One-Rep MaxPublished Last updated 6 sources

    1RM Calculator: Estimate Your One-Rep Max

    A one-rep max (1RM) is the most weight you can lift once with good form, and you rarely need to test it directly to know it: formulas built on reps-to-failure sets, most commonly the mean of Epley's and Brzycki's equations, estimate it accurately enough to program from, especially at low rep counts. The number itself is only useful once it is translated into a percentage-based training zone — roughly 80%+ for strength, a wide but practically 65-85% band for hypertrophy, and 30-70% moved fast for power, per the 2026 ACSM resistance-training position stand.

    Key takeaways

    • A one-rep max (1RM) is the heaviest single, technically sound rep you can complete — and rep-max formulas estimate it from a lighter set without you needing to actually test it [1].
    • Reppit's own estimate, and the calculator below, use the mean of the Epley and Brzycki equations — correlate well with a measured 1RM (r > 0.95) but carry real, exercise-specific bias [1].
    • Accuracy degrades as the set moves further from failure: R² for predicted 1RM fell from about 0.99 at a 5-rep test to about 0.96 at a 20-rep test in one direct comparison [2].
    • Directly-tested 1RM is itself a highly reliable measurement (median test-retest ICC 0.97 across 32 studies), so it is a legitimate number to estimate toward, not a moving target [4].
    • The number only matters once you turn it into a percentage-based training zone — current guidance puts strength work at 80%+ of 1RM, power at 30-70% moved fast, and hypertrophy across a wide range provided sets are close to failure [5].

    A one-rep max is simply the most weight you can lift for one full, technically sound repetition of a given exercise. It matters less as a number to chase and more as a reference point: percentages of it are how strength and conditioning coaches assign load without guessing. The useful question is rarely "what is my 1RM" in isolation — it is "how do I get a trustworthy 1RM without grinding through an actual maximal attempt, and what do I do with it once I have it."

    Estimate it from a set you've already done

    You do not need to test a true 1RM to get a usable number. Rep-max formulas take a weight and a rep count from a set taken to or near failure and project what one rep at that effort would have weighed. The two most common are Epley's `weight × (1 + reps/30)` and Brzycki's `weight × 36/(37 − reps)`; Reppit, and the calculator below, use the mean of the two [6].

    Estimated 1RM: 114.6 kg (from 100 kg × 5 reps to failure)

    Load (kg) by training goal, from this estimated 1RM
    Goal% 1RMLoadTypical reps
    Strength80-100%91.7-114.6 kg1-6
    Hypertrophy65-85%74.5-97.4 kg6-15
    Power30-70%34.4-80.2 kg1-6
    Rough reps-before-failure at each %1RM — a sanity check, not a per-exercise prediction
    % 1RMTypical repsLoad (kg)
    100%1114.6
    95%2-3108.9
    90%3-5103.1
    85%5-797.4
    80%7-1091.7
    75%9-1385.9
    70%12-1680.2
    65%15-2074.5
    60%18-2468.8

    This assumes the set you entered was taken to or within about a rep of failure — that is what every rep-max formula is built on. Estimates drift further from a true max the more reps you enter, and carry exercise-specific bias, so treat the number as a starting point rather than a measurement.

    Enter a set taken to or within about a rep of failure. The 1RM estimate is the same averaged Epley/Brzycki model Reppit's API uses for estimated rep maxes [S6].

    How accurate is the estimate?

    In the study most directly testing these formulas, LeSuer and colleagues measured actual 1RM on the bench press, squat and deadlift and compared it against seven prediction equations, including Epley's and Brzycki's. Correlations between predicted and measured 1RM were uniformly high (r > 0.95) across all seven — but high correlation is not the same as no error: average predicted values differed significantly from the true 1RM for most equations on the bench press and squat, and every equation significantly underestimated the deadlift [1].

    That exercise-specific bias is the main reason to compare a lift against its own history rather than across lifts, and it is also why a squat 1RM estimate and a deadlift 1RM estimate built from the same formula deserve different amounts of trust.

    If you already log an RPE alongside your sets, the same formula pair can build a load table from any rep/RPE combination, not just a failure set — see the RPE-based load calculator for that version.

    Is it worth testing a true 1RM at all?

    A directly measured 1RM is itself a reliable number: a systematic review pooling 32 studies and roughly 1,600 participants found a median test-retest reliability of ICC 0.97, with 92% of reported reliability coefficients at 0.90 or higher [4]. So a real maximal attempt is not a fundamentally noisy measurement — it is the accepted reference standard that prediction equations are validated against. The tradeoff is practical rather than statistical: a true max attempt costs technical risk, fatigue and recovery time that an estimate from a submaximal set avoids entirely, which is exactly why the estimate is the more common tool for ongoing programming.

    Roughly how many reps a given %1RM allows

    Classic percentage charts assign a fixed number of reps to each %1RM, but a 2023 meta-regression pooling 952 reps-to-failure tests from over 7,000 people found people generally manage more reps at a given %1RM than those older charts assumed, with substantial variation between individuals and a real difference by exercise — the same %1RM allowed noticeably more reps on the leg press than on the bench press. Sex, age and training status barely moved the relationship [3].

