About one rep max estimation
A one rep max (1RM) is the heaviest load a lifter can move through one complete repetition of a given exercise with correct form. It is the anchor that strength coaches use to write training programs, score powerlifting meets, and compare lifters across bodyweights via Wilks or DOTS coefficients. The challenge is that testing 1RM directly is risky and exhausting, so coaches estimate it from a submaximal set using a regression formula.
This calculator runs five of the published formulas in parallel (Epley 1985, Brzycki 1993, Lander 1985, Lombardi 1989, O'Conner 1989) and averages them. Averaging cancels out individual formula bias: Brzycki under-predicts above 10 reps, Lombardi over-predicts at very low reps, Epley sits roughly in the middle. The combined estimate is generally within 2 to 5 percent of a measured 1RM for sets of 3 to 6 reps.
How the formulas work
Each formula maps the pair (weight lifted, reps completed) to a predicted single-rep maximum. They share the assumption that strength endurance follows a predictable curve below about 10 reps: every additional rep at the same load implies that load is a smaller fraction of true 1RM.
Epley 1RM = w x (1 + r / 30) Brzycki 1RM = w x 36 / (37 - r) Lander 1RM = (100 x w) / (101.3 - 2.67123 x r) Lombardi 1RM = w x r^0.10 O'Conner 1RM = w x (1 + 0.025 x r) w = weight lifted r = clean reps to technical failure
- Reps must be true reps to failure, not reps in reserve. Leaving 1 rep in the tank under-predicts 1RM by 3 to 5 percent.
- Form must hold. A grindy, partial-range squat is not a rep for prediction purposes.
- Best window: 3 to 6 reps. Below 3 you may as well attempt the max. Above 8 to 10 fatigue physiology dominates and accuracy collapses.
Worked example: bench press
A lifter benches 80 kg for 5 clean reps to failure. Plug those values into each formula:
- Epley: 80 x (1 + 5/30) = 80 x 1.1667 = 93.3 kg.
- Brzycki: 80 x 36 / (37 - 5) = 80 x 1.125 = 90.0 kg.
- Lander: 100 x 80 / (101.3 - 13.356) = 8000 / 87.94 = 91.0 kg.
- Lombardi: 80 x 5^0.10 = 80 x 1.1746 = 94.0 kg.
- O'Conner: 80 x (1 + 0.125) = 90.0 kg.
- Average: (93.3 + 90.0 + 91.0 + 94.0 + 90.0) / 5 = 91.7 kg.
Percent of 1RM training table
Once you have a reliable 1RM estimate, percent-based programming becomes straightforward. The table below maps load to expected rep range and primary training adaptation.
| % of 1RM | Approx reps | Primary adaptation | Typical use |
|---|---|---|---|
| 100% | 1 | Maximal strength | Powerlifting meet |
| 95% | 2 | Strength | Peaking week |
| 90% | 3 to 4 | Strength | Strength block top set |
| 85% | 5 to 6 | Strength + size | 5x5, Texas Method |
| 80% | 7 to 8 | Hypertrophy | Bodybuilding compound |
| 75% | 9 to 10 | Hypertrophy | Accessory volume |
| 70% | 11 to 12 | Hypertrophy + endurance | High-rep growth |
| 60% | 15 to 20 | Muscular endurance | Conditioning sets |
Common pitfalls
- Counting reps in reserve. A 5RM means 5 reps and the 6th would fail. If you stop one rep short the formula reads as if your true 5RM is heavier, inflating the estimate.
- Using reps above 10. Past 10 the limit is cardiovascular fatigue, not strength. Sets of 12 to 15 produce wildly different 1RM predictions across formulas.
- Mixing assistance lifts with main lifts. A 1RM estimated from a barbell row or incline press does not transfer to a flat bench. Estimate each lift separately.
- Stale estimates. 1RM drifts up with training (2 to 5 percent per month for novices) and down with deloads, illness, or a poor sleep week. Re-estimate every 4 to 8 weeks.
- Equipment differences. A 1RM in a power rack with a competition bar is not the same as a 1RM on a Smith machine, in a sleeve-stiffness bar, or with belt and wraps. Note your setup.
- Single-rep ego attempts. Maxing out every week interferes with hypertrophy adaptation, raises injury rates, and beats up the central nervous system. Reserve true 1RM tests for meets or peaking weeks.
Related calculators
Frequently asked questions
Which 1RM formula is most accurate?
For 2 to 5 reps Epley and Brzycki are within roughly 2 percent of each other and both track measured 1RM well in trained lifters. Above 10 reps every formula loses accuracy because aerobic fatigue contaminates the strength signal. The 3Tej tool averages five formulas (Epley, Brzycki, Lander, Lombardi, O'Conner) to smooth the noise.
How many reps should I use to estimate 1RM?
3 to 5 reps is the sweet spot. Below 3 reps you are already near the true max and gain little from estimation. Above 6 reps the relationship between reps and load becomes nonlinear and formulas overestimate by 5 to 15 percent in untrained lifters.
Should I actually test my 1RM in the gym?
For most lifters no. A direct 1RM test demands a deload week, a spotter, full warm-up ramping, and adds injury risk on the failed attempt. Estimating from a 3 to 5 rep submaximal set gives 95 percent of the information for 5 percent of the risk. Powerlifters test 1RM only in competition or in the final week of a peaking block.
How do I use percent-of-1RM training tables?
Pick a target intensity (e.g. 80 percent for hypertrophy, 90 percent for strength) and multiply your estimated 1RM by that percent to get the working weight. The reps column tells you a realistic max set length at that load. Trained lifters typically work at 70 to 85 percent for volume and 87 to 95 percent for peaking.
Does the same 1RM formula work for bench, squat, and deadlift?
Not exactly. Deadlift 1RM is usually 3 to 5 percent lower than the formula predicts at high reps because grip and lower-back fatigue cap reps before muscular strength does. Bench press matches the formulas well. Squat sits between the two. Treat the estimate as a ceiling, not a guarantee.
Sources
- Epley, B. (1985). Poundage Chart. Boyd Epley Workout. Body Enterprises.
- Brzycki, M. (1993). Strength testing: predicting a one-rep max from reps-to-fatigue. JOPERD, 64(1), 88 to 90.
- Lander, J. (1985). Maximum based on reps. NSCA Journal, 6, 60 to 61.
- Lombardi, V. P. (1989). Beginning Weight Training. Wm. C. Brown.
- LeSuer, D. A. et al. (1997). The accuracy of prediction equations for estimating 1-RM performance in the bench press, squat, and deadlift. JSCR, 11(4), 211 to 213.