How to Test Your 1RM Safely (and Why You Often Shouldn't)

Your one-rep max is the heaviest load you can lift once with acceptable technique. It matters less as a badge than as a unit of measurement: nearly every structured strength programme prescribes work as a percentage of it. Five sets of three at 85% is meaningless without a number to take 85% of.

So the practical question is not whether you can lift a maximal single. It is whether you have a reliable enough estimate to programme from — and for that, an estimate is usually better than a test.

A true max attempt carries real costs. Technique degrades under maximal load, which is where most lifting injuries happen. It is neurologically fatiguing enough to compromise the following few days of training. It needs competent spotters on some lifts. And it is highly sensitive to sleep, stress and how the day is going, so a single bad attempt can under-report your actual capacity by a substantial margin.

An estimate from a set of three to five reps has none of those problems. It can be taken during normal training, repeated every few weeks, and it is accurate enough that the difference rarely changes what you would programme.

The One Rep Max Calculator does the conversion using both the Epley and Brzycki formulas and averages them.

Epley (1985) is 1RM = weight × (1 + reps/30). Brzycki (1993) is 1RM = weight × 36 / (37 − reps). Both were fitted to real lifting data and both are widely used.

They diverge in a predictable direction: Epley reads slightly higher, Brzycki slightly lower, and the gap widens as reps increase. At five reps with 100 kg, Epley gives 116.7 kg and Brzycki 112.5 kg — a 4 kg spread. At three reps the two agree closely.

Averaging them is a reasonable way to avoid committing to either one's particular bias, which is what the calculator reports as its headline figure.

Both formulas assume a fairly consistent relationship between reps and percentage of maximum. That assumption holds well at low reps and breaks down as reps climb, because sets of fifteen or twenty are limited by muscular endurance and fatigue tolerance rather than by maximal strength.

Individual variation compounds it. Two lifters with an identical 1RM can differ substantially in how many reps they manage at 80% — muscle fibre composition and training history both feed in. Someone trained on high-rep work will out-rep someone trained on singles at the same relative load.

The practical rule: use five reps or fewer for the best estimate, treat six to ten as usable, and treat anything above twelve as a rough indication. The calculator warns above ten reps and refuses above fifteen, where Brzycki's denominator starts heading toward zero and the maths stops being meaningful.

The estimate is only as good as the set behind it. A set taken well short of failure will under-report; a set with degraded technique will over-report.

Once you have a 1RM, the percentage table converts it into working weights. The broad conventions are well established: strength work sits at 85–95% for one to five reps, hypertrophy work at 67–85% for six to twelve, and muscular endurance below 67% for twelve or more.

These bands overlap and the boundaries are soft. Meaningful muscle growth happens across a wide range of loads provided sets are taken close enough to failure, and strength develops at moderate loads too. The percentages are a planning convenience, not a physiological law.

The table on the calculator runs from 50% to 100% in 5% increments with approximate rep counts alongside, which is enough to programme most sessions directly.

There are legitimate reasons to attempt an actual maximum. Competitive powerlifters need to practise maximal singles because that is the sport. Some programmes are built around tested maxes and periodise toward a test week. And occasionally it is worth knowing where you genuinely stand.

If you do test, do it properly: full warm-up with progressively heavier singles, long rests between attempts, competent spotters, and a hard stop after two or three failed attempts. Test one lift per session, not three.

Do not test when under-slept, ill, or in a steep calorie deficit. All three reduce maximal output enough to produce a misleadingly low number, which then propagates into every training weight derived from it.

Once you have a 1RM, the obvious next question is whether it is any good. Published strength standards express lifts as multiples of body weight — a common intermediate benchmark is a bodyweight bench press, a 1.5× bodyweight squat and a 2× bodyweight deadlift, with novice and advanced tiers either side.

They are worth treating loosely. Standards vary by source, by limb length, by training age and by sex, and they say nothing about health. A tall lifter with long femurs will squat less than a shorter one of identical strength, simply because the bar travels further.

Their genuine use is as a rough compass rather than a scoreboard: if one of your lifts sits far below the others relative to standards, that is usually a signal about where to put your attention next.

Every four to eight weeks is sensible for most people, or whenever a working weight starts feeling noticeably easier than the prescribed percentage should.

Newer lifters progress fast enough that a 1RM estimate goes stale within weeks — for them, adding weight when the prescribed reps become easy is a more practical guide than any percentage table. Experienced lifters change slowly enough that quarterly re-estimation is plenty.

The estimate is a tool for programming, not a measure of your worth as a lifter. Re-estimate when it stops describing your training accurately, and otherwise get on with the training.