Bar speed tells you what percentage of your 1RM you are lifting (R² = 0.961)
In short
On the full squat, mean propulsive velocity predicts relative load (%1RM) with R² = 0.961 and a standard error of estimate of 0.069. Mean velocity is effectively as good at R² = 0.958, while peak velocity is the weakest of the three at R² = 0.795. Across 85 players split into senior, under-18 and under-16 groups, no velocity variable differed significantly between age categories at any relative load — the same bar speed means the same %1RM regardless of age.
Measure how fast the bar moves on a full squat and you have read off what percentage of your one-rep max is on it. The relationship between mean propulsive velocity (MPV) and relative load came out at R² = 0.961, with a standard error of estimate of 0.069. You can work backwards to today's relative load without ever lifting a true max.
The numbers come from a progressive loading test taken all the way to 1RM, with a linear velocity transducer on the bar. All three velocity variables related to relative load at p < 0.001 — but they were not equally accurate.
Which of the three velocities should you watch?
- Mean propulsive velocity (MPV) — R² = 0.961, standard error of estimate 0.069. The most accurate of the three.
- Mean velocity (MV) — R² = 0.958, standard error of estimate 0.062. Effectively equivalent to MPV, and its error band is marginally tighter.
- Peak velocity (PV) — R² = 0.795, standard error of estimate 0.132. Nearly double the error. If your device only reports peak velocity, it is the weakest basis for prescribing relative load.
Does the velocity standard shift with age?
It did not. The 85 players came from one club, split into senior, under-18 and under-16 categories, and at no relative load did MPV, MV or PV differ significantly between groups. Mixed-effects polynomial models likewise showed no consistent effect of age on the overall level of the load-velocity relationship. The same velocity means the same %1RM, and nothing in this data justifies a separate chart per age group.
Can this cut how often you test a max?
A relative strength score is built from squat, bench press and deadlift one-rep maxes. Testing those for real is expensive — injury risk on the attempt, fatigue for days after. If velocity can estimate relative load, you can drop true testing to roughly once a cycle and use bar speed in between to pick the weight that matches the day you are actually having. On timing the tests themselves, see when to test your max.
Is this the same as knowing when to end a set?
No. Here velocity is choosing the load — what percentage of your max is on the bar right now. Whether you cut the set at 10% velocity loss or push to 20% is about ending the set. Same device, different question. The set-termination side is covered in velocity loss: 10% vs 20%.
The participants were 85 female soccer players with at least six months of squat training, twice a week. Load-velocity relationships are individual, so what this study establishes is that the relationship is strong and age-independent — not a universal velocity-to-%1RM lookup table you can copy off the page. Your own baseline still needs one progressive loading test.
Frequently asked questions
Can bar speed tell you what percentage of your 1RM you are lifting?
On the full squat, quite accurately. Mean propulsive velocity related to relative load at R² = 0.961 with a standard error of estimate of 0.069. You still need one progressive loading test to establish your own baseline.
Should you use mean propulsive velocity or peak velocity?
Use mean propulsive velocity (R² = 0.961) or mean velocity (R² = 0.958). Peak velocity came in at R² = 0.795 with a standard error of estimate of 0.132 — close to double the error of the other two.
Does the same velocity mean a different %1RM as you get older?
Not in this data. Senior, under-18 and under-16 groups showed no significant velocity differences at any relative load, and mixed-effects models found no consistent effect of age. The same speed can be read as the same relative load.
Does velocity measurement replace 1RM testing entirely?
Not entirely. Velocity estimates relative load (%1RM), but turning that percentage into an actual weight — or into a relative strength score — still requires a reference 1RM value. Treat it as a way to test less often, not never.
Source: PubMed