Summary Your one-rep max (1RM) is the heaviest load you can lift for a single rep. This calculator estimates it by averaging three established formulas: Epley 1985, Brzycki 1993, and Lombardi 1989. A lifter who hits 100 kg (220 lb) for 5 reps on the bench press has an estimated 1RM of roughly 116 kg (256 lb), with the three formulas spanning 113 to 117 kg. Accuracy is roughly plus or minus 2 percent at 3 reps and degrades to plus or minus 10 percent at 10 reps for compound lifts (Reynolds et al. 2006, J Strength Cond Res). Use the percentage table to program hypertrophy work at 65 to 75 percent, max strength at 85 to 95 percent, and power at lighter loads moved fast.

Estimate your 1 rep max

All inputs update the result instantly. Default values shown for a 100 kg bench press for 5 reps on a compound lift, taken close to failure.

The load you used on a recent set.
Best accuracy at 3 to 5 reps. Above 10 the estimate degrades.
Compound lifts estimate roughly 50 percent more accurately.
Helms and Zourdos RPE scale. Adjusts up if you stopped short.
Estimated 1 rep max
116kg
Spread across formulas: 113 to 117 kg
Epley
117 kg
Brzycki
113 kg
Lombardi
117 kg
Confidence
± 5%

Average of Epley 1985, Brzycki 1993, and Lombardi 1989. Accuracy roughly ±5% at 5 reps for compound lifts (Reynolds et al. 2006, J Strength Cond Res).

Conceptual illustration of a loaded barbell with weighted plates representing one rep max strength estimation across three peer-reviewed formulas
Three formulas, three answers. The calculator runs them in parallel and shows the spread so you see how confident you can actually be.

Most online 1RM calculators run a single formula and hand back a single number. That number gets treated as gospel and plugged into a percentage program, when in reality the underlying equations disagree by several percent and the prediction error widens with every additional rep beyond a true triple.

This calculator runs the three most-validated formulas in parallel, shows you each estimate, and averages them. The spread itself is information: a tight cluster means the estimate is more trustworthy. A wide cluster, especially above 8 reps, is a signal to retest with a heavier load.

How this calculator works

Each formula was developed independently and published in the strength and conditioning literature. None is universally best. LeSuer et al. 1997 in the Journal of Strength and Conditioning Research compared multiple equations against measured 1RMs in the bench press, squat, and deadlift, and found Brzycki most accurate for bench and squat, Epley most accurate for deadlift, and Lombardi performing better at higher reps. Averaging the three reduces the impact of any one formula's blind spot.

For accuracy, Reynolds et al. 2006 in the Journal of Strength and Conditioning Research found mean prediction error of roughly ±2 percent at 3 reps, ±5 percent at 5 reps, and ±10 percent at 10 reps for compound barbell lifts in trained adults. Isolation lifts carry roughly 50 percent wider error bands. The calculator displays a confidence band that shifts to match your inputs.

Why averaging beats any single formula

The problem with a single number

A single estimate hides uncertainty. If you bench 100 kg for 5 reps and Epley returns 117 kg while Brzycki returns 112.5 kg, your true 1RM is somewhere in that 4 to 5 kg window. A program that prescribes "85 percent of 1RM for triples" wants 95.6 kg if you trust Brzycki, or 99.4 kg if you trust Epley. Same set, almost a 4 kg programming difference based on which formula you pick.

Showing the spread directly is what fixes this. If the formulas cluster within 3 percent of each other, your estimate is solid. If they spread more than 7 percent apart, the rep count is probably too high to give a reliable answer, and you should retest with a heavier load taken to a true 3 to 5 reps.

Silhouette of an athlete performing a heavy back squat with a loaded barbell representing a compound lift used for accurate one rep max estimation
Compound barbell lifts are the original validation context for these formulas. Isolation movements predict less reliably and need wider error bars.

Why exercise type matters

Reynolds 2006 and several follow-up studies found that prediction equations work best on multi-joint compound lifts (squat, bench press, deadlift, overhead press, row) where total-body neural drive and force output are the limiting factors. For isolation movements (biceps curl, leg extension, lateral raise), local muscular endurance dominates fatigue, and the relationship between sub-max reps and true 1RM is noisier. The calculator widens the confidence band when you select isolation, which is the honest reflection of the literature.

Worked examples (for quick reference)

Eight common scenarios with the calculator output for each formula and the average. Use this as a sanity check against your own numbers.

Set performed Type Epley Brzycki Lombardi Average
60 kg × 5 reps Compound 70 kg 68 kg 70 kg 69 kg
80 kg × 3 reps Compound 88 kg 85 kg 89 kg 87 kg
100 kg × 5 reps Compound 117 kg 113 kg 117 kg 116 kg
100 kg × 8 reps Compound 127 kg 124 kg 123 kg 125 kg
140 kg × 3 reps Compound 154 kg 148 kg 156 kg 153 kg
180 kg × 2 reps Compound 192 kg 185 kg 193 kg 190 kg
50 kg × 12 reps Compound 70 kg 72 kg 64 kg 69 kg
20 kg × 10 reps Isolation (curl) 27 kg 27 kg 25 kg 26 kg

Notice how the formulas agree closely at low reps and start to diverge above 8. That divergence is the calculator telling you the estimate has more room for error, not that the math is broken.

