One Rep Max Calculator
Estimate your 1RM from any set with seven published formulas — and see the honest spread between them.
One Rep Max Calculator
Results recalculate instantly on every keystroke. Nothing you type is transmitted.
What this result does not account for
- All seven equations are population regressions fitted on bench, squat and deadlift data; five of the seven do not even return the lifted weight when given a single repetition.
- Accuracy is roughly 2-5% at 1-6 reps and degrades steeply above 10, where the formulas increasingly measure muscular endurance.
- Transfer to isolation exercises is poor because local endurance dominates those movements.
- The input assumes consistent tempo and full range of motion; partial or bounced reps invalidate it.
- Individual fibre-type distribution determines rep capacity at a given percentage and cannot be inferred from a single set.
In short: From 100 kg for 5 reps the seven standard equations return anything from 112.5 kg (Brzycki, O'Conner) to 119.0 kg (Mayhew) — a mean of 115.5 kg and a 6.5 kg spread. Any calculator showing one number is hiding that disagreement. The spread widens sharply with reps: at 15 reps it is 21.9% of the estimate, which is why you should test at 3-6 reps.
Formula
Three functional shapes are represented: linear (Epley, O'Conner, Lander), hyperbolic (Brzycki), power-law (Lombardi) and exponential (Wathan, Mayhew). They agree closely at low reps because that is where all of them were fitted, and diverge sharply above ten, which is the practical argument for testing at three to six repetitions.
Worked Example
- Enter the weight and the reps you completed with it.
- State any reps left in reserve honestly — the equations assume you went to failure.
- Read the mean, then read the spread: the spread is the real precision of the estimate.
- Use the training max at 90%, not the raw estimate, for programming percentages.
- If the spread looks uncomfortable, retest at three to six reps and it will narrow.
100 kg for 5 reps to failure: Brzycki and O'Conner say 112.5 kg, Mayhew says 119.0 kg, and the seven-formula mean is 115.5 kg with a 6.5 kg spread — 5.6% of the estimate. At 15 reps the same calculation spreads 21.9%, which is the argument for low-rep testing in one number.
Strengths & Limits Of This Model
Where this engine is strong
- Runs seven published equations instead of hiding behind one
- Reports the spread, which is the estimate's real precision
- Publishes the identity test that five of the seven formulas fail
- Handles reps in reserve properly, before the equations rather than after
Where it stops
- Cannot replace a tested maximum for competition purposes
- Depends entirely on the input set being genuinely near failure
- Unreliable for isolation movements
- Refuses high-rep inputs by design, which some users will want anyway
Practical Use Cases
Setting programme percentages without testing a max
Most percentage-based programmes need a 1RM you would rather not test. A set of three to six to failure gives an estimate within a few percent, and taking 90% of it as a training max builds in the margin that makes the programme completable on an ordinary day.
Deciding meet attempts
Use the conservative end of the range rather than the mean. An opener you can triple, a second attempt near the low estimate and a third beyond it is the standard structure, and the spread shown here tells you how much genuine uncertainty sits under those numbers.
Tracking progress between test days
Estimated maxes from consistent submaximal sets track real strength changes well, provided you use the same formula and the same rep range every time. Switching from Brzycki to Mayhew mid-cycle will manufacture a 6% gain that never happened.
Comparing lifts across a training group
Estimated maxes let a coach compare athletes without putting everyone under a maximal bar on the same afternoon, which is both a time and a safety consideration. Keep the rep range identical across athletes or the comparison inherits the formula divergence.
Methodology & Editorial Standards
Seven published submaximal prediction equations are computed in their original forms: Epley (1985), Brzycki (1993), Lombardi (1989), O'Conner et al. (1989), Lander (1985), Wathan (1994) and Mayhew et al. (1992). The headline figure is the unweighted mean, which cancels individual formula bias, and the minimum, maximum and spread are reported alongside it because the disagreement between equations is the honest expression of the estimate's precision. Reps in reserve are added to performed reps before the equations are applied, since all seven assume a set carried to momentary failure. Inputs above 20 reps are refused: every equation was fitted on low-to-moderate rep data and Brzycki's denominator approaches zero at 37 repetitions.
Computation runs in IEEE-754 double precision at full internal precision; rounding to two decimal places occurs strictly at the display layer, so no cumulative drift enters the result. All monetary outputs use accounting presentation — grouped thousands, two decimals, negatives in parentheses — so figures can be transcribed directly into a model or working paper. Division-by-zero and out-of-domain inputs return an em-dash rather than a misleading number.
This engine was reconciled against an independent reference implementation and hand-verified for the worked example above before release. Our full five-stage review process is published on the About Us page.
Disclaimer. This calculator is provided for informational and modelling purposes only and does not constitute financial, tax, legal, medical, or engineering advice. Verify all figures with a qualified professional before acting on them.
One Rep Max Calculator — 20 Expert FAQs
20 analyst-written answers to the questions practitioners actually ask — optimised for voice and answer-engine retrieval.
How do you calculate your one rep max?
