Guide
One-rep max formulas, explained.
What a 1RM estimate actually is.
A one-rep max (1RM) estimate answers this: "I just lifted w pounds for r reps — what's the heaviest single I could plausibly lift?" The three formulas below all model the same observed fact: as reps go up, the fraction of your max you can lift goes down, roughly linearly, in the low-rep range. They disagree slightly on the slope — which is why you get three different answers and why averaging them is the sensible move.
Epley: w × (1 + r/30).
The most popular formula in strength programming, published by Boyd Epley in 1985. It assumes each additional rep is worth 1/30 of the weight. Simple, and it runs slightly high — lifters often find their tested max lands a few pounds below the Epley number.
Brzycki: w × 36/(37 − r).
Matt Brzycki's 1993 formula divides by (37 − reps), which makes it slightly more conservative than Epley. It's the default in several research studies precisely because it errs on the low side — a safer bet when the estimate drives your training percentages.
Lander: w × 100/(101.3 − 2.67123r).
Jeff Lander's formula fits a curve to measured rep-percentage data, which is why the constants look odd. In practice it lands between Epley and Brzycki — a touch more conservative than Epley without Brzycki's pessimism.
Worked example: 185 lb × 5 reps.
Epley
185 × (1 + 5/30) = 185 × 1.16667 = ≈ 216 lb.
Brzycki
185 × 36/(37 − 5) = 185 × 36/32 = 185 × 1.125 = ≈ 208 lb.
Lander
185 × 100/(101.3 − 2.67123 × 5) = 185 × 100/87.94 = ≈ 210 lb.
Three formulas, three answers: 216 / 208 / 210. The average — ≈ 211 lb — is the number to program from. That's the whole method: run a recent hard set through all three, average them, and use that as your estimated max for percentage-based programming.
Where the formulas stop being trustworthy.
Above ~10 reps, all bets are off. A 15-rep set measures muscular endurance, not maximal strength, and the rep-to-percentage curve bends away from the straight line every formula assumes. Plug 225 × 15 into Epley and you'll get a fantasy number.
Estimates are exercise-specific. They were validated on the squat, bench press, and deadlift. Smaller lifts (overhead press) are more technique-sensitive, so estimates run noisier there.
Fatigue, tempo, and technique shift results. A grindy 5-rep set after a hard workout predicts less than a fresh one. Paused reps, slow eccentrics, and sloppy form all change what the numbers mean.
An estimate is never a guaranteed safe max. The formulas say what you could plausibly lift, not what you should attempt. Don't test a true 1RM without spotters or safety arms, and treat any estimate as a programming input — not a dare.
Common questions.
Which 1RM formula is the most accurate?
Epley is the most widely used and generally lands within 5% of a tested max on the big barbell lifts. Brzycki tends to read slightly low, Lander sits in between. In practice, averaging all three is more useful than arguing over which single formula is best.
Should I test my true 1RM instead of estimating it?
A true max test is more accurate, but it demands spotters or safety arms, a full warm-up ramp, and a body that's recovered and injury-free. For programming percentages, a formula estimate from a recent 3–5 rep set is usually close enough — reserve true max testing for a few times a year at most.
Why do 1RM formulas break down above 10 reps?
Because a 12-rep set tests muscular endurance more than maximal strength. The relationship between reps and percentage of max is roughly linear in the 1–10 range, then curves away — plug 225 × 15 into Epley and you'll get a fantasy number.
Do 1RM estimates work for every lift?
They're most reliable for the squat, bench press, and deadlift. Smaller lifts like the overhead press are more technique-sensitive, so estimates run noisier — and the formulas were never designed for bodyweight or machine movements.