Knowing your max is half the equation. Programming it intelligently is the other half.

FitCraft's AI coach Ty builds a personalized strength program around your numbers, your schedule, and your training history. Free version available.

Take the Free Assessment Free • 2 minutes • No credit card

Programming with your 1RM: the percentage table

The point of estimating your 1RM is to anchor a percentage-based training plan. The 2017 meta-analysis by Schoenfeld et al. in the Journal of Strength and Conditioning Research showed that hypertrophy gains are similar across a wide load range when sets are taken close to failure, but the practical sweet spot for muscle growth still lives at moderate intensities. The classic load-rep zones from the NSCA Essentials of Strength Training and Conditioning (Haff and Triplett, 4th ed.) map percentages to reps and to training purpose as follows.

% of 1RM Reps possible Primary purpose Weight at default 116 kg 1RM
100% 1 Max strength test 116 kg
95% 2 Max strength 110 kg
90% 3 to 4 Max strength 104 kg
85% 5 to 6 Strength 99 kg
80% 7 to 8 Strength & hypertrophy 93 kg
75% 9 to 10 Hypertrophy 87 kg
70% 11 to 12 Hypertrophy 81 kg
65% 13 to 15 Hypertrophy & endurance 75 kg
60% 16 to 20 Endurance 70 kg

For most lifters, most of the time, the productive zone for general strength and physique work sits between 65 and 90 percent of 1RM. Anything below 60 percent rapidly tilts into local muscular endurance work, and anything above 95 percent should be reserved for peaking blocks or actual testing days, not regular training. We covered the broader case for staying inside this productive band in our piece on progressive overload for beginners, and the load-versus-volume tradeoff for hypertrophy is laid out further in how rest periods affect muscle growth.

RPE and reps in reserve

The formulas above all assume the rep count you entered was a true rep-max, meaning the next rep would have failed. Most well-programmed sets stop short of failure to manage fatigue. The Helms and Zourdos RPE scale (described in Helms et al. 2016 in Strength and Conditioning Journal, validated by Zourdos et al. 2016 in the Journal of Strength and Conditioning Research) maps perceived effort to reps in reserve as follows.

If you select an RPE under 10, the calculator adds the implied reps in reserve to your rep count before running the formulas, then applies the same averaging. So 100 kg × 5 reps at RPE 8 is treated as a 100 kg × 7 rep-max equivalent, which gives a higher 1RM estimate than the same set treated as a true 5RM. This is the same logic strength coaches use when programming with auto-regulation.

Three myths the calculator deliberately ignores

Myth 1: a single formula gives a precise number

No formula was ever intended to. Brzycki's original 1993 paper called the equation a "useful approximation" for testing, not a substitute. Reynolds 2006 confirmed that even the best equation carries roughly ±2 percent error at low reps and degrades sharply above 10. A 1RM estimate is a working number for programming, not a competition entry.

Myth 2: high-rep sets predict 1RM as well as low-rep sets

They do not. The biology is different: a 3-rep set is fundamentally a maximum-strength expression. A 15-rep set is largely a local muscular endurance expression with strength as one input. The further the rep count drifts from a true triple, the more your estimate is reflecting your endurance rather than your peak force production. If you want a clean 1RM estimate, build up to a heavy 3 to 5 and use that set, not a high-rep down-set.

Myth 3: testing your true 1RM regularly is optimal

For most lifters, true 1RM testing is high-fatigue, technique-dependent, and offers limited new information beyond what a heavy 3 to 5 already tells you. Programs that auto-regulate by RPE or use estimated 1RMs almost always outperform programs anchored to monthly true-max tests, because the test itself eats into recovery and demands near-perfect technique under maximal load. Save true testing for the end of dedicated peaking blocks, and use estimates for everything in between. Older lifters in particular benefit from staying off true-max attempts, as we covered in strength training after 60.

Abstract conceptual illustration of training intensity progression showing layered platforms ascending with weighted plates representing percentage-based programming for hypertrophy strength and power
Percentage-based programming uses your estimated 1RM as the anchor. The same lifter trains at different fractions of that anchor depending on the goal: hypertrophy, strength, or power.

When to ignore this calculator

The calculator is built for healthy adults with at least a few months of consistent training on the lift in question. A few situations require professional judgment instead.

For everyone else, the calculator is a defensible starting point. Even when an actual max test is on the calendar, plug your most recent heavy set in first to set realistic warm-up jumps. And whenever the actual programming calls for a true 1RM (a meet, a benchmark test, a peaking finisher), bring in a qualified coach for the attempt itself.