    A practical, rounded range at each %1RM — not one study's exact figures, but a synthesis that leans toward the higher counts the updated evidence supports. Individual variation is large; treat this as a sanity check for the calculator above, not a substitute for it.
    % 1RMTypical reps before failure
    100%1
    90%3-5
    80%7-10
    70%12-16
    60%18-24

    What to actually do with the number

    The 2026 ACSM resistance-training position stand — an overview of 137 systematic reviews covering over 30,000 participants, and the first major update to this guidance in 17 years — sets out load ranges by goal. For maximal strength, loads at or above about 80% of 1RM most reliably drove further strength gains. For power, moderate loads of roughly 30-70% of 1RM moved as fast as technique allows were most effective. For hypertrophy, the picture is more permissive: loads anywhere from about 30% to 100% of 1RM produced muscle growth, provided each set was taken close to failure — load itself matters much less than effort, though heavier loads reach that effort in fewer, more time-efficient reps [5].

    A practical band within the ACSM ranges above — not the only loads that work, but a reasonable default per goal.
    Goal% 1RMTypical repsWhy
    Strength80-100%1-6Heaviest loads most reliably drive further 1RM gains [5].
    Hypertrophy65-85%6-15A practical middle of a much wider range that works, as long as sets are close to failure [5].
    Power30-70%1-6Light-to-moderate load moved as fast as technique allows [5].

    Where this breaks down

    How Reppit uses this

    Reppit estimates 1RM through 12RM directly from your logged sets, using the mean of the Epley and Brzycki formulas, and refuses to project past 36 implied reps because the Brzycki denominator breaks down there [6]. Because it runs on every qualifying set automatically, your estimated max updates as your training does, rather than requiring a dedicated max-out session to stay current.

    Is it safe to test my actual one-rep max?

    Directly tested 1RM is a reliable measurement (median test-retest ICC 0.97 across 32 studies) and is commonly used with healthy trained lifters [4]. The more common reason to avoid it day-to-day is practical, not safety: it costs technical risk, fatigue and recovery time that an estimate from a submaximal set avoids, which is why estimation from a lighter set is the more common approach for ongoing programming.

    How accurate is a 1RM calculator?

    Correlations between predicted and measured 1RM are consistently high (r > 0.95), but there is real exercise-specific bias — the same formula can be close on one lift and consistently off on another, with the deadlift most often underestimated [1]. Accuracy is best when the set you enter has roughly 3-8 reps taken close to failure [2].

    What percentage of my 1RM should I lift for muscle growth?

    Anywhere from about 30% to 100% of 1RM builds muscle as long as the set is taken close to failure, per the 2026 ACSM position stand — load matters much less than effort for hypertrophy specifically. A practical default is 65-85% of 1RM for 6-15 reps, which reaches that effort without needing extremely light or extremely heavy loads [5].

    How often should I re-check my 1RM?

    There is no fixed universal interval in the evidence here, but because the estimate is a snapshot that shifts with training progress, fatigue and recovery, treating it as a periodically-refreshed reference rather than a fixed number is more accurate than testing once and assuming it holds for months.

    Sources

    1. [1]The Accuracy of Prediction Equations for Estimating 1-RM Performance in the Bench Press, Squat, and Deadlift — Journal of Strength and Conditioning Research, 1997. observationalSupports: Validates Epley, Brzycki and other rep-based prediction equations against measured 1RM in bench press, squat and deadlift: high correlations (r > 0.95) alongside significant exercise-specific bias, with every equation underestimating the deadlift.
    2. [2]Prediction of One Repetition Maximum Strength from Multiple Repetition Maximum Testing and Anthropometry — Journal of Strength and Conditioning Research, 2006. observationalSupports: Prediction accuracy declines as the test set moves further from failure-adjacent rep ranges: R2 for predicted 1RM fell from 0.993 (chest press) / 0.974 (leg press) at 5RM to 0.955 / 0.915 at 20RM.
    3. [3]Maximal Number of Repetitions at Percentages of the One Repetition Maximum: A Meta-Regression and Moderator Analysis of Sex, Age, Training Status, and Exercise — Sports Medicine, 2023. meta-analysisSupports: Meta-regression of 952 reps-to-failure tests across roughly 7,000 people: more reps are achievable at a given %1RM than older charts assumed, with large between-individual variability, little effect of sex/age/training status, and a real exercise-specific difference between leg press and bench press.
    4. [4]Test-Retest Reliability of the One-Repetition Maximum (1RM) Strength Assessment: a Systematic Review — Sports Medicine - Open, 2020. systematic-reviewSupports: Directly tested 1RM is highly reliable test-retest: median ICC 0.97 across 32 studies pooling roughly 1,600 participants, with 92% of reliability coefficients at 0.90 or higher.
    5. [5]American College of Sports Medicine Position Stand: Resistance Training Prescription for Muscle Function, Hypertrophy, and Physical Performance in Healthy Adults: An Overview of Reviews — Medicine & Science in Sports & Exercise, 2026. position-standSupports: 2026 ACSM position stand, synthesising 137 systematic reviews: loads at or above roughly 80% of 1RM most reliably drive further strength gains; loads of about 30-70% of 1RM moved at high concentric velocity are most effective for power; hypertrophy occurs across a wide load range (roughly 30-100% of 1RM) provided sets are taken close to failure.
    6. [6]Reppit estimated rep maxes: Epley, Brzycki and their average — Reppit (shipped product behaviour), 2026. product-sourceSupports: Reppit's 1RM-12RM estimation from a completed set using Epley, Brzycki and their mean, with a 37-repetition limit.

    Related

    ← All Learn articles