring devices.

A practical clarifying distinction:

The further you travel from measured quantity toward inferred construct, the more layers of assumption you accumulate. When two devices report “different” PlayerLoad values for the same drill, the answer is usually that they are using different proprietary derivation formulas, not that one is wrong and one is right.

Rather than ranking brands, the most durable way to teach the device market is by category. The table below is indicative; specific models change quickly, but the categories and the questions they invite are stable.

Three plain-language definitions every coach needs:

Validity asks whether a device measures what it claims to measure. A GPS unit that reports total distance is valid for that purpose if it agrees with a gold-standard reference (a measured track, for example) within an acceptable margin.

Reliability asks whether a device gives consistent results when conditions are stable. A device that gives different total-distance readings for the same athlete running the same route on two consecutive days is unreliable.

Ecological validity asks whether validity and reliability established in a laboratory hold up in real coaching conditions: wind, sweat, contact, indoor versus outdoor, gripped objects, weather.

Research evidence. Sperlich and Holmberg (2017) wrote a brief but influential editorial in the British Journal of Sports Medicine warning that “wearable, yes, but able…?” The gap between marketed performance and validated performance is often large. Düking, Hotho, Holmberg, Fuss, and Sperlich (2016) reviewed commercially available wearables and found that the accuracy, reliability, or validity of only nine devices had been evaluated scientifically. Akenhead and Nassis (2016) documented that coaches’ decisions about monitoring tools are made under significant uncertainty about validation status.

How to read a validation study abstract in five minutes. Look for: (a) sample: who was tested, in what sport, at what intensity; (b) gold standard: what the device was compared against; (c) statistic: mean bias and limits of agreement, not just correlation; (d) ecological match: were conditions like yours? If a device was validated only at slow treadmill speeds, do not assume it generalises to sprint training.

What every coach must understand before turning a device on.

Under the European Union’s General Data Protection Regulation (Regulation 2016/679), personal data is any information relating to an identifiable natural person. Health data (heart rate, biomechanical data, sleep, training load) is treated under Article 9 as a special category of personal data, requiring stricter handling. The minimum operational implications:

• You must have a clear lawful basis for processing the data (Article 6) and an explicit basis for processing the health data (Article 9), most commonly explicit consent in a sport-coaching setting that is not a medical context.
• Consent must be informed (the athlete understands what is collected, why, with whom it is shared, for how long) and freely given (no coercion from selection pressure).
• Minors require additional protections: parental or guardian consent, simpler explanations adapted to the athlete’s age.
• You must apply data minimisation: collect only what you need.
• You must define and communicate retention periods (how long the data is kept).
• You must have a plan for what happens if there is a breach: under GDPR, certain breaches must be reported within 72 hours.

Module 5 deepens the operational treatment of governance. For Module 1, the operational rule is simple: do not deploy a device without an informed consent process in place.