How do you calculate your one-rep max (1RM)?

Why estimate instead of test a true max

A true one-rep max is the heaviest weight you can lift for exactly one rep with correct form. Finding it by trial and error means working up through near-maximal attempts until you fail, which is where most injuries around heavy lifts happen, a missed bench rep without a spotter or safety bars, a rounded back on a max deadlift attempt, a knee that buckles under a max squat.

Estimating from a submaximal set avoids that. Lift a weight you can move for a genuine 3 to 8 reps with clean form, feed the weight and rep count into a formula, and you get a number close to your true max without ever approaching failure. For almost every training purpose, programming percentages, tracking progress, deciding how much to load the bar, the estimate is all you need.

The two formulas: Epley and Brzycki

Epley: 1RM = w x (1 + r / 30). Brzycki: 1RM = w x 36 / (37 - r). Both take the weight you lifted (w) and the reps you completed (r) and project what you could have done for one rep.

They're built differently and don't always agree. Epley assumes a fairly steady drop-off in reps as weight increases; Brzycki comes from a lookup table originally built for sets of up to about 10 reps. At low rep counts (2 to 5) the two formulas usually land within a few pounds of each other. The two formulas give the same answer at exactly 10 reps. Below 10 reps Epley runs slightly higher, and above 10 reps Brzycki pulls ahead and the gap widens quickly. Many lifters simply average the two for a middle-ground number.

Worked example: 185 lb bench press for 5 reps

Say you bench 185 lb for 5 clean reps.

Epley: 185 x (1 + 5/30) = 185 x 1.1667 = 215.8, which rounds to about 216 lb.

Brzycki: 185 x 36 / (37 - 5) = 185 x 36/32 = 185 x 1.125 = 208.1, or about 208 lb.

Average the two: (216 + 208) / 2 = 212 lb. That 212 lb figure is your working estimate of a true one-rep max, a number you never actually had to attempt.

Turning your estimate into a training plan

The whole reason to know your 1RM is to program percentages of it for specific goals. Using the 212 lb estimate from above:

• Strength (3 to 5 reps): about 85 to 90% of 1RM = 180 to 191 lb
• Hypertrophy (6 to 10 reps): about 75 to 85% of 1RM = 159 to 180 lb
• Endurance (12 to 15 reps): about 60 to 70% of 1RM = 127 to 148 lb

How accurate are these formulas, and where they break down

Both formulas are most accurate in the range they were built for, roughly 2 to 10 reps. Below that, at 1 to 3 reps, the estimate sits close to a real max because you're barely extrapolating. Above about 10 reps, accuracy drops off, because the relationship between reps and percentage of max stops being linear, fatigue and technique breakdown start to matter more than raw strength, and neither formula accounts for that.

There's also person-to-person variation. Lifters with a higher proportion of fast-twitch muscle fiber tend to fatigue faster at high reps, so a formula run off a high-rep set underestimates their true max; slow-twitch-dominant lifters hold up longer at higher reps, so it can overestimate theirs. The lower your rep count when you test, the less this matters, which is another reason to stay in the 3 to 8 rep range.

Common mistakes

A few habits either inflate the estimate or put lifters at unnecessary risk. Round both back down to something safer and more accurate before you program off the number.

• Testing a true 1RM without a spotter or safety bars. A failed heavy attempt on a bench, squat, or overhead lift is where serious injuries happen; a submaximal-set estimate gets you a usable number without that risk.
• Plugging in a set of more than about 10 reps. Both formulas lose accuracy at high rep counts, so a 15-rep set will overstate or understate your real max more than a 5-rep set will.
• Treating one estimate as permanent. Strength changes over weeks of training, so a number from two months ago won't reflect where you are now. Recalculate periodically rather than programming off an old figure.
• Counting a rep that wasn't clean. A partial-range press or a bounced rep inflates the rep count without representing real strength, which pushes the formula's result higher than your actual capability.

How the estimate climbs as reps go up

The same bar weight produces very different estimates depending on how many reps you got, and seeing the pattern helps you judge which numbers to trust. Take a set at 135 pounds and run both formulas. For 3 reps, Epley gives about 148.5 pounds and Brzycki about 142.9. For 5 reps the pair reads 157.5 and 151.9. At 8 reps they read 171.0 and 167.6. At 10 reps both land on exactly 180.0. At 12 reps the order flips, with Epley at 189.0 and Brzycki at 194.4.

That crossover at 10 reps is worth remembering. Between 2 and 10 reps, Epley runs higher and Brzycki is the more conservative of the two. Past 10, Brzycki pulls ahead and accelerates, because its denominator shrinks toward zero as reps approach 37, which makes it nonsensical for very long sets. Epley also has a quirk at the other end: fed a true single, it returns 103 percent of the weight, so for a genuine 1-rep lift you simply use the weight itself.

