The Science Behind Movement Onset: How We Measure When Athletes Start Moving

Defining the Origin of Motion

The definition of "movement start" represents a fundamental challenge in sports science and biomechanics. The onset of movement is not a single, universally-agreed-upon instant, but rather a process that can be defined and detected across three distinct, sequential domains: the neurological, the kinetic, and the kinematic.

The Three Domains of Movement Onset

1. Neurological Onset: The earliest detectable precursor originates in the central nervous system—the brain's preparatory phase and volitional command to move, detectable via EEG/MEG signals that precede observable movement by hundreds of milliseconds.

2. Kinetic Onset: The first instant an athlete applies measurable force against their environment (e.g., ground contact on a force plate). This is considered the "gold standard" in laboratory settings but requires specialized equipment.

3. Kinematic Onset: The first observable change in a body segment's or center of mass's position, velocity, or acceleration in space. This is the consequence of kinetic force application and is measured by field-based technologies including radar, laser, IMU, GPS/LPS, video systems, and timing gates.

Ledsreact's Approach: Kinematic Detection

Like other field-based sports technologies (IMU/GPS systems, video analysis, and timing gates), the Ledsreact Pro focuses on kinematic onset detection. We measure the first observable change in the athlete's position, velocity and acceleration using our radar-based tracking system. This approach balances practical field applicability with scientific rigor, recognizing that kinematic onset occurs milliseconds after kinetic onset but provides the most practical and reliable measurement for athletic performance assessment.

Comparison with Timing Gates: Addressing Systematic Bias

Traditional timing gates use infrared or laser beams that trigger when broken by the athlete's body. However, this technology introduces significant systematic errors that affect measurement accuracy.

False Triggers and Measurement Errors

Timing gates detect when any part of the athlete's body breaks the beam—arms, legs, or forward lean—not necessarily when the center of mass crosses the gate line. Research demonstrates that:

• False signals occur in 32% of trials in 10-meter sprinting with single-beam systems (Earp & Newton, 2012)
• Single-beam timing errors can reach ±0.06 seconds, three times the smallest worthwhile change in sprinting (Haugen et al., 2014)
• For short sprints (5-10 m), false signals can represent over half of expected performance changes, masking actual training adaptations
• Beam height significantly affects measurements: dual-beam systems at 0.60 m produce faster times than those at 0.80 m, making results measured at different heights not comparable (Altmann et al., 2017)

Bridging Historical Data

Many teams have extensive timing gate datasets but face operational limits with legacy systems. Ledsreact offers two robust solutions to bridge this gap:

• Standard bias correction: Validated, protocol-specific adjustments ensure continuity with established timing gate methodologies.
• Custom computer vision analysis: Quantifies your laboratory or field timing gate setup’s systematic bias relative to Ledsreact radar using AI-based video evaluation—ideal for non-standard or variable setups.

Both methods allow direct, statistically valid comparison between new Ledsreact results and existing historical timing gate data.

Ledsreact's Advantage

The Ledsreact Pro radar system tracks the athlete's entire body movement continuously, measuring movement onset based on center of mass velocity rather than discrete beam interruptions. This eliminates false triggers and provides more accurate, reliable start detection. Our validation studies show high correlation (r > 0.95) and low mean differences (< 30 ms) with timing gates after bias correction (see our case study).

Countdown vs. Player-Initiated Modes

The Ledsreact Pro supports two distinct start modes, each serving different training and assessment purposes.

Countdown Mode

In countdown mode, the system provides an audible countdown (typically three beeps) followed by a green LED signal (the 4th beep). This mode enables reaction time measurement.

• Reaction Time Calculation: The time between the start of the green signal (4th beep) and the first detected movement is measured as reaction time

• Use Cases: Standardized testing protocols, comparing athletes under identical conditions, assessing reaction time as a performance component

• Protocol Consistency: Ensures all athletes respond to the same external stimulus, reducing variability from self-initiated starts

Player-Initiated Mode (No Countdown)

In player-initiated mode, the athlete starts when ready, without an external countdown signal.

• No Reaction Time: Since there's no external signal, reaction time is not measured

• Use Cases: Focus on pure movement performance without reaction time component, training scenarios where athletes control their own start timing

• Natural Movement: Allows athletes to start when they feel optimally prepared

Understanding Reaction Time

Reaction time measurement is fundamental to assessing athletic performance, particularly in sprint starts and agility tasks. Understanding how reaction time relates to movement onset detection helps coaches interpret performance data accurately.

