Yes. A 2026 study in the Journal of Strength and Conditioning Research found that a maximal-speed eccentric (lowering) phase raised mean propulsive velocity by roughly 6-23% in the jump squat and bench press throw compared with starting from a pause. It also raised estimated 1RM by about 5-6 kg, and test reliability stayed the same. The takeaway from this strength training research is that how you start a rep can change your test numbers as much as weeks of training.
This matters well beyond velocity-based training labs. Any lifter who tests, estimates a max, or compares sessions over time is comparing protocols as well as performances. If the protocol drifts, the data drifts with it.
Key Finding
The researchers compared two ways of building a load-velocity (L-V) profile. In one, lifters started from a still position and only moved upward (concentric-only, or CON). In the other, they lowered the load as fast as they could and then reversed into the lift (maximal intended eccentric velocity, or MAX).
The MAX condition produced faster reps at almost every load and higher estimated maxes. Its reliability was as good as or slightly better than the pause-start method. The authors conclude that eccentric speed instructions substantially change the L-V profile. They suggest a standardized fast-eccentric protocol as a complement to the usual concentric-only test, not a replacement for it.
Study Details
- Participants: 18 men
- Exercises: jump squat (JS) and bench press throw (BPT), which are ballistic movements where the load leaves the body or the lifter leaves the ground
- Loads: 20, 40, 60, and 80% of one-repetition maximum (1RM)
- Design: each L-V assessment was performed on two separate days to measure reliability
- Conditions: concentric-only (CON) vs. maximal intended eccentric velocity (MAX)
- Primary metric: mean propulsive velocity (MPV), the average bar speed during the accelerating part of the lift
From each lifter's velocity data the researchers derived three things. L0 is the theoretical load at zero velocity. V0 is the theoretical velocity at zero load. The third was an estimated 1RM for the back squat and bench press.
Results
Velocity at each load
- Jump squat: MPV was significantly higher in the MAX condition at every load, by 8.9-10.9%.
- Bench press throw: MPV was significantly higher at every load except 20% 1RM, by 6.1-23.3%.
Reliability
- MAX: ICC 0.70-0.87, CV 4.2-11.2% (described as high)
- CON: ICC 0.61-0.86, CV 4.8-10.4% (described as moderate to high)
The fast-eccentric method did not add noise. Its lower-bound ICC was actually a little higher than the pause-start method's.
Derived parameters
- V0 was higher with MAX in both the jump squat (+9.0%) and the bench press throw (+3.4%).
- L0 was higher with MAX in the bench press throw (+5%).
Estimated 1RM
| Exercise | Concentric-only | Max eccentric velocity | Difference |
|---|---|---|---|
| Back squat | 117.8 kg | 124.1 kg | +6.3 kg (~5.3%) |
| Bench press | 83.9 kg | 89.2 kg | +5.3 kg (~6.3%) |
The same lifters, tested within a short window, "gained" about 5-6 kg on paper because of the descent instruction alone.
Why a Faster Descent Raises Velocity
A fast lowering phase followed by an immediate reversal uses the stretch-shortening cycle. Research consistently suggests that a quick eccentric-to-concentric transition improves concentric output compared with starting from rest. Proposed mechanisms include stored elastic energy, greater muscle activation before the reversal, and more time to build force before the propulsive phase.
A pause-start removes most of that help. Neither method is wrong. They measure somewhat different qualities: concentric-only output versus the ability to use a rapid stretch-shortening action.
Limitations
- Small, all-male sample. Eighteen men is typical for this kind of reliability work, but the results may not transfer directly to women, adolescents, or lifters with very different training backgrounds.
- Estimated, not measured, 1RM. The squat and bench press maxes were projected from ballistic L-V profiles. A higher estimate does not prove a higher true 1RM. It shows the estimate depends on the protocol.
- Ballistic exercises only. Jump squats and bench throws let the lifter accelerate through the whole movement. It is not yet clear how big the effect would be in traditional squats or bench presses, where lifters decelerate near lockout.
This is a measurement study, not a training intervention. It tells us nothing directly about strength or hypertrophy adaptations over time. It is about how to test reliably, which is the foundation that any evidence-based training plan relies on.
