One-Rep Max Calculator

This calculator compares three validated formulas: Epley (1RM = weight x (1 + reps/30)), Brzycki (1RM = weight x 36 / (37 - reps)), and Lombardi (1RM = weight x reps^0.10). The Epley formula, published by Boyd Epley at the University of Nebraska in 1985, is the most widely cited in strength training literature. Brzycki tends to produce slightly lower estimates at higher rep counts (8-12 reps), making it more conservative. Lombardi uses an exponential model rather than linear. All three agree within 2-3% for sets of 3-6 reps but diverge at 10+ reps. The calculator presents all three so you can see the range and use the most conservative estimate for safety.

Research by Reynolds, Gordon, and Robergs (2006) in the Journal of Strength and Conditioning Research found that estimation accuracy is highest for sets of 2-6 reps, where predictions typically fall within 3-5% of a tested 1RM. At 7-10 reps, accuracy drops to within 5-8%. Above 10 reps, estimates can overpredict by 10-15% because muscular endurance becomes a larger factor and the linear relationship between reps and maximal capacity breaks down. For the most reliable estimate, use a recent set of 3-5 reps performed at true maximal effort with good form.

For most training purposes, an estimated 1RM is safer and more practical than a true tested maximum. Actual 1RM attempts carry meaningful injury risk, particularly on compound lifts like squats and deadlifts, and produce significant central nervous system fatigue that can impair training for several days afterward. The NSCA recommends that only experienced lifters with spotters and proper equipment attempt true 1RM testing. For programming purposes, an estimated 1RM derived from a 3-5 rep set provides sufficient accuracy to set training loads.

Different exercises have different rep-to-max relationships based on muscle groups involved and movement patterns. Isolation exercises and machine movements tend to allow more reps at a given percentage of 1RM than compound barbell movements. Deadlifts often produce lower rep-to-max ratios than squats because grip fatigue and spinal loading accumulate faster. The formulas were primarily validated on compound barbell movements, so they are most accurate for squat, bench press, deadlift, and overhead press.

Update your estimated 1RM every 4-6 weeks, typically at the end of a training block before starting a new cycle. Intermediate lifters can expect 3-5% improvements per block on major lifts, while beginners may see 5-10% gains. If your estimated 1RM has not improved across two consecutive training blocks (8-12 weeks), your programming likely needs adjustment in volume, intensity, exercise selection, or recovery management.

A training max is typically set at 85-90% of your estimated true 1RM and is used as the reference point for calculating working weights in many popular programs like 5/3/1 by Jim Wendler. Using a training max instead of a true 1RM provides a built-in buffer that reduces injury risk and allows for productive training even on suboptimal days. If your estimated 1RM is 150 kg, your training max at 90% would be 135 kg, and your working sets would be calculated from that lower number.

No. The Epley, Brzycki, and Lombardi formulas are purely mathematical relationships between weight lifted and reps completed. They do not account for body weight, training age, sex, or exercise type. The accuracy remains consistent across these variables for sets of 2-10 reps. For bodyweight-relative strength comparison, use the result in the Strength Percentile Calculator.

You can, but the estimate will not transfer directly to free-weight equivalents. Smith machine movements remove stabilization demands, typically allowing 10-15% more weight than the free-weight version. If you estimate a Smith machine bench press 1RM of 100 kg, your free-weight bench press 1RM is likely 85-90 kg. Use this calculator with the exercise variation you actually train and program for.

Yes. AI Fit Hub tools are free, no-signup browser tools. All calculations run locally in your browser and no data is sent to any server.