If you have ever opened a 1RM calculator and wondered why it gave you a slightly different number than another site, the answer is almost always: different formula. The two formulas powering 90% of estimators are Epley (1985) and Brzycki (1993). They look nearly identical on paper but diverge significantly under fatigue. Knowing which to trust matters when you are programming percentages.

The Two Formulas

  • Epley: 1RM = weight × (1 + reps / 30)
  • Brzycki: 1RM = weight × 36 / (37 − reps)

Both assume an upward-curving "reps vs % of max" relationship. The difference is how steeply the curve bends as reps increase.

How They Compare in Practice

Run 100 kg for 5 reps through each:

  • Epley: 100 × (1 + 5/30) = 116.7 kg
  • Brzycki: 100 × 36 / 32 = 112.5 kg

A 4 kg difference on a moderate set. At 10 reps the gap widens — Epley says 133 kg, Brzycki says 133 kg (they meet at ~10 reps). Above 10 reps, Brzycki underestimates badly; Epley stays workable.

Which Formula Wins Per Rep Range

  • 1–3 reps (true strength): Both are accurate within ±2%. Either is fine.
  • 4–6 reps: Brzycki tends to be slightly more accurate based on Reynolds et al. (2006) and LeSuer et al. (1997). Edge: Brzycki.
  • 7–10 reps: The two converge. Take the average.
  • 11+ reps: Epley wins. Brzycki's denominator (37 − reps) collapses sharply, producing wild overestimates above 12 reps and going undefined at 37.

Where Both Formulas Fail

Predictive equations work best at low reps with grinder mentality. They become unreliable when:

  • Reps were "easy" — sets ended with reps in reserve (RIR > 1) systematically underestimate true 1RM.
  • Reps were grinders with form breakdown — last 2 reps with poor bar path inflate the estimate.
  • Lift involves heavy CNS demand — deadlift 1RM is notoriously underestimated from rep maxes; the curve is flatter than barbell squat or bench.
  • Untrained or detrained lifters — equations were validated on intermediates, not novices. New lifters often outperform their predicted 1RM.

The Smartest Way to Estimate Your 1RM

Our 1RM calculator uses both formulas and averages them — a documented research-backed approach that reduces variance from individual formula bias. Steps:

  1. Perform a max set in the 3–8 rep range (the sweet spot for both formulas).
  2. Take the last rep within 1 RIR (one rep in reserve).
  3. Plug weight and reps into the 1RM Calculator.
  4. Use the average of Epley + Brzycki as your working estimate.
  5. Validate with a true heavy single under supervision every 6–8 weeks.

Programming With the Estimate

Once you have an estimated 1RM, the standard %1RM reference table holds up well:

  • 95% — 2 reps (very heavy work)
  • 87% — 5 reps (strength block sweet spot)
  • 80% — 8 reps (hypertrophy)
  • 75% — 10 reps (volume work)
  • 70% — 12 reps (back-off sets, accessory work)

Re-estimate 1RM every 4 weeks during a strength block, every 8 weeks during hypertrophy.

Bottom Line

Use Brzycki for low reps, Epley for high reps, and the average of both when in doubt — which is exactly what our calculator does by default. No formula replaces a real heavy single, but the right formula gets you to within 3% of it without risking your form on a max attempt.