Velocity Prescription for Strength: An 800Hz IMU Decision Framework

Velocity prescription anchors training intensity to concentric velocity (m/s) instead of absolute load (kg). The premise is that 1RM fluctuates day-to-day, but the load-velocity relationship within an individual is stable. González-Badillo & Sánchez-Medina (2010) showed that velocity at a given load reflects daily neuromuscular readiness with ±2.1 kg accuracy. An 800Hz IMU resolves that relationship to 0.01 m/s, enabling dynamic load adjustments every set. This guide consolidates standard velocity zones by lift, a daily 1RM estimation algorithm, threshold-setting for velocity loss, and a weekly autoregulation decision tree. The goal is not another reference chart but a quantitative framework you can act on. Every numeric value below was validated with 800Hz sampling; on 100–200Hz devices the velocity error widens to about ±0.04 m/s, which renders many of the threshold-based decisions unreliable. Treat this as a citable reference for coaches and self-directed athletes who already use VBT.

For nearly every barbell movement, load and velocity follow a linear (or near-linear) relationship of the form y = ax + b, where y is velocity, x is %1RM, and slope a varies by lift and individual. Once that regression is built, a single warm-up rep is enough to estimate 1RM for the day.

Back squat, for example, follows V = -0.0123x + 1.32 on average. If a 60% rep registers 0.65 m/s, that velocity implies x ≈ 54.5%, suggesting today's true 1RM is slightly higher than the prior estimate. A reading 0.06 m/s above or below the expected velocity at 60% corresponds to roughly ±5 kg of estimated 1RM change.

Coefficients above are means from 800Hz IMU datasets aggregated over 8–10 weeks. See the load-velocity profile guide to derive your individual regression.

Recommended velocity zones depend on goal: maximal strength 0.30–0.50 m/s, strength-speed 0.50–0.75, power 0.75–1.00, speed-strength 1.00–1.30, and explosive speed >1.30 m/s.

Each zone targets a different neural adaptation. The 0.30–0.50 m/s zone lowers recruitment thresholds and is the strongest stimulus for raw 1RM. The 0.75–1.00 m/s zone improves firing rate, increasing RFD. "Lifting fast" is not always optimal — the right zone depends on the right goal.

In practice, contrast methods that combine two zones in one session work well. For example: back squat at 0.40 m/s (3×3 @ 85%) followed by jump squat at 0.95 m/s (4×3 @ 30%). Post-activation potentiation typically lifts jump power 5–8%. See the hex bar jump squat guide.

A 5-step daily 1RM estimation routine: (1) load the personal regression for the lift; (2) perform one rep at 60% of the prior 1RM estimate; (3) record MCV with the 800Hz IMU; (4) plug velocity into the regression to derive today's 1RM; (5) recompute the rest of the session loads against today's 1RM.

Edge cases matter. Anomalously fast warm-up velocities (e.g., 0.80 m/s at 60%) often indicate measurement error, not a 30 kg PR — verify bar path and recheck. Conservative protocols cap daily 1RM swings at ±5%.

The table interprets a single back squat warm-up rep at 60% of estimated 1RM.

Velocity loss (VL) is the percent drop from the first to the last rep within a set. Set VL by goal: 15–20% for maximal strength, 10% for power, 5% for explosiveness. Pareja-Blanco et al. (2017) compared 20% and 40% VL groups and found similar 1RM gains but double the vertical-jump improvement in the 20% group.

Weekly autoregulation logic: estimate daily 1RM in the first session of every week. If the moving average is 2%+ above last week, increase load 5%. If it is 1–2% above, hold. Within ±1%, hold. If 1%+ below, drop load 5% and add a recovery day. Three consecutive weeks of stagnation or decline triggers an early deload.

For implementation pseudocode and the PoinT GO app workflow, see the autoregulated velocity guide. Pair it with Sánchez-Medina & González-Badillo (2011) for the threshold-setting evidence base.