The implied percentages tell the same story. Epley assumes a 5-rep set is about 86 percent of your max and a 10-rep set is 75 percent. Brzycki puts 5 reps at about 89 percent and 10 reps also at 75 percent. Those two formulas agree most near 10 reps and disagree most at the extremes, so keep your test set between 3 and 8 reps to minimize the spread.

Counting reps in reserve instead of only reps completed

Both formulas assume the set ended at or near failure, meaning you could not have done another clean rep. Many training sets do not, and plugging in a set that stopped early understates your strength. The fix is to add the reps you believe you had left to the reps you completed.

Using the 185 pound bench example from the main guide: if you stopped at 5 reps but honestly had 2 more in the tank, treat it as 7 reps. Epley then gives about 228 pounds and Brzycki about 222, averaging roughly 225 pounds. The 5-reps-to-failure version averaged 212. That 13 pound gap shows how much a single judgment call moves the output.

The weak spot is the judgment itself. Many lifters misjudge how many reps they had left, especially on a hard final rep that looked slow. If you cannot say confidently how many were left, use the reps you completed, accept a slightly low estimate, and let later sessions raise it. A conservative number you repeat is more useful than an optimistic one you miss.

A metric example: a 140 kg squat for 4 reps

The formulas do not care about units, as long as you keep them consistent, so a lifter working in kilograms uses the same math. Suppose a squat of 140 kg for 4 clean reps. Epley: 140 x (1 + 4/30) = 140 x 1.1333 = about 158.7 kg. Brzycki: 140 x 36 / (37 - 4) = 140 x 36/33 = about 152.7 kg. The average is about 155.7 kg.

A barbell loaded in kilograms usually moves in 2.5 kg jumps with the smallest plates, so a practical working estimate rounds to 155 kg. Do not chase a decimal place the method cannot support. Day-to-day readiness, sleep, and how deep you squatted each rep can shift a true max by several percent, so a figure to the nearest 2.5 kg is already more precise than the underlying measurement.

Working backward: what weight for a target rep count

Once you have an estimate, you can flip the formulas to choose a load for a given number of reps. Epley rearranged gives weight = 1RM / (1 + r/30). Brzycki rearranged gives weight = 1RM x (37 - r) / 36. Say your estimated max is 300 pounds and you want a hard set of 8.

Epley: 300 / (1 + 8/30) = 300 / 1.2667 = about 236.8 pounds. Brzycki: 300 x 29/36 = about 241.7 pounds. The two sit about 5 pounds apart, so loading 235 to 240 pounds is a sensible target, and your first set tells you which side of the range suits you. Round to the plate jumps you actually have rather than the decimal the math produced.

This is the practical payoff of knowing your max: you can plan a rep target and load without guessing. If the planned load turns out too easy or too hard, treat that as new data and update your estimate rather than forcing the plan.

Training max and a simple three-week wave

Some coaches and popular programs do not train directly off the estimated max. They set a training max at around 90 percent of it, which builds in a margin for bad days and keeps rep quality high. Using the 212 pound estimate, a training max would be about 191 pounds.

From there, a simple wave might use percentages of the training max across three weeks. At 65 percent the working weight is about 124 pounds, at 75 percent about 143, and at 85 percent about 162. In practice a lifter rounds to available plates, perhaps 125, 145, and 160 pounds.

The point is not that these exact percentages are special. It is that the training max is a deliberate buffer, and the lighter weeks leave room for the heavier ones. When your top sets feel easy for a couple of cycles, recalculate from a fresh set and nudge the training max up.

If you do want to test a true max safely

Nothing here says you must never find a real one-rep max. If you are preparing for a meet or simply want to see it, the conditions matter far more than the number. Test on a lift where failure is safe, such as a bench press with a competent spotter or safety arms set correctly, or a squat inside a rack with safeties set below your lowest depth. Choose a day when you are rested and warmed up.

A common pattern ramps from light weights with falling reps. Using the 212 pound estimate, it might look like 45 pounds for 8, 95 for 5, 135 for 3, 165 for 2, 185 for 1, 200 for 1, then 210 for 1, with 3 to 5 minutes of rest between the heavy singles. Then make a single attempt around the estimate. Stop when bar speed drops sharply or form breaks, instead of forcing a grinder.

If a true attempt lands well above or below the estimate, that is information about how your body responds to rep ranges. Use it to decide which formula, or which average, fits you best, and apply that adjustment to future estimates.