Definition of Reaction Time

In sports science, reaction time is defined as the time interval between the presentation of a stimulus (signal) and the initiation of a motor response. For agility exercises, this stimulus-response time is critical as it reflects an athlete's ability to quickly respond to a changing environment.

Types of Reaction Time:

• Simple Reaction Time (SRT): Involves responding to a single, known stimulus, such as a sprint start after hearing a buzzer or seeing a light signal

• Choice Reaction Time (CRT): Requires deciding between multiple stimuli and responses, as in complex agility tasks like T-tests, X-drills (agility box), or 5-10-5 tests with a random signal

The Ledsreact Pro offers various standardized and customizable agility tests for measuring both reaction time types.

To measure simple reaction time, put the device in countdown mode, where athletes respond to the green LED signal (4th beep). Enable "random start" mode to minimize anticipatory movements, as explained in the video below.

To measure choice reaction time (CRT) with the Ledsreact Pro, choose a multi-directional agility test—for example:

• T-test
• Agility box (X-drill)
• 5-10-5 shuttle
• Y-drill
• Front-and-back
• Or any other customized protocol that fits your training needs

These exercises can also be performed with a ball or relevant equipment to better replicate real game or training scenarios.

Below is a video playlist demonstrating how Thierry Barnerat uses the Ledsreact Pro for high-performance goalkeeper agility drills:

Reaction Time and the Three Domains of Movement Onset

Reaction time measurement bridges the gap between the neurological and kinematic domains of movement onset:

1. Neurological Domain: The brain processes the stimulus (visual/auditory signal) and generates the motor command—this occurs in the first 50-100 milliseconds after stimulus presentation

2. Kinetic Domain: The athlete applies force against the ground, initiating movement—this typically occurs 100-200 milliseconds after the stimulus

3. Kinematic Domain: Observable movement begins—this is what the Ledsreact Pro detects as movement start, typically occurring 100-250 milliseconds after the stimulus depending on athlete level

What We Measure: The Ledsreact Pro measures reaction time from the stimulus presentation (green LED signal) to kinematic onset detection (first observable movement). This captures the complete stimulus-response cycle, including neural processing, force application, and observable movement initiation.

Expected Reaction Times by Athlete Level

Reaction times vary significantly by athlete profile, training level, and type of activity:

Elite Athletes (SRT):

• Sprint Start: 100-120 milliseconds (ms)

• Agility Task: 150-200 ms

Sub-Elite Athletes:

• Sprint Start: 120-180 ms

• Agility Task: 200-250 ms

Recreational Athletes:

• Sprint Start: 180-250 ms

• Agility Task: 250-350 ms

Factors such as fatigue, training load, specific sport discipline, and environmental conditions can slightly alter these benchmarks. Consistent measurement allows coaches to track changes in reaction time as an indicator of performance readiness and training adaptations.

False Start Detection: The 100ms Threshold

Definition: A false start is defined as any movement initiated within 100 milliseconds after the signal.

Scientific Rationale: The 100ms threshold is based on human neural processing speed. Research demonstrates that the minimum time required for sensory information to travel from the stimulus (visual or auditory) through the nervous system to generate a motor response is approximately 100ms. This includes:

• Sensory processing time (visual/auditory cortex): ~20-40ms

• Neural transmission and decision-making: ~30-50ms

• Motor command generation: ~20-30ms

• Muscle activation and force production: ~30-50ms

Any reaction faster than 100ms is considered a preemptive movement, indicating that the athlete anticipated the signal rather than genuinely reacting to it. This threshold is recognized as the gold standard across multiple sports.

Sport-Specific Examples:

• Track and Field (Sprint Events): The International Association of Athletics Federations (IAAF) defines a false start as any reaction time below 100ms after the starting gun

• Swimming: The threshold is also generally set at 100ms, with starts under this time considered false

• Agility Testing: The same 100ms threshold applies to ensure fair and valid reaction time measurements

Ledsreact Pro False Start Detection: Our system automatically flags any detected movement occurring within 100ms of the stimulus signal as a false start, ensuring measurement validity and preventing anticipatory movements from contaminating performance data. This automatic detection helps coaches identify athletes who may be anticipating the signal rather than reacting to it, which is important for both training and competition preparation.

