Are one-rep max calculators accurate?

A one-rep max calculator gets you close, not exact. The two best-known formulas agree to within about 4 kg per 100 kg on the bar for sets of 2 to 12 reps. Studies find the estimate is better the fewer reps you use, with ten as the ceiling. Treat the result as an estimate to track over time, not a lift you are owed. The One-Rep Max Calculator shows six formulas side by side.

What a one-rep max calculator does

It takes two numbers — the weight you lifted and the reps you got — and scales up to what one rep might be. Epley's formula multiplies: estimated max = weight × (1 + reps ÷ 30). Brzycki's divides: estimated max = weight ÷ (1.0278 − 0.0278 × reps), which comes to the same thing, give or take rounding, as weight × 36 ÷ (37 − reps).

So a set of 5 at 100 kg becomes 116.7 kg under Epley and 112.5 kg under Brzycki. Same set, two answers.

Neither equation began as a large laboratory study. According to a 2026 preprint that reviewed these formulas (not yet peer reviewed), Epley's equation came from a poundage chart in a 1985 training manual for athletes at the University of Nebraska, and Brzycki published his in 1993 in an article for practitioners. The same paper notes that the classic equations were derived from small samples doing a single exercise, usually the bench press. That is worth knowing before you treat either figure as a measurement. You can read the 2026 preprint on the formulas yourself.

Epley and Brzycki, side by side

Run both formulas on one 100 kg set and the quarrel settles itself. They give the same answer at 10 reps. Below 10 reps Epley reads higher, by at most 4.2 kg per 100 kg on the bar. Above 10 reps Brzycki reads higher, and the gap widens fast. The table uses a 100 kg set; the gaps scale in proportion for your own lifts.

The two cross at 10 reps, where both give 133.3 kg. From 2 to 9 reps Epley sits above Brzycki, widest at 4 and 5 reps with 4.2 kg per 100 kg. At 11 and 12 reps the order flips: Brzycki is higher by 1.8 kg and then 4.0 kg. Past 12 reps the gap grows quickly. At 15 reps Epley gives 150.0 kg and Brzycki 163.6 kg — a difference of 13.6 kg on the same 100 kg set.

What the studies found

Three findings matter. Fewer reps make for a better estimate. Error grows as the reps climb. And people differ in how many reps they get at the same percentage of their max, so one equation cannot fit everybody, or every lift, equally well.

Reynolds, Gordon and Robergs (2006, Journal of Strength and Conditioning Research) had 70 adults — 34 men and 36 women, aged 18 to 69 — test their 1, 5, 10 and 20-rep maxes on the barbell bench press and a plate-loaded leg press. The load fell in a curve, not a straight line, as the reps rose. Predictions built from the 5-rep max came out most accurate, and the authors concluded that no more than 10 repetitions should be used in linear equations to estimate a one-rep max. Their 2006 bench and leg press study is the clearest statement of the ceiling.

Mayhew and colleagues (2008, same journal) tested the bench press of 103 young women before and after 12 weeks of training. The 2008 training study found prediction equations were more accurate when fewer than 10 repetitions to fatigue were used.

The same group (2002, Journal of Sports Medicine and Physical Fitness) tested 260 college American-football players who pressed 225 lb (102.3 kg) for as many reps as they could. Of nine prediction equations then in use, only two gave results that were not significantly different from the players' real maxes. Even the two equations built for that 225 lb test each put only about half the players, 52% and 51%, within 4.5 kg of their real max. The authors write that the accuracy of prediction decreases at higher repetitions. That is the 2002 study of 260 players.

Then there is the spread between people. Nuzzo and colleagues (2024, Sports Medicine) pooled 952 repetitions-to-failure tests from 269 studies and 7,289 people. The number of reps a person could do at a given percentage of their max varied from one individual to the next, and was higher on the leg press than on the bench press across the whole range of loads. Sex, age and training status made little difference. Read the 2024 pooled analysis for the detail. A formula built on bench press data is being asked to speak for your leg press too.

Even a tested max moves

Testing a real one-rep max does not hand you a fixed number either. Grgic and colleagues (2020, Sports Medicine – Open) reviewed 32 studies covering 1,595 people to see how repeatable the test is. It was reliable: the median correlation between test and retest was 0.97. But the typical variation from one test to the next was 4.2%, with studies ranging from 0.5% to 12.1%.

On a 100 kg max, 4.2% is 4.2 kg. So the same lifter, tested twice, can reasonably come out a few kilos apart. Their 2020 reliability review sets the margin you are comparing any estimate against. A number with a margin is still useful, but it is not a line in the record book.

Three rules for using the number

Keep the reps low, keep the formula fixed, and read the trend rather than the single figure. None of them require you to test anything. They only require you to be consistent about what you type in and what you compare it to.

1. Use a hard set of ten reps or fewer, and nearer five when you can. Reynolds and colleagues put the ceiling at 10 reps for linear equations; Mayhew's 2008 work found accuracy better under 10 reps; the 2002 study noted accuracy falls as reps rise. A set of 5 is better input than a set of 12.
2. Use the same formula every time. Switching between Epley and Brzycki moves the answer by up to 4.2 kg per 100 kg with no change in you. If your logged max jumps because you changed websites, you have learned nothing about your strength.
3. Watch the trend, not the single figure. The same formula on the same lift, week after week, tells you whether progressive overload is working. A rising line over two months means more than any one reading. If you are adding reps before weight, as in double progression, each added rep shows up as a higher estimate.

Masheen is a workout log for people who lift. It works out an estimated one-rep max after each set using Epley's formula and draws the trend for each lift; strength trends are part of Masheen Pro. Its own one-rep max calculator shows Brzycki's figure beside Epley's as a second opinion. The app is coming soon to iPhone and Android. The One-Rep Max Calculator on this site does the same sum: Epley as the headline, with Brzycki and four other published formulas beside it, and the spread between them.