Velocity-Based Training vs Percentages: What a New Study Found
Quick answer: In a four-week trial published 11 June 2026 in Frontiers in Physiology, Wang and colleagues put 52 trained men through 12 back squat sessions. Groups that trained by bar speed, or were simply told to push every rep as fast as possible, gained roughly three times more jump height than the group following fixed percentages of their one-rep max.
Most programs tell you a number. Squat 75% of your one-rep max, four sets of six, and that is your Tuesday.
The number assumes you are the same person every Tuesday. You are not. You slept badly, you skipped lunch, you had a stressful morning, and the bar that flew last week now grinds. The percentage does not know any of that.
What the researchers actually did
Fifty-two trained men were randomly split into four groups of 13. Every group squatted three times a week for four weeks, and every group did the same accessory work: Romanian deadlifts, leg press and bench press, with identical sets and reps.
The only thing that changed was how the back squat itself was prescribed.
- PBT (percentage-based training). The traditional method. 75% of estimated 1RM for weeks one and two, 80% for weeks three and four, fixed reps, no feedback on how the bar moved.
- MIV (maximum intended velocity). Same sort of load, 70% to 75%, six fixed reps, but with one instruction repeated before and during every set: move the bar as fast as you physically can on the way up.
- VLT (velocity-loss threshold). Load picked from a slower bar-speed range of 0.55 to 0.65 metres per second, roughly 75% to 80% of 1RM. Sets ended when rep speed dropped 20% below a reference rep, somewhere between four and eight reps.
- TVZ+VLT (target velocity zone plus velocity loss). Load picked so the first rep moved at 0.75 to 0.85 metres per second, roughly 60% to 70% of 1RM, with sets ending once speed fell 15% from the first rep.
Before and after, everyone was tested on countermovement jump height, 10 and 30 metre sprints, and an isometric mid-thigh pull, which measures how much force you can produce against an immovable bar.
What they found
The jump test produced the clearest split, and it was not subtle.
Jump height gained over four weeks, by how the squat was prescribed
View the data
| Group | Percentages (PBT) | Velocity loss (VLT) | Max intent (MIV) | Speed zone (TVZ+VLT) |
|---|---|---|---|---|
| Change in countermovement jump height | 1.4 cm | 2.1 cm | 3.79 cm | 4.21 cm |
The gap between the speed-zone group and the percentage group was large by the standards of training research: a Hedges' g of 1.63, which is a way of saying the two groups barely overlapped. The max-intent group beat percentages too (g = 1.26).
Sprint times moved the same way, though less dramatically.
Seconds shaved off a 30 metre sprint
View the data
| Group | Percentages (PBT) | Velocity loss (VLT) | Max intent (MIV) | Speed zone (TVZ+VLT) |
|---|---|---|---|---|
| Change in 0 to 30 m sprint time | -0.05 s | -0.08 s | -0.14 s | -0.15 s |
Now the part that gets left out of most summaries. On raw force production, nothing separated the groups at all.
Maximum force improved in every group, and the method made no difference
View the data
| Group | Percentages (PBT) | Velocity loss (VLT) | Max intent (MIV) | Speed zone (TVZ+VLT) |
|---|---|---|---|---|
| Change in isometric mid-thigh pull peak force | 289 N | 256 N | 189 N | 198 N |
So the honest headline is narrower than "velocity training wins". Four weeks of squatting made everyone stronger. What changed with prescription method was how fast that strength could be expressed.
Why would telling someone to push faster change anything?
The loads were similar. The sets were similar. The one thing that separated the max-intent group from the percentage group was an instruction.
Intent is a training variable. When you decide to move a heavy bar explosively, your nervous system recruits motor units harder and earlier, even though the bar itself still moves slowly under 75% of your max. The bar speed you see is the outcome. The effort you put into accelerating is the input, and the input is what adapts.
The two velocity-guided groups got something extra: a number after every rep. When a screen tells you the last rep was slower than the one before, you either push harder or you stop. Both responses keep the quality of the set high, which is exactly what a fixed "four sets of six" cannot do on a bad day.
That is the researchers' proposed explanation, not something the study measured. They recorded no muscle activity, no motor unit behaviour and no neural drive, and they say so plainly. Nobody watched a nervous system here.
What this study doesn't prove
Four weeks is a mesocycle, not a training age. The authors state directly that this duration cannot support conclusions about muscle growth or long-term strength. Twelve sessions is enough to shift how you express force. It is not enough to change how much muscle you have.
The groups differed in several ways at once. Load, target speed, feedback, movement intent and the rule for ending a set all varied together. The paper calls its own results "package-level effects" and refuses to credit any single component. You cannot read this as proof that velocity loss thresholds work, or that feedback works, because nothing was isolated.
Nobody recorded what actually happened in the sessions. Completed reps, total tonnage and realised velocity loss were not retained, so the study cannot verify that the groups did equal work. If the speed-zone group happened to do more or less volume, that would be invisible here.
It was 52 young men at a sports university. Average age 21.5, all male, all athletes. The authors list female athletes, older adults, recreational lifters and elite lifters as groups this does not generalise to.
Nine outcomes were tested, which inflates the odds of a fluke. The researchers applied a Holm correction to guard against that, and were honest about which findings survived it and which did not. The rate-of-force-development result, often quoted as a velocity training benefit, did not survive (p = 0.245).