If the system detects a false start, you have two options: quickly retry the current attempt, or skip ahead to the next player in your test sequence. This is illustrated in the following video:

Ledsreact's Detection Approach

Accurately detecting movement start is fundamental to measuring reaction time, acceleration, and all subsequent performance metrics. Yet most field-based sports technologies rely on simple threshold-based methods—defining movement start as the moment velocity or acceleration crosses a fixed value. Research reveals fundamental limitations with these approaches that compromise measurement accuracy.

Why Simple Thresholds Fall Short

Simple threshold-based systems face three critical problems:

The Validity vs. Reliability Trade-off: Low velocity thresholds (e.g., 0.2 m/s) capture early motion but are extremely sensitive to pre-movement noise—swinging arms, weight shifts, postural adjustments—resulting in poor inter-system reliability (Stattin, 2019). Higher thresholds (e.g., 1.0 m/s) improve reliability but intentionally miss the critical first phase of acceleration (Haugen et al., 2014).

Fixed Acceleration Thresholds Fail: Systems using fixed acceleration thresholds (e.g., 2.78 m/s²) face a fundamental flaw: maximal acceleration decreases linearly with initial speed. From a standing start, athletes achieve ~5.7 m/s², but from a fast jog (15 km/h), maximal acceleration drops to ~2.4 m/s²—below the threshold. The system misses maximal-effort sprints entirely (World Conference on Science and Soccer, 2019).

Systematic Biases: Simple threshold crossing introduces biases that vary with movement amplitude, step cycle timing, and signal noise, making consistent measurement across conditions impossible (PMC, 2011).

Our Multi-Criteria Solution

The Ledsreact Pro solves these problems with a sophisticated multi-criteria detection system that goes beyond simple threshold crossing. Rather than relying on a single threshold, our system employs multiple independent detection criteria working in parallel:

• Sustained acceleration analysis: Identifies when forward acceleration exceeds exercise-specific thresholds and is maintained over time

• Velocity threshold confirmation: Detects when speed crosses a minimum threshold with confirmation from subsequent data points

• Position change detection: Recognizes meaningful displacement patterns indicating intentional movement initiation

• Advanced signal processing: Kalman filtering and noise reduction ensure clean data before detection analysis

The system evaluates all criteria independently and uses the most conservative (latest) detection among those that trigger—minimizing false positives while ensuring we capture true movement onset. Detection parameters are tailored to each exercise type (sprints, change-of-direction tests, agility drills), ensuring optimal accuracy for each specific use case.

This approach balances capturing early movement (validity) with maintaining consistent measurements (reliability), providing robust detection across different movement patterns, athlete sizes, and exercise types—without the fundamental limitations of simple threshold-based systems.

Starting Position Requirements

Critical Requirement: The athlete must start in the correct position, or movement start will not be detected and the exercise will not complete.

Starting Position by Exercise Type

Sprint and COD Tests, T-Tests:

• Athletes must start in front of the starting line (between the Ledsreact Pro device and the starting line)

• The system detects movement when the athlete crosses forward from this position

• Starting behind the line or too far forward will prevent proper detection

Example of setup:

5-10-5, Agility Box, and Custom Tests:

• Athletes must start inside the designated start checkpoint

• The system uses checkpoint-based detection for these multi-directional tests

• Proper positioning ensures accurate movement onset detection

Why This Matters

The radar system requires the athlete to be within its detection range and moving in the expected direction. Starting outside the designated area means:

• The system cannot establish a baseline position

• Movement onset cannot be reliably detected

• The exercise will not complete, and no data will be recorded

Always ensure athletes understand and follow the starting position requirements for their specific exercise type.

Data Visualization and Chart Interpretation

Understanding how movement start detection affects data visualization is crucial for accurate performance analysis.

Smoothened Speed and Acceleration Charts

The Ledsreact Pro displays smoothened speed and acceleration charts that provide a clear visualization of the athlete's performance throughout the exercise. These charts:

• Use advanced signal processing to create smooth, interpretable curves

• Remove high-frequency noise while preserving the essential movement characteristics

• Enable coaches to identify acceleration patterns, peak velocities, and deceleration phases

Time Zero (t = 0s) Definition

In our charts, t = 0s represents the time of first detected movement—the moment our system identifies that the athlete has begun the exercise.

