What a one-rep max estimate is doing
Your one-rep max is the heaviest weight you could lift once with acceptable technique. Rather than finding it by attempting it, these formulas infer it from a submaximal set: if you can do 100 kg for five hard reps, there is a fairly predictable weight you could do for one.
Epley — 1RM = w × (1 + reps/30) — and Brzycki — 1RM = w × 36/(37 − reps) — are the two most widely used. At five reps they differ by about 4%. At twelve they differ by considerably more, which is the honest argument for showing you all four rather than picking one and implying a precision that is not there.
Why the estimate drifts at high reps
These equations assume everyone has the same rep-to-percentage relationship. They do not. Training history, muscle fibre distribution and the lift itself all change how many reps a given percentage allows — a squat typically permits more reps at 80% than a bench press does — and the differences compound as the set gets longer.
Estimate from a heavy set, not a long one
A set of 3–5 reps taken close to failure gives the most reliable estimate. A set of 15 is measuring your endurance and will read high. If the only set you have is a long one, treat the result as a rough ceiling and re-test with something heavier before you program from it.
Turning a max into a week of training
The percentage table above converts the estimate into working weights: roughly 80–95% of max for strength work in the 1–5 rep range, 65–80% for hypertrophy work at 6–12. Treat those as opening offers. Your effective max moves with sleep, stress and how the previous session went, so a fixed percentage will feel wrong on a fair number of days — the weight on the bar should answer to how the first working set moves, not only to arithmetic.
Which is most of the argument for not programming by hand. Dietly builds the week from your latest scan, the weak point it named, and the equipment you actually have — and when the rack is taken or a shoulder is complaining, you say so in a sentence and the session is rewritten around it.
Common questions
- How accurate is a 1RM calculator?
- Good within a few percent for sets of 1–5 reps, and increasingly unreliable above that. The formulas assume a fixed relationship between reps and percentage of maximum, but that relationship varies by person, by lift and by training history — someone with high work capacity will get more reps at 85% than the equations predict, and their estimate will read high as a result.
- Which 1RM formula is best?
- None of them is reliably best, which is why this page shows four. Epley and Brzycki are the most cited and agree closely at low reps — they diverge as reps climb, with Brzycki reading lower. Use the average, treat the spread as the error bar, and re-estimate from a heavy set rather than a light one.
- Should I actually test my one-rep max?
- Rarely, and not as a beginner. A true max attempt is technically demanding and costs days of recovery, and the estimate from a heavy set of 3–5 is close enough to program from. Testing has a place for competitive lifters and for peaking blocks; for everyone else it is risk without much return.
- How do I use percentages of my 1RM?
- As a starting point, not a prescription. Strength work generally sits at 80–95% for 1–5 reps, hypertrophy work at 65–80% for 6–12. Your max changes week to week with sleep, food and fatigue, so a fixed percentage will sometimes be too heavy and sometimes too light — adjust by how the first working set actually feels.