One Rep Max Calculator
Estimate your one-rep max from the weight and reps of a set, using several established formulas, with the rep range where the estimate is actually reliable.
estimated one-rep max
| % of max | Weight | ~Reps |
|---|---|---|
| 100% | — | 1 |
| 95% | — | 2 |
| 90% | — | 4 |
| 85% | — | 6 |
| 80% | — | 8 |
| 75% | — | 10 |
| 70% | — | 12 |
| 65% | — | 16 |
| 60% | — | 20 |
At one rep they should all agree, and they differ by nine percent
If you lift 100 kg once, your one-rep max is 100 kg. That is not an estimate, it is the definition — and it is the one input where every formula can be checked without any theory at all. Most of them fail it:
| Formula | Says your 1RM is | Off by |
|---|---|---|
| Epley | 103.3 kg | +3.3 kg |
| Brzycki | 100.0 kg | exact |
| Lombardi | 100.0 kg | exact |
| O'Conner | 102.5 kg | +2.5 kg |
| Wathan | 101.3 kg | +1.3 kg |
| Mayhew | 108.9 kg | +8.9 kg |
Only Brzycki and Lombardi return the weight you actually lifted. The others are up to 8.9% apart on the easiest question there is.
That is not quite the scandal it looks like — these are regression fits to population data rather than derivations, so nothing forces them through the obvious point. But it is worth knowing that a formula which cannot reproduce a single rep is being trusted with twelve.
And the disagreement explodes as the reps go up
Lifting 100 kg for a given number of reps, here is what the six formulas say your maximum is:
| Reps | Epley | Brzycki | Lombardi | O'Conner | Wathan | Mayhew | Spread |
|---|---|---|---|---|---|---|---|
| 1 | 103 | 100 | 100 | 102 | 101 | 109 | 9% |
| 3 | 110 | 106 | 112 | 108 | 109 | 114 | 8% |
| 5 | 117 | 113 | 117 | 113 | 117 | 119 | 6% |
| 8 | 127 | 124 | 123 | 120 | 128 | 126 | 6% |
| 10 | 133 | 133 | 126 | 125 | 135 | 131 | 8% |
| 12 | 140 | 144 | 128 | 130 | 141 | 135 | 12% |
| 15 | 150 | 164 | 131 | 138 | 151 | 142 | 25% |
| 20 | 167 | 212 | 135 | 150 | 164 | 151 | 57% |
At five reps they sit within 6% of each other. At twenty they are 57% apart — nearly 77 kg of disagreement about the same set. They stay within 10% of one another up to about 11 reps and diverge sharply after.
Brzycki is the extreme case and fails outright rather than gently. It divides by (37 − reps), so it climbs to infinity at 37 reps and turns negative beyond. The other five stay finite and positive out to fifty, so this is one formula's breakdown rather than a general limit.
The practical reading: estimate from a low rep count. Three to five keeps the formulas within about six percent of one another, which is roughly the day-to-day variation in what anyone can actually lift on a given morning. Past about ten reps the estimate is measuring the formula rather than the lifter, and which one your app happens to use becomes the largest term in the answer.
How to use
- Enter the weight lifted and the reps completed.
- Use a set taken close to failure for a meaningful estimate.
- Read the estimated maximum and the percentage table.
- Treat it as a training guide, not a number to attempt blindly.
Frequently asked questions
How does the estimate work?
From the observed relationship between how much weight you can lift and how many times you can lift it. The common formulas — Epley, Brzycki and others — all express roughly the same curve with slightly different constants, which is why they agree closely at low reps and diverge at high ones.
How many reps give a reliable estimate?
Under about six, and ideally two to five. Beyond ten reps the estimate becomes increasingly unreliable, because performance at high repetitions depends on muscular endurance and conditioning rather than maximal strength — two things that vary hugely between people.
Does the set need to be to failure?
It needs to be close to it for the estimate to mean anything. A set with several reps left in reserve produces a substantial underestimate, since the formula assumes you could not have done another. Being honest about how hard the set was matters more than which formula you use.
Why do the formulas disagree?
Because they were fitted to different groups of lifters. Individual endurance at submaximal loads varies considerably, so someone who can grind out many reps at 85 per cent will be underestimated by formulas calibrated on people who cannot. Your own history is a better guide than any equation.
Does it work for every lift?
Better for some than others. It is most reliable for large compound lifts where technique holds up under load — squat, deadlift, bench press. For isolation work and technically demanding lifts, fatigue and form break down differently, and the estimate is looser.
Should I actually test my maximum?
Only with proper preparation, spotters or safety equipment, and enough training experience to hold form under a maximal load. Estimating is safer and sufficient for programming, which is why most training plans work from percentages of an estimated maximum rather than a tested one. Nothing here is a substitute for coaching.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.