Strength training loads and your one rep max
7 min read · Updated 2026-10-12 · FitCalc
Testing a true one rep max is expensive and risky, so strength is usually estimated from a set you can already do comfortably. The estimates are useful, but only if you understand that they are equations rather than measurements, and that they drift as your fatigue state and your progress change them.
What a one rep max means and when it matters
A one rep max is the largest load you could lift once, in a given exercise, under ideal conditions. In practice it matters less than the percentage of it you can train at, and that is exactly why the number is tracked: the percentages are what program design is built on.
A bench press one rep max of 100 means the working weight for five reps sits around 78, and for three reps around 93. Without the anchor, every set in a program becomes guesswork.
Why the formulas disagree
The common estimates — Epley, Brzycki, Lombardi, Lander, O'Conner, Wathan — all assume that the relationship between load and repetitions is roughly linear, and then pick a different line through the data. Epley is simply load times one plus reps divided by thirty. Brzycki is load times thirty-six divided by reps plus thirty-seven.
On a set of five, these give 78.6 and 85.7 for the same lift: an 8.5 per cent spread. On a set of ten they diverge much further, and near a true max they are all unreliable because fatigue, technique and motivation are not constant across the range. The formulas agree best in the middle of the range and disagree most at the extremes, which is the opposite of what people assume.
Getting a more honest number
The estimate improves when the test set is hard but not maximal, and when you record it accurately — the load lifted, the repetitions completed cleanly, and the grip in the straps if you used them. A set stopped one rep early understates the true max, and so does a set where technique broke down on rep eight.
Re-test every four to six weeks rather than weekly. Day-to-day variation in sleep, nutrition and motivation is larger than any progress you make in that time, and a number that moves constantly tells you nothing.
Using the figure well
Use the estimated max to set working loads, not to decide whether you are making progress. Rate of increase on your actual training loads — the bench going from 80 to 82.5 across eight weeks — is the signal worth reacting to.
And leave the margin. Training at a true one rep max builds very little and risks a lot. A rough rule is that a new personal best belongs in a single top set of a session where you are not already near failure on anything else.
Run your own numbers
Free calculators that answer the “how much” version of this question. They run entirely in your browser — nothing you type is uploaded or stored.
Common questions
How do I calculate my one rep max?
Estimate it from a hard but safe set: multiply the load by the number of reps using one of the standard formulas. Epley is load × (1 + reps ÷ 30), which is the simplest and widely used.
How accurate are one rep max calculators?
Within a few percent for sets of three to five reps, and considerably worse above ten reps or near a true maximum. A formula is an estimate from a submaximal set, not a measurement.
How often should I retest my one rep max?
Every four to six weeks. Day-to-day variation from sleep, nutrition and motivation is larger than the progress you make in between, so more frequent testing just produces noise.
FitCalc provides general information only and is not medical advice, diagnosis or treatment. Figures are illustrative and depend on your own circumstances. Always consult a qualified health professional about your situation. See our full disclaimer.