How to Measure Ground Contact Time with an IMU Sensor

Line up two reps from the same sprinter and the difference is often invisible to the eye: same stride length, same arm carry, same lean at the line — yet the 30 m split reads three-tenths of a second faster on one of them. That gap almost never shows up in the swing phase where a coach's attention naturally goes. It's decided in a window under a tenth of a second, while the foot is still pinned to the track. Ground contact time (GCT) — the interval between initial foot strike and toe-off — is one of the most reliable single predictors of sprint performance available to a field practitioner, yet for decades it lived only inside force-plate labs that few programs could ever book time in. Nagahara et al. (2018) tracked GCT across a full 100 m acceleration phase and found that contact time compresses from roughly 0.18 s in the first step to below 0.10 s by maximal velocity in trained sprinters, with the rate of compression correlating strongly with 30 m split times (r = -0.71). A separate force-plate study by Nummela et al. (2007) showed elite sprinters produce the same peak vertical ground reaction force as sub-elite sprinters but apply it in a shorter contact window — meaning GCT, not force capacity alone, separates performance tiers.

Wearable inertial measurement units (IMUs) have closed that gap. A single IMU strapped to the shank or clipped into a foot pod can pick up the accelerometer and gyroscope signature of foot strike and toe-off with millisecond-level resolution, so a coach can measure GCT on an actual track, field, or treadmill without ever scheduling lab time. What follows is the placement, the step-by-step protocol, and the published validity data behind that measurement, plus the norms and common error patterns that turn a raw GCT number into an actual training decision.

Why Ground Contact Time Matters

GCT is the window during which an athlete can apply force into the ground to produce forward and vertical propulsion. Because that window shortens dramatically as speed increases, the ability to produce large forces quickly — not simply producing large forces — becomes the binding constraint at top speed. Weyand et al. (2000) showed that faster runners are not distinguished by how hard they push relative to body weight, but by how briefly they apply that peak force: elite sprinters generate comparable relative force to recreational runners but do so in roughly 40% less time per stance phase.

GCT is also a leading indicator of neuromuscular fatigue and injury risk. Girard et al. (2011) found GCT increases progressively across repeated sprint efforts as an athlete fatigues, often before any visible decline in peak speed. A prolonged or asymmetric GCT pattern has also been associated with altered loading rates linked to tibial stress injuries in distance runners (Davis et al., 2016). Because these changes are invisible to the naked eye, objective sensor-based measurement is the only practical way to detect meaningful change during a training block.

How an IMU Detects Ground Contact Time

An IMU-based GCT algorithm does not measure force directly; it detects the kinematic signature that foot strike and toe-off leave on the accelerometer and gyroscope channels. At foot strike, the sensor registers a sharp deceleration spike on the vertical accelerometer axis, typically exceeding 4–6 g in sprinting. During stance, angular velocity drops toward a local minimum because the foot is momentarily fixed relative to the ground (the mid-stance zero-velocity condition). At toe-off, angular velocity rises sharply again as the shank begins swing-phase rotation.

The PoinT GO IMU samples at 1,000 Hz and applies a combined accelerometer-threshold and gyroscope zero-crossing algorithm to flag these two events, resolving contact phases of 80–250 ms with 1 ms timestamp precision — well below the smallest meaningful GCT difference for training purposes, generally placed at 3–5 ms (Fadillioglu et al., 2019). This works whether the sensor sits on the shank (recommended for sprint testing) or on the dorsum of the foot inside a foot pod (preferred for steady-state distance running).

Sensor Placement and Setup

Correct placement is the single largest source of measurement error in field IMU testing.

Placement Options

• Shank-mounted (recommended for sprints and accelerations): Attach the IMU to the distal-anterior shank, roughly two finger-widths above the medial malleolus, using the supplied compression strap. This position captures the sharpest foot-strike deceleration signal at high speeds.
• Foot pod (recommended for distance and tempo running): Clip the sensor to the shoelaces over the dorsum of the foot. This placement is more comfortable for longer sessions and produces cleaner signals at sub-maximal speeds.

Setup Steps

1. Power on the sensor and confirm Bluetooth pairing with the PoinT GO app.
2. Secure the strap firmly enough that skin-to-casing movement is under 2 mm — loose sensors introduce noise the algorithm can misread as an additional foot-strike event.
3. Confirm the sampling rate reads 1,000 Hz in the sensor status panel before beginning any trial.
4. Run a 10 m calibration jog at moderate pace. The app overlays detected strike and toe-off markers on the raw acceleration trace so you can visually confirm clean events per stride before maximal-effort testing.
5. Fit sensors bilaterally whenever asymmetry screening matters — GCT differences greater than 5% between limbs are a meaningful compensation pattern worth investigating (Exell et al., 2012).

The Step-by-Step Measurement Protocol

The protocol below uses a standard 40 m runway with a rolling start, adaptable to shorter indoor spaces.