Related reading

References

  1. Epley B. "Poundage Chart." In: Boyd Epley Workout. Lincoln, NE: Body Enterprises; 1985:86. Original publication of the Epley equation.
  2. Brzycki M. "Strength testing: predicting a one-rep max from reps to fatigue." JOPERD. 1993;64(1):88-90. doi:10.1080/07303084.1993.10606684
  3. Lombardi VP. Beginning Weight Training: The Safe and Effective Way. Dubuque, IA: William C Brown; 1989. Original publication of the Lombardi equation.
  4. Reynolds JM, Gordon TJ, Robergs RA. "Prediction of one repetition maximum strength from multiple repetition maximum testing and anthropometry." J Strength Cond Res. 2006;20(3):584-592. doi:10.1519/R-15694.1
  5. LeSuer DA, McCormick JH, Mayhew JL, Wasserstein RL, Arnold MD. "The accuracy of prediction equations for estimating 1-RM performance in the bench press, squat, and deadlift." J Strength Cond Res. 1997;11(4):211-213. doi:10.1519/00124278-199711000-00001
  6. Schoenfeld BJ, Grgic J, Ogborn D, Krieger JW. "Strength and hypertrophy adaptations between low- vs. high-load resistance training: a systematic review and meta-analysis." J Strength Cond Res. 2017;31(12):3508-3523. doi:10.1519/JSC.0000000000002200
  7. Haff GG, Triplett NT (eds). Essentials of Strength Training and Conditioning. 4th ed. Champaign, IL: Human Kinetics; 2016. National Strength and Conditioning Association.
  8. Helms ER, Cronin J, Storey A, Zourdos MC. "Application of the Repetitions in Reserve-Based Rating of Perceived Exertion Scale for Resistance Training." Strength Cond J. 2016;38(4):42-49. doi:10.1519/SSC.0000000000000218
  9. Zourdos MC, Klemp A, Dolan C, et al. "Novel Resistance Training-Specific Rating of Perceived Exertion Scale Measuring Repetitions in Reserve." J Strength Cond Res. 2016;30(1):267-275. doi:10.1519/JSC.0000000000001049

Frequently Asked Questions

What is a 1 rep max and why does it matter?

Your one-rep max (1RM) is the heaviest load you can lift for a single repetition with technically sound form. It matters because nearly every evidence-based strength program prescribes daily loads as a percentage of your 1RM. Hypertrophy work typically lives at sixty-five to seventy-five percent of 1RM for eight to twelve reps. Maximum strength work runs at eighty-five to ninety-five percent for one to five reps. Power work uses lighter loads moved fast. Without a 1RM number, percentage-based programming has nothing to anchor to.

Why average three formulas instead of just using Epley?

Each formula has known biases. LeSuer et al. 1997 in the Journal of Strength and Conditioning Research compared multiple equations and found Brzycki most accurate for the bench press and squat, Epley most accurate for the deadlift, and Lombardi performing better at higher rep counts. Averaging the three reduces the impact of any one formula's blind spot. Showing each output side by side also reveals the spread, which is itself information: a tight cluster means the estimate is more trustworthy than a wide one.

How accurate is a 1RM calculator?

Reynolds et al. 2006 in the Journal of Strength and Conditioning Research found accuracy of roughly plus or minus two percent at three reps, plus or minus five percent at five reps, and plus or minus ten percent at ten reps for compound barbell lifts in trained adults. Above ten reps, local muscular endurance becomes the limiting factor rather than maximum strength, so estimates degrade further. Isolation lifts carry roughly fifty percent wider error bands than compounds. Use the calculator for programming reference, not for proof of a true max.

What is RPE and how does it improve the estimate?

RPE (Rate of Perceived Exertion) on the Helms and Zourdos resistance-training scale runs from six to ten. RPE 10 means you could not have done another rep. RPE 9 means you had one rep in the tank. RPE 8 means two reps. RPE 7 means three reps. If you stop a set short of failure, the formulas underestimate your true 1RM because they assume the rep count was a true rep-max. The calculator adds the implied reps in reserve to your rep count before running the formulas, which corrects for stopping early.

Should I just go test my actual 1 rep max?

True 1RM testing is appropriate for experienced lifters with sound technique on the lift in question, ideally under the supervision of a qualified coach. For most lifters most of the time, a calculated estimate from a clean set of three to five reps is safer, less fatiguing, and accurate enough for programming. Novices, anyone with active injuries, and anyone who has not trained the lift to technical proficiency should not test a true max and should use a coach-supervised assessment instead.

What percentage of my 1 rep max should I train at?

It depends on the goal. The 2017 meta-analysis by Schoenfeld and colleagues in the Journal of Strength and Conditioning Research showed that hypertrophy gains are similar across a wide load range when sets are taken close to failure, but the practical sweet spot for hypertrophy is sixty-five to seventy-five percent of 1RM for eight to twelve reps. Maximum strength training runs at eighty-five to ninety-five percent for one to five reps per the NSCA Essentials of Strength Training and Conditioning. Power work uses lighter compound loads moved with maximal intent, often forty to sixty percent of 1RM for explosive triples.

Does this calculator work for the deadlift, squat, and bench press?

Yes. The three big barbell lifts are the original validation context for these formulas. LeSuer et al. 1997 specifically validated Brzycki, Epley, and Lombardi against measured 1RMs on the bench press, squat, and deadlift in trained college athletes and found mean errors under five percent at three to five reps. The calculator uses the same formulas, weights kilograms or pounds identically, and applies the same compound versus isolation accuracy disclaimer.