Take a set to failure, then apply a prediction equation. Epley gives weight × (1 + reps/30); Brzycki gives weight × 36/(37 − reps). For 100 kg × 5 that is 116.7 and 112.5 kg respectively. This page runs seven published equations and reports the mean plus the spread, because the disagreement is part of the answer.
Which 1RM formula is most accurate?
None universally. Brzycki is NCAA-validated and conservative, favoured in research; Epley is the most widely implemented; Wathan tends to fit power athletes well. Within one to six reps they agree within a few percent, so the choice barely matters. Above ten reps they diverge enough that the choice matters more than the lifting.
How accurate are one rep max calculators?
Within roughly 2-5% of a tested max for sets of one to six reps in trained lifters. Accuracy degrades steeply above ten reps as muscular endurance rather than maximal strength determines the outcome. At 15 reps the seven formulas here disagree by nearly 22%, which is a fair statement of the real precision.
Why do 1RM formulas give different answers?
They were fitted to different populations with different functional shapes — linear, hyperbolic, power-law and exponential. Each fits its own data well and extrapolates differently outside it. A wide spread between them is a signal that your rep count is too high for reliable prediction.
How many reps should I use to estimate my 1RM?
Three to six, taken to or within one rep of failure. Below three the formulas add almost nothing to what you already lifted, and you have exposed yourself to a near-maximal load anyway. Above ten the estimate becomes a statement about your endurance rather than your strength.
What is a training max and why is it 90%?
A training max is a deliberately conservative figure — about 90% of your estimated or tested maximum — used as the base for programme percentages. Popularised by Wendler's 5/3/1. Your genuine capacity fluctuates daily with sleep, stress and food, and programming from a best-ever number produces sessions you fail.
Do these formulas work for the deadlift?
Reasonably, though the deadlift tends to produce fewer reps at a given percentage than the squat or bench, so estimates from high-rep deadlift sets run high. Grip and lower-back fatigue also end deadlift sets before the prime movers are truly finished, which contaminates the input.
Can I use a 1RM calculator for bicep curls?
Not reliably. These equations were fitted on bench, squat and deadlift data, and isolation movements are dominated by local muscular endurance. Someone curling 20 kg for 15 reps does not have the maximum the formulas predict, and testing a true curl max is rarely worth the elbow risk anyway.
What does reps in reserve mean for my estimate?
Every one of these equations assumes the set ended at failure. If you stopped with two reps left, the honest input is your reps plus those two. Each unreported reserve rep understates your maximum by roughly 3-4%, which is why a comfortable set of five produces a disappointing prediction.
Is it safe to test a true one rep max?
It carries more risk than submaximal testing, particularly on squat and deadlift and particularly for novices, where technique degrades under maximal load. The NSCA and NASM both recommend submaximal protocols for general fitness clients. If you do test, use safety bars or a competent spotter and build up in small increments.
Why does Brzycki break at high reps?
Its denominator is 37 − reps, so at 37 reps you divide by zero and the prediction goes to infinity. Well before that it becomes absurd: at 36 reps it claims a max of 3,600 kg from a 100 kg set. This page refuses inputs above 20 reps for exactly this reason.
How often should I recalculate my 1RM?
Every four to eight weeks, or at the end of a training block. Programming from a stale maximum means gradually underloading as you get stronger, which quietly stalls progress. A single submaximal test set at the end of a block costs one exercise and keeps the percentages honest.
What percentage of 1RM should I train at?
For maximal strength, 85% and above for low reps. For hypertrophy, roughly 65-85% with sets taken near failure — and the evidence suggests load matters less than proximity to failure across a surprisingly wide band. Below 50% you are training technique and endurance rather than driving much adaptation.
Do 1RM formulas work the same for men and women?
The equations themselves are not sex-specific and are applied identically. Some research suggests women perform more reps at a given percentage of maximum on average, particularly on upper-body lifts, which would make these formulas run slightly conservative. The difference is smaller than the spread between the formulas.
Should beginners use a 1RM calculator?
Use it, but treat the number as soft. Novice lifters have poor technique under load and limited ability to judge failure, and both distort the input. Early progress is also fast enough that any estimate is stale within weeks, so recalculate often and program conservatively.
What is a good one rep max for bench press?
Bodyweight for one rep is a common intermediate benchmark for men, with 1.5× bodyweight considered advanced; women's benchmarks sit near 0.6× and 1.0× respectively. These are rough population markers, not targets — limb lengths alone change bench leverage enough to make cross-person comparison unfair.
Is this one rep max calculator free to use?
Yes. It is free, requires no account, and has no usage limits. ApexConverter is funded by contextual advertising, never by selling user data.
Is my data sent to a server?
No. The engine runs as Vanilla JavaScript inside your browser under our Zero-Server Client-Side Execution model. Your figures are computed locally and are never transmitted, logged, or stored.
How accurate is this calculator?
It applies the standard closed-form formula in IEEE-754 double precision, rounding only at the display layer. The engine is reconciled against an independent reference implementation before release.
Does it work on mobile?
Yes. The interface is mobile-first with numeric keypad hints and is tested down to a 320-pixel viewport with no horizontal scrolling.