What This Means for Your Training
The distinctive lesson is this: your test is only comparable to itself. A 6 kg jump in estimated squat 1RM could reflect real progress, or it could reflect bouncing out of the hole on test day when you paused last time. Here is how to apply that.
- Pick one descent style for testing and write it down. For example, use a "2-second pause at the bottom" or a "fast descent, immediate reversal." Keep that exact cue every time you test or estimate a max.
- Never compare across protocols. A pause-start rep and a touch-and-go rep at the same load are not the same data point. If you change protocols, treat the first new session as a new baseline.
- Expect about 5-10% higher velocity or e1RM from a fast stretch-reflex start in ballistic lifts. That margin is about the size of several months of progress for an intermediate lifter, so do not mistake it for a breakthrough.
- Consider testing both. The authors suggest the two protocols complement each other. A concentric-only test reflects starting strength. A fast-eccentric test shows how well you use the stretch-shortening cycle. A widening gap between the two may say something about your reactive qualities.
- Control the other variables too. Keep the warm-up, time of day, rest between attempts, and depth or range of motion consistent. The study's CV values of 4-11% show that even well-controlled tests have noise. Look at trends across several sessions rather than reacting to one result.
- Apply the same logic without a velocity device. Many lifters work out estimated maxes from rep performance. Tempo changes how many reps you can complete at a given load, so log the tempo with the set. If you do not, your rep-based estimates are comparing two different lifts.
How This Fits Into Your Training Log
Kenso does not measure bar velocity, but its AI Coach can review your logged sets and help you check whether an estimated max jump lines up with consistent conditions. Ask the Kenso AI coach how this applies to your program.
The Bigger Picture
It is easy to treat a number from a test as objective truth. This study is a reminder that numbers come from procedures. Change the procedure and the number moves, even when the athlete has not changed.
For lifters who care about progression, that is useful to know rather than discouraging. Consistent conditions make your data trustworthy, and trustworthy data makes good programming decisions possible. Kenso's rule-based progression engine works from the sets you log, so the more consistently you perform and record them, the better its recommendations reflect what is really happening in your training.
Conclusion
Hernandez-Davo and colleagues showed that a maximal-speed eccentric phase raises measured velocity by 6-23% and estimated 1RM by roughly 5-6 kg in the jump squat and bench press throw. Reliability held steady. Both protocols are valid, but they are not interchangeable. Choose one, standardize it, and compare like with like. Over months and years, that discipline separates real progress from noise in the measurement.
If you want a log that rewards that kind of consistency, Kenso is built for lifters who train with intention. Download Kenso on iOS, record your sessions the same way every time, and let your data show you where you are actually improving.
FAQ
Does lowering the bar faster make you stronger?
Not by itself, according to this study. It measured acute performance during testing, not long-term adaptation. The fast descent raised velocity and estimated 1RM within the same testing period, which reflects how the rep was performed rather than a change in strength.
Which load-velocity protocol should I use, concentric-only or fast eccentric?
Either can be reliable. The fast-eccentric protocol showed ICCs of 0.70-0.87 and the concentric-only protocol 0.61-0.86. What matters most is using the same protocol every time. The authors suggest the fast-eccentric version can complement, not replace, standard concentric-only testing.
How much does eccentric speed change estimated 1RM?
In this study, estimated back squat 1RM rose from 117.8 kg to 124.1 kg and estimated bench press 1RM from 83.9 kg to 89.2 kg with the fast-eccentric protocol. Both increases are roughly 5-6%.
Do these findings apply to regular squats and bench presses?
Not directly. The study used ballistic exercises, the jump squat and bench press throw, where lifters accelerate through the full movement. The general principle that tempo affects measured performance likely carries over, but the size of the effect in traditional lifts was not tested here.
Can I use this research without a velocity tracker?
Yes. The practical lesson is to standardize rep tempo and pauses whenever you test or estimate a max, and to record those details. That applies whether you estimate maxes from bar speed or from reps completed at a given load.
Citation
Hernandez-Davo JL, Sabido R, Asencio P, Boullosa D. Comparison Between Concentric-Only and Eccentric-Concentric Load-Velocity Relationships and its Derived Parameters in the Jump Squat and Bench Press Throw Exercises. Journal of Strength and Conditioning Research. 2026. DOI: 10.1519/JSC.0000000000005731. PubMed: https://pubmed.ncbi.nlm.nih.gov/42945128/