Optimal Sled Load for Resisted Sprints: The Velocity Decrement Research

Ask five sprint coaches how heavy a sled should be and you will get five different answers — 10% bodyweight, 20%, 50%, sometimes numbers north of 80%. Part of the confusion is honest: for two decades the literature itself was split, with older guidelines built around light loads to preserve sprint mechanics and a newer wave of research arguing that meaningfully heavier loads are where the real adaptation happens. Both camps have data behind them, which is exactly the problem — they were answering different questions.

The variable that actually resolves the argument is not the sled's weight in kilograms or a fixed percentage of bodyweight. It is velocity decrement — how much the sled slows the athlete down relative to an unloaded sprint. Two athletes pulling a sled loaded to 20% of bodyweight on different surfaces can experience completely different resistance because friction coefficients vary by a factor of two or more between rubber tracks, turf, and worn asphalt. Velocity decrement strips friction and hardware out of the equation and measures the one thing that matters: how much the load actually changed the athlete's sprint mechanics and force output.

Bodyweight-percentage prescriptions assume a constant relationship between sled weight and resistance experienced by the athlete. That assumption breaks the moment you change surfaces. A sled loaded to 20% of a 90 kg athlete's bodyweight (18 kg of added mass, plus the sled's own weight) generates meaningfully more drag on artificial turf than on a polished indoor track, because the coefficient of friction between sled runners and turf fibers is roughly double that of a smooth hard surface. Two teams following the identical '20% bodyweight' program card could be training completely different stimuli without anyone noticing.

Velocity decrement (often written %Vdec) sidesteps this entirely. You time or track an athlete's peak velocity over a short unloaded sprint — say, a 6.4 m/s peak over 20 m — then add sled resistance until peak velocity drops to a target percentage of that baseline. If loaded peak velocity falls to 5.1 m/s, that is a 20% velocity decrement, regardless of what the sled weighs or what surface you are training on. It is a self-calibrating measure that travels between athletes, surfaces, and equipment.

Three studies define most of what we currently know about sled load selection, and they were not all measuring the same outcome — which is why the recommendations from each can look contradictory at first glance.

Notice that Cross et al. answered a mechanical question (what load makes an athlete's legs produce the most power against resistance) and found the answer was surprisingly heavy — around 50% Vdec. Petrakos and colleagues answered a transfer question (what load actually makes athletes run faster in competition) using mostly light-load studies available at the time, and found small but real benefits. Neither study is wrong. They tell you that heavy loads build a different quality — horizontal force capacity — than light loads, which better preserve the velocity and technical qualities of unloaded sprinting.

Sprinting is, mechanically, a repeated application of horizontal force against the ground within a very short ground contact time. Samozino and Morin's force-velocity profiling work has shown that sprinters vary in whether their limiting factor is force capability (how hard they can push) or velocity capability (how fast they can cycle their limbs) — and that most athletes have some degree of imbalance between the two.

A heavy sled (35–50% velocity decrement) forces a much longer ground contact time and a more pronounced forward trunk lean, which biases the movement toward the acceleration-specific mechanics that build horizontal force: greater hip and knee extension torque per stride, higher activation of the posterior chain, and — critically — practice producing force at slower velocities, which is exactly where force-deficient athletes need to improve. A light sled (0–15% Vdec) barely changes trunk angle or contact time; the athlete is essentially sprinting normally with a small added resistance, which preserves stride frequency and the neuromuscular pattern of max-velocity running.

This is why prescribing sled load off a single fixed percentage for an entire roster is a mistake independent of the percentage you pick. An athlete who is already force-dominant and velocity-deficient (common in taller, more powerful athletes who accelerate well but plateau at top speed) gains little from more heavy sled work — they need overspeed and max-velocity exposure instead. An athlete who is velocity-dominant and force-deficient (common in athletes with high stride rate but a mechanically weak drive phase) is the one who benefits most from the heavier 30–50% Vdec zone that Cross et al. identified.

