How to Calculate One Rep Max: Why Two Formulas Disagree

2026-10-24

To calculate your one rep max without testing it, do a heavy set to failure of 10 reps or fewer, then put the weight and the reps into a prediction formula. The two most common are Epley, 1RM = weight × (1 + reps ÷ 30), and Brzycki, 1RM = weight × 36 ÷ (37 − reps). A set of 100 lb for 5 reps gives 116.7 lb by Epley and 112.5 lb by Brzycki.

That gap is the part most guides skip. They list the formulas, sometimes five or six of them, and leave you to pick one. This article shows where the two main formulas agree and where they don't, why 10 reps is a hard limit, and what actually makes an estimate wrong, with examples you can check.

How to calculate one rep max with Epley and Brzycki

Both formulas need the same two inputs: the total weight you lifted, bar included, and the number of full reps you completed before you couldn't do another one with good form. Units don't matter. Put in pounds and you get pounds; put in kilograms and you get kilograms.

Epley comes from a poundage chart published at the University of Nebraska in 1985. Its logic is linear: every rep adds a thirtieth of the weight, or about 3.3%. Brzycki was published in 1993 in the Journal of Physical Education, Recreation and Dance. You'll also see it written as weight ÷ (1.0278 − 0.0278 × reps), which is the same formula with the constants rounded.

Here's what the two formulas give for a few sets taken from published examples:

• 100 lb × 5 reps: Epley 116.7 lb, Brzycki 112.5 lb.
• 100 lb × 6 reps: Epley about 120 lb, Brzycki about 116 lb.
• 80 kg × 5 reps: Epley about 93 kg, Brzycki about 90 kg.
• 75 kg × 10 reps: Epley about 100 kg.

The last one is a useful sanity check. A weight you can move exactly 10 times sits at roughly three quarters of your estimated max.

Why the two formulas give different answers

Epley grows in a straight line with reps. Brzycki starts out lower and catches up, and the two meet at exactly 10 reps. There they agree: 100 lb for 10 reps comes out at about 133 lb by either formula. At every rep count below 10, Epley gives the higher number.

That crossing point has a misleading side effect. Because the formulas agree at 10 reps, it's tempting to think a 10-rep set gives the most trustworthy estimate. It doesn't. The curves simply cross there, and the research points the other way: a review of the literature published in 2026 cites work showing that a 5-rep max predicts the true max better than a 10-rep or 20-rep max.

So which formula is right? Research doesn't crown either. A 2013 validation study by Menêses and colleagues found that how well a prediction equation works depends on the exercise. No single formula won across lifts, and on the leg press every equation they tested overestimated the measured max, by 6–15%.

A practical reading is to treat the two numbers as a range rather than two competing answers. If you're using the estimate to plan working weights, the lower Brzycki value is the cautious choice.

The single-rep quirk and other small differences

Plug 1 rep into Epley as written and you get the weight times 1.0333, a max about 3% above what you just lifted. That makes no sense: a single you completed is your max for that day. This is why Wikipedia gives the Epley formula with the condition "assuming r > 1", and why our calculator returns the weight itself for one rep under both formulas. Brzycki doesn't need the fix, because 36 ÷ (37 − 1) is exactly 1.

Some of the disagreement between calculators comes from the Epley constant. Some references write it as 0.0333 per rep, others as 0.033, and the original chart isn't openly available. The difference is real but small: for 200 lb × 10, a 0.033 coefficient gives 266.0 lb and the 1/30 version 266.7 lb. When two calculators disagree by under a pound, this is usually why. It's tiny next to the error that comes from the set itself.

Why one rep max formulas stop at 10 reps

Many online calculators accept 15 or 20 reps. Ours stops at 10, because that's where the sources draw the line. Brian Mackenzie's coaching reference calls the estimates reasonable "providing the number of repetitions does not exceed 10." The 2026 review notes that accuracy falls above roughly 10 reps. Wikipedia says the formulas "greatly diverge after about 10 reps."

The reason is physiological as much as mathematical. A set of 15 or 20 depends heavily on muscular endurance, so two lifters with the same max can get very different rep counts with the same weight. For some lifts the reliable range is narrower still. The US Defense Department's Human Performance Resources site (HPRC) draws on the NSCA's strength training textbook and says multiple-rep tests for exercises like the back squat or power clean aren't very reliable above 5RM, because form breaks down as fatigue builds.

The best set to estimate from is a heavy one, something you can manage for about 3–5 good reps.

What really makes an estimate wrong

Picking a formula changes the answer by a few percent. These things can change it by more:

• Stopping short of failure. The formulas assume you couldn't do another rep. If you had two or three left, the estimate comes out low.
• Counting reps that shouldn't count. A bench rep bounced off the chest or a squat cut short makes the set look better than it was, so the estimate comes out high.
• The exercise. Accuracy depends on the lift. On the leg press, the 2013 study found every equation overestimated the max.
• Treating one estimate as a fact. As Wikipedia puts it, the estimate may vary by 10% or more from the actual 1RM.

None of these show up in the formula, which is why an estimated max works as a planning number, not a personal record.

Turning the estimate into working weights

An estimated max is mostly useful as a base for percentages. HPRC walks through the NSCA chart with a bench press max of 230 lb: 85% of it is about 196 lb, a load that lifter could handle for about 6 reps. The same guidance puts strength work at 4–6 reps with at least 85% of 1RM, and puts a 5-rep max at about 87% of 1RM. Enter 230 lb for 1 rep in the calculator and its 85% row comes within half a pound of that 196 lb.

Be careful going the other way. Charts that map percentages to rep counts are estimates too. If you invert the formulas, Epley and Brzycki predict different rep counts for the same load: about 3.3 reps at 90% of max by Epley and about 4.6 by Brzycki.

An estimate exists so you don't have to test a heavy single, but sets to failure carry risk too. Use a spotter or safety bars. If you have an injury or a health condition, check with a doctor or a qualified coach before training to failure.

What to calculate next

The one rep max calculator below does the arithmetic from this article. Enter the weight and your reps to failure, from 1 to 10, and choose Epley or Brzycki for the main result. The other formula is shown next to it, so you always see the range. Below that is a table of working weights from 100% down to 50% of the estimate. You can enter pounds or kilograms. Switching the unit only changes the label, because the formulas work the same either way. When your training changes, run it again with a fresh heavy set rather than keeping one estimate as your number for good.