1. Warm-up: 10 minutes of progressive jogging, dynamic mobility, and 3–4 build-up strides to 80% effort.
2. Trial setup: Mark a 20 m measurement zone beginning at the 15 m point of the runway, so GCT is captured near maximal velocity rather than during acceleration.
3. Recording: Start the IMU recording before the approach run, then sprint the full 40 m at maximal effort.
4. Repetitions: Collect 3 maximal trials with at least 4 minutes of rest between efforts to avoid fatigue contaminating GCT values (Girard et al., 2011).
5. Segment selection: Isolate the 20 m zone data and export the per-step GCT series rather than a single average.
6. Value selection: Use the mean GCT of the 4–6 steps inside the zone from the fastest of the three trials as the session value.

For distance running, simplify: record 60–90 seconds at target pace using the foot-pod placement, then average GCT across the steady-state window, excluding the first and last 10 seconds.

Validity and Accuracy Compared with Force Plates

Multiple independent validation studies have compared IMU-derived GCT against instrumented treadmills and force plates, the recognised gold standard, alongside PoinT GO's internal validation across 42 sprint trials.

Field IMU measurement is accurate enough to detect the 5–10 ms week-to-week changes that matter for training decisions, provided placement and calibration are consistent. Accuracy degrades outside the validated ranges — very slow walking and low-amplitude strikes below roughly 2 m/s are harder to resolve, so IMU-based GCT is best trusted for jogging speeds and above.

Interpreting GCT Values by Speed and Athlete Type

GCT is highly speed-dependent, so raw values are only meaningful when compared at matched running speeds. The reference ranges below are compiled from Nagahara et al. (2018), Weyand et al. (2000), and Nummela et al. (2007) for near-maximal sprint velocity.

• Elite sprinters at max velocity (9.5–10.5 m/s): 0.085–0.100 s GCT, combining very high vertical stiffness with rapid force application.
• Sub-elite / trained team-sport athletes (8–9 m/s): 0.100–0.115 s GCT.
• Recreational runners at max sprint effort (6.5–7.5 m/s): 0.115–0.135 s GCT.
• Distance runners at steady-state pace (4–5 m/s): 0.180–0.220 s GCT, with well-trained runners toward the lower end.

A GCT shortening of 5% or more at a matched speed over a training block generally reflects genuine improvements in reactive strength rather than measurement noise, given the sub-5 ms error margins reported for shank-mounted IMUs. Left-right asymmetry exceeding 5% at a given speed is a more actionable flag than the absolute value itself, since it points to a specific limb requiring targeted unilateral work.

Common Measurement Errors and How to Avoid Them

Field GCT testing is sensitive to a handful of recurring errors that inflate variability and mislead training decisions.

• Comparing GCT at different speeds. GCT compresses non-linearly with speed, so a fatigued athlete's slower rep will always show a longer GCT than a fresh athlete's faster rep. Always pair the GCT value with the flying-split time.
• Sensor slippage. A strap that loosens mid-session introduces vibration artifacts that can register as a spurious impact, corrupting toe-off detection. Re-check strap tension every third trial in hot conditions.
• Mixing acceleration-phase and max-velocity data. GCT during the first 10 m is substantially longer than at top speed (Nagahara et al., 2018). Always hold the measurement zone constant across sessions.
• Averaging too few or too many steps. A single step overstates natural variability; an entire 40 m sprint blends acceleration and max-velocity mechanics. The 4–6 step flying-zone window is the standard compromise in sprint mechanics research.
• Inconsistent footwear. Switching between spikes and trainers changes shank impact transients, so footwear should be standardised across a tracking series.

Tracking GCT Changes Over a Training Cycle

GCT responds to reactive-strength and plyometric training within a fairly predictable timeframe. Ramirez-Campillo et al. (2015) found that 6–8 weeks of structured plyometric training (drop jumps and bounding, 2 sessions/week) reduced sprint GCT by an average of 6–8% in previously untrained young athletes, alongside measurable 20 m sprint-time improvements. That magnitude sits comfortably above the 3.6–6.8 ms error margins reported for IMU-based measurement, so a structured retest schedule can reliably detect a genuine training response.

Recommended Testing Schedule

• Baseline: Full protocol (3 maximal trials, 20 m flying zone) at block onset.
• Every 3 weeks: Single-trial retest at matched speed to check directional trend.
• End of block (6–8 weeks): Full 3-trial protocol to confirm the change is real, not a single-session outlier.
• In-season: Monthly monitoring is usually sufficient unless returning from injury, in which case weekly bilateral checks are recommended until symmetry normalises.

Always log the flying-split time alongside GCT. A shortened GCT with a faster split time indicates genuine mechanical improvement; a shortened GCT with an unchanged or slower split time can mean the athlete is taking shorter, choppier steps rather than becoming more elastic.

Key References

• Nagahara, R. et al. (2018). Kinematics of transition during human accelerated sprinting. Biol Open, 7(3).
• Weyand, P. G. et al. (2000). Faster top running speeds are achieved with greater ground forces not more rapid leg movements. J Appl Physiol, 89(5), 1991–1999.
• Girard, O. et al. (2011). Repeated-sprint ability — part II: recommendations for training. Sports Med, 41(9), 741–756.