In practice, most programs benefit from using velocity decrement zones rather than a single fixed load, rotating emphasis across a training block. A simplified zone structure used by several sprint programs:

A sample undulating microcycle for an off-season block might run heavy sled (35% Vdec, 6 x 20 m, full recovery of 3–4 minutes between reps) on the primary strength day, moderate sled (15% Vdec, 6 x 30 m) on a secondary day, and unloaded max-velocity work (flying 20s or similar) on a third day to keep the top-end speed qualities from atrophying while the force work is emphasized elsewhere. Volume for heavy sled work should stay conservative — 6 total reps of 15–20 m is a reasonable ceiling per session, since contact times and eccentric loading on the hip extensors climb sharply as load increases and recovery demand rises with it.

The most frequent error is not choosing the wrong percentage but never checking whether the percentage you picked is producing the stimulus you think it is. A coach who loads every athlete on the roster to '20% bodyweight' without testing surface friction may be delivering a 12% velocity decrement to one athlete and a 24% decrement to another, simply because one runs in worn spikes on damp turf and the other runs in fresh cleats on dry turf. Test velocity decrement directly, at least at the start of a training block and again whenever the surface or footwear changes.

A second common mistake is chasing the heaviest possible load because '50% Vdec maximized power in the research,' without checking whether that athlete's limiting factor is actually force. Loading a velocity-deficient sprinter — someone with a strong drive phase who can already produce force well but struggles to hold stride frequency at top speed — into the heavy zone all off-season does not address their bottleneck and can reinforce a slower, more force-dominant stride pattern they need to move away from, not toward.

A third mistake is skipping the unloaded baseline sprint before adding load. Without a fresh baseline peak velocity measured that same session (fatigue, surface, and weather all shift baseline), the percentage decrement calculated against an old number will be wrong, sometimes by a wide enough margin to put an athlete in the wrong training zone entirely.

This is a straightforward field test that takes about 15 minutes per athlete once you have a velocity-tracking device (GPS unit, radar, or a phone app with sprint tracking):

1. Step 1 — Baseline. Run one or two unloaded 20 m sprints from a standing start, full recovery between. Record peak velocity from the faster effort. This is your 0% Vdec reference for the day.
2. Step 2 — Incremental loading. Attach a light sled load (roughly 10% bodyweight) and run a single 20 m sprint. Calculate the velocity decrement against the baseline: (baseline velocity − loaded velocity) / baseline velocity × 100.
3. Step 3 — Adjust and repeat. If decrement is under your target zone, add load (typically in 10% bodyweight increments) and repeat, with 3–4 minutes recovery between attempts. If it overshoots, remove load.
4. Step 4 — Record the working load. Once you land within 2–3 percentage points of your target zone (light, moderate, or heavy), record that load in kilograms for that athlete on that surface. Re-test every 4–6 weeks or after a surface/footwear change — the load in kilograms will drift as the athlete gets stronger even if the target %Vdec stays the same.

Most athletes will not need more than three or four loading attempts to land in the target zone once you have a rough starting estimate from Step 1's baseline speed.

1. Cross, M.R., Brughelli, M., Samozino, P., Brown, S.R., & Morin, J.B. (2017). Optimal Loading for Maximizing Power During Sled-Resisted Sprinting. International Journal of Sports Physiology and Performance, 12(8), 1069–1077.
2. Petrakos, G., Morin, J.B., & Egan, B. (2016). Resisted Sled Sprint Training to Improve Sprint Performance: A Systematic Review. Sports Medicine, 46(3), 381–400.
3. Cahill, M.J., Oliver, J.L., Cronin, J.B., Clark, K.P., Cross, M.R., & Lloyd, R.S. (2019). Sled-Pull Load-Velocity Profiling and Implications for Sprint Training Prescription in Young Athletes. Journal of Science and Medicine in Sport, 22(6), 668–672.