The funding looks clean. Support came from the Shaanxi Provincial Department of Education and a provincial social sciences programme, with no commercial ties and no declared conflicts of interest. No velocity-tracker company paid for this.
How to test this in your own training
The cheapest finding in the paper is the one that costs nothing. The max-intent group used no equipment at all. They were just told to push hard, and they added 3.79 cm to their jump.
- Pick one compound lift and one working weight. Squat, bench or deadlift, at a load you can already handle for the reps you normally do. Do not change the weight yet.
- Change the instruction, not the program. For the next six weeks, drive every concentric rep as fast as you can while keeping the same technique. The bar will not look fast at 75%. That is fine. The intent is the variable.
- Log the reps you actually complete at that weight. Not how it felt. The number. Same lift, same load, week by week, so you have something to compare.
- Add a one-word note on bar speed.
snappy,normal, orgrindyon your top set. Two seconds to write, and it is the closest thing to a velocity tracker you already own. - Compare at week six. If reps at the same weight have gone up, or
grindysets have becomenormalones, the instruction is doing something for you. If nothing moved, you have your answer just as cheaply. To compare weeks where the weight changed, run each top set through the one rep max calculator so you are looking at one number instead of two.
How the four methods compare
| Prescription method | How the load is chosen | When a set ends | Kit needed |
|---|---|---|---|
| Percentages (PBT) | Fixed share of your 1RM, planned in advance | Prescribed rep count, regardless of speed | None |
| Max intent (MIV) | Fixed share of your 1RM, same as PBT | Prescribed rep count, pushed explosively | None |
| Velocity loss (VLT) | Whatever weight moves at a target speed | When rep speed drops past a set threshold | Bar-speed tracker |
| Speed zone plus velocity loss | Whatever weight makes rep one hit a speed window | When rep speed drops 15% from rep one | Bar-speed tracker |
Running your own version of this without a tracker
You do not need a linear position transducer to act on this. You need six weeks of reps at a known weight, written down, and the willingness to look at them.
That second part is where it usually falls apart. Six weeks of squats live in your notes app as 24 separate scribbles, and answering "did my reps at 100 kg go up?" means scrolling back through all of them.
Gym Note Plus is built for exactly that gap. Keep typing sessions the way you already do, in plain text, and it turns them into per-exercise history without you formatting anything:
- Per-exercise trends so a six-week comparison is one screen, not an evening of scrolling.
- Plain text in, structure out. No dropdowns, no picking exercises from a list between sets.
- Notes stay attached to the set, so a
grindytag sits next to the weight it belongs to.
Paste your last squat session into the free notes to workout translator if you want to see what your own notes look like as a chart before you install anything.
Four weeks of the same squat, at nearly the same weight, produced three times the jump improvement when the lifters were simply told to push harder. The instruction was free. Finding out whether it works for you costs a logbook.
Frequently Asked Questions
What did the 2026 study on velocity-based training find?
Wang and colleagues, publishing 11 June 2026 in Frontiers in Physiology, compared four back squat prescriptions in 52 trained men over four weeks. Velocity-guided and maximum-intent groups improved countermovement jump height by 3.79 to 4.21 cm, against 1.40 cm for percentage-based training. Maximum isometric force improved equally in all groups.
Is velocity-based training better than percentage-based training?
For explosive qualities over a short block, this study says yes. Bar-speed-guided squatting produced larger jump and sprint improvements than fixed percentages. For maximum force production, it found no difference at all. Velocity-based training changes how quickly you express strength more than how much strength you have.
Do you need a velocity tracker to benefit from this research?
No. The maximum intended velocity group in Wang et al. 2026 used no measuring device. They trained at 70% to 75% of their one-rep max with fixed reps and a single instruction to move every rep as fast as possible, and beat percentage-based training on jump height and power.
What is a velocity loss threshold in weight training?
A velocity loss threshold ends a set once rep speed drops by a set percentage below the fastest rep. In this study, one group stopped at a 20% drop from a reference rep and another at 15% from rep one, producing sets of four to eight reps instead of a fixed count.
Does pushing the bar fast make you stronger?
Not measurably stronger in maximum force, according to this trial. Isometric mid-thigh pull force rose by roughly 189 to 289 newtons across all four groups with no reliable difference between them. What explosive intent changed was jump height, sprint time and peak power.
Final Takeaways
- The instruction is free, so test it. The max-intent group used no equipment and still tripled the percentage group's jump improvement.
- Explosive qualities moved, maximum force did not. Every group got stronger on the isometric pull by a similar amount, so do not read this as a strength result.
- Four weeks in 52 young male athletes. Nothing here says anything about hypertrophy, women, older lifters, or what happens after month one.
- Percentages assume you are identical every session. If that assumption bothers you, our guides to when to increase weight lifting and RPE vs RIR cover the low-tech ways to autoregulate without a bar-speed sensor.
- Your log is the experiment. Reps at a fixed weight, six weeks, one lift. That is the whole protocol.
Josh Ibbotson
Josh is the creator of Gym Note Plus, building tools that make workout tracking as simple as taking notes.
Ready to try it out?
Download Gym Note Plus and start tracking your workouts the way you want.
Download on App Store
