e1RM
Menu
Menu

Original data asset

1RM Formula Accuracy Study

Which 1RM formula is most accurate depends on rep range, lift, population, and set quality. This page publishes e1RM's transparent formula-spread benchmark and summarizes the public validation literature behind the seven formulas.

Seven formulas
Rep-range benchmark
Dataset schema

Short answer

For most lifters, the most reliable 1RM estimate comes from a clean set of three to six reps. In that range, Brzycki, Lander, and the formula mean are usually safer than the highest estimate. Above seven reps, use the lower formula cluster or retest with heavier weight.

This first version is not claiming private clinical validation data. It combines published 1RM validation evidence with an original e1RM formula-spread benchmark. Anonymous user-submitted true-max comparisons will be added only after consented collection is available.

Best formula by rep range

Scroll table horizontally

Rep range Best practical choice Formula spread Recommendation
1-3 reps Formula mean or conservative cluster Low set-noise; formula outliers still possible The set is already close to max strength, but do not chase the highest formula.
4-6 reps Brzycki, Lander, or formula mean Lowest practical uncertainty Best practical range for estimating 1RM from normal strength work.
7-10 reps Brzycki, O'Conner, or lower cluster Higher biological noise Use conservative estimates or retest with heavier weight.
10+ reps Do not rank formulas from this input Very high biological noise Muscular endurance dominates. Retest with 3-5 reps.

e1RM formula-spread benchmark

For each rep count below, e1RM calculates all seven formulas using a fixed 100 kg working weight. Because all formulas scale linearly with weight, the percentage spread is the same at any load. The "closest formula" is closest to the trimmed formula mean after removing the highest and lowest outputs.

Scroll table horizontally

Input Trimmed mean Lowest Highest Spread Closest formula
100 kg x 2 105.6 kg Brzycki: 102.9 kg Mayhew: 111.4 kg 8.0% Wathan
100 kg x 3 109.1 kg Brzycki: 105.9 kg Mayhew: 114 kg 7.4% Wathan
100 kg x 4 112.3 kg Brzycki: 109.1 kg Mayhew: 116.5 kg 6.6% Wathan
100 kg x 5 115.4 kg Brzycki: 112.5 kg Mayhew: 119 kg 5.6% Wathan
100 kg x 6 118.7 kg O'Conner: 115 kg Mayhew: 121.5 kg 5.5% Lombardi
100 kg x 7 122 kg O'Conner: 117.5 kg Wathan: 124 kg 5.3% Lombardi
100 kg x 8 125.1 kg O'Conner: 120 kg Wathan: 127.7 kg 6.2% Lander
100 kg x 9 128.2 kg O'Conner: 122.5 kg Wathan: 131.2 kg 6.8% Brzycki
100 kg x 10 131.5 kg O'Conner: 125 kg Wathan: 134.7 kg 7.4% Mayhew

Published validation evidence

Scroll table horizontally

Source Population Lift / test Takeaway
Mayhew et al. 1992 College men and women after resistance training Bench press Bench-press prediction depends on relative muscular endurance and the selected percentage of 1RM.
LeSuer et al. 1997 Healthy men and women Bench press, squat, deadlift Different equations perform differently by lift; no single formula is universally best.
Knutzen et al. 1999 Older adults Machine resistance exercises Prediction equations can be useful but acceptable error varies by exercise and population.
Reynolds et al. 2006 Resistance-trained adults Bench press and leg press Lower-rep multiple-RM tests are strong predictors; higher-rep tests add more noise.

Method and limitations

The benchmark measures formula disagreement, not direct biological truth. It is useful because formula spread is the uncertainty the user actually sees when entering one set into a calculator. Lower spread usually means the input set is more stable for programming.

The published studies show why the answer changes by lift and population. A formula that behaves well for bench press in one group may not be best for deadlift, machine exercises, older adults, or high-rep sets. That is why e1RM shows all seven formulas instead of hiding the disagreement.

Future versions will add anonymized user-submitted comparisons where the same lifter reports a recent submaximal set and a confirmed true 1RM for the same exercise and movement standard.

Practical recommendation

Use three to six reps when possible. If the seven formulas differ by more than about 5%, pick the lower estimate for training or repeat the test with a heavier set. If the set was above ten reps, treat the output as a rough planning number, not a max-strength fact.

Sources

  • Brzycki 1993 - Strength testing: predicting a one-rep max from reps-to-fatigue.
  • Mayhew et al. 1992 - Relative muscular endurance performance as a predictor of bench press strength.
  • LeSuer et al. 1997 - The accuracy of prediction equations for estimating 1-RM performance.
  • Reynolds et al. 2006 - Prediction of one repetition maximum strength from multiple repetition maximum testing and anthropometry.
  • Knutzen et al. 1999 - Validity of 1RM prediction equations for older adults.
  • Wood et al. 2002 - Accuracy of seven equations for predicting 1RM performance of apparently healthy, sedentary older adults.