Important Understanding:

• Theoretically, for a standing start, velocity should be 0 m/s at t = 0s

• In practice, you may observe a small initial velocity (typically around 0 up to 2 m/s) and acceleration at t = 0s

• This may occur when bias correction is applied to align measurements with historical timing gate data

Why Initial Velocity May Not Be Zero

The initial velocity you observe at t = 0s may reflect the impact of bias correction when comparing historical timing gate data:

Bias Correction Impact: When bias correction is applied to align Ledsreact Pro measurements with historical timing gate data, the system filters out data points from the early acceleration phase and shifts timestamps accordingly. This means that t = 0s in the visualization represents the point in time after the bias period has been removed, at which point the athlete may already be moving at a measurable velocity. The velocity shown at t = 0s reflects the athlete's actual speed at that corrected time point, not an artifact of the measurement system.

Visualization showing how bias correction affects initial velocity. The red dashed line shows the original data starting near 0 m/s, while the blue solid line shows the kept data after bias correction, which begins at approximately 1.7 m/s. The filtered region (yellow shaded area) between the reaction time (0.377s) and cutoff time (0.627s) represents the 0.250s bias period that has been removed. After bias correction, t = 0s corresponds to the cutoff time, where the athlete is already moving at a measurable speed.

Additional Factors: In cases where bias correction is not applied, the initial velocity may also reflect:

• Multi-Criteria Detection: Our detection system identifies movement onset using multiple criteria, which may occur when the athlete has already begun accelerating

• Mathematical Reconstruction: Advanced motion analysis including backward interpolation and spline fitting may show an initial velocity due to boundary conditions in mathematical models

• Signal Processing Pipeline: Kalman filtering and coordinate transformations applied during processing can affect how initial velocity appears in the final visualization

• Flying (or False) Starts: For protocols where the athlete is already moving at the start (e.g., flying sprints), the initial velocity at t = 0s reflects their actual running speed as they cross the start line.

Reaction Time Not Shown on Charts

For clarity and focus on movement performance, initial reaction time is not displayed on the speed and acceleration charts. Reaction time is measured and reported separately as a distinct metric, allowing coaches to:

• Analyze movement performance independently from reaction time

• Compare movement metrics across different start conditions

• Focus on acceleration and velocity profiles without reaction time confounding the visualization

Conclusion

Movement start detection is a complex challenge that requires balancing multiple competing priorities: capturing reaction time and the earliest meaningful movement, filtering out noise, detecting false starts, maintaining reliability across different conditions, and providing accurate performance metrics.

The Ledsreact Pro's radar-based approach, combined with sophisticated signal processing and multi-factor detection algorithms, provides a superior solution compared to traditional timing gates. By tracking the athlete's entire body movement continuously and using advanced filtering techniques, we eliminate the systematic biases inherent in beam-based systems while capturing the critical early acceleration phase that coaches need for performance analysis.

Our system also provides accurate reaction time measurement, automatically detecting false starts (<100ms) to ensure measurement validity. Understanding how movement start is detected—including the relationship between neurological, kinetic, and kinematic domains—and why initial velocity may not be exactly zero in visualizations helps coaches and athletes interpret performance data accurately and make informed training decisions based on reliable, scientifically-grounded measurements.

References

Altmann, S., Neumann, R., & Woll, A. (2018). Influence of alternative starting procedures on sprint performance in timing gate systems. Journal of Strength and Conditioning Research, 32(10), 2935–2941.

Altmann, S., Spielmann, M., Engel, F. A., Neumann, R., Ringhof, S., Oriwol, D., & Haertel, S. (2017). Validity of Single-Beam Timing Lights at Different Heights. Journal of Strength and Conditioning Research, 31(7), 1994–1999.

Bond, C. W., Willaert, E. M., & Noonan, B. C. (2017). Comparison of Three Timing Systems: Reliability and Best Practice Recommendations in Timing Short-Duration Sprints. Journal of Strength and Conditioning Research, 31(4), 1062–1071.

Earp, J. E., & Newton, R. U. (2012). Advances in electronic timing systems: considerations for selecting an appropriate timing system. Journal of Strength and Conditioning Research, 26(5), 1245–1248.

Haugen, T., Buchheit, M., & Sprint Research Group (2014). Sprint measurement methods: reliability and validity considerations. Sports Medicine, 44(5), 619–638.

Stattin, S. (2019). Concurrent validity and reliability of a time-of-flight camera on measuring muscle's mechanical properties during sprint running. Examensarbete, Umeå University.