Brzycki vs Epley is a choice between two curves that agree at 10 reps and disagree either side of it: Brzycki, weight × 36 / (37 − reps), and Epley, weight × (1 + reps / 30). For sets of 1 to 12 reps the two are within a few percent of each other; beyond 12 they part company fast. Zyra Workouts uses both — Brzycki through 12 reps, Epley from 13 — and this page shows the arithmetic behind that choice. It does not crown a winner, because the honest answer is that accuracy depends on the lift, the lifter and the rep count, and we are not citing a trial here.
The two formulas
Brzycki was published in 1993 as a way to predict a one-rep max from a set taken to the point where no more reps were possible1. Its usual form is:
1RM = weight / (1.0278 − 0.0278 × reps)
which simplifies to the version in the app's source:
1RM = weight × 36 / (37 − reps)
Epley is the older and simpler of the two, usually attributed to Boyd Epley's coaching charts from the mid-1980s. It is a straight line:
1RM = weight × (1 + reps / 30)
Every extra rep adds one thirtieth of the weight to the estimate. Ten reps adds a third; thirty reps would double it.
The shape is the whole difference. Epley's line rises at a constant rate for ever. Brzycki's curve rises slowly at first, then faster, because its denominator shrinks toward zero as the reps climb — at 37 reps it divides by zero. That is not a flaw so much as a statement about where the formula was meant to be used: sets of a few reps, close to the max.
In plain terms: Brzycki is a curve built for heavy sets; Epley is a ruler that keeps going.
Where they agree and where they diverge (table)
At 100 kg, rep by rep. The bold rows are where the app switches; everything above the line is what Zyra Workouts actually computes, everything below is shown so you can see why it switches.
| Reps | Brzycki | Epley | Difference | What the app uses |
|---|---|---|---|---|
| 1 | 100.0 kg | 103.3 kg | −3.3 | Brzycki |
| 2 | 102.9 kg | 106.7 kg | −3.8 | Brzycki |
| 3 | 105.9 kg | 110.0 kg | −4.1 | Brzycki |
| 4 | 109.1 kg | 113.3 kg | −4.2 | Brzycki |
| 5 | 112.5 kg | 116.7 kg | −4.2 | Brzycki |
| 6 | 116.1 kg | 120.0 kg | −3.9 | Brzycki |
| 8 | 124.1 kg | 126.7 kg | −2.6 | Brzycki |
| 10 | 133.3 kg | 133.3 kg | 0.0 | Brzycki (identical) |
| 12 | 144.0 kg | 140.0 kg | +4.0 | Brzycki (last rep count) |
| 13 | 150.0 kg | 143.3 kg | +6.7 | Epley (first rep count) |
| 15 | 163.6 kg | 150.0 kg | +13.6 | Epley |
| 20 | 211.8 kg | 166.7 kg | +45.1 | Epley |
Three things stand out.
- A single rep. Brzycki returns the weight itself for a 1-rep set (
36 / 36 = 1); Epley adds 3.3% even to a true single. If you log an actual max, Brzycki reports it as-is. - The crossover. At 10 reps both give
133.3 kg, because36 / 27and1 + 10 / 30are both1.333. Below that, Brzycki is the more conservative estimate; at 11 and 12 it is slightly the higher one. - The runaway. From 13 reps on, Brzycki pulls ahead by a widening margin — 45 kg at 20 reps. No serious reading of a 20-rep set at 100 kg says the lifter could single 212 kg.
One honest side effect of the switch: an estimate from 12 reps (Brzycki, 144.0 kg) is fractionally higher than one from 13 reps (Epley, 143.3 kg). Thirteen reps at the same weight should mean a stronger lifter, and the chart will show a 0.7 kg dip instead. It is a known, small discontinuity; it only appears if you log sets on both sides of 12 at the same weight, and it is far smaller than the error either formula carries at those rep counts.
Why we switch at 12 reps
The switch is a guard, not a claim that one formula is truer. Its job is to stop long sets from painting false records.
Consider a lifter whose heavy work is 5 × 112.5 kg (Brzycki e1RM 126.6 kg) and who finishes with a back-off set of 20 × 90 kg. Under Brzycki alone, the back-off set estimates 90 × 36 / 17 = 190.6 kg and becomes the session's e1RM — a "record" 64 kg above anything the lifter has approached. Under Epley from 13 reps, the same set estimates 90 × (1 + 20 / 30) = 150 kg. Still high, but no longer absurd, and still below what a genuinely heavier day would produce.
Twelve is a natural place to cut because it is roughly where the two formulas stop agreeing (within 3% at 12, 4.7% at 13, 9% at 15) and where sets stop being limited mainly by strength. Some apps cut at 10; some never switch. The app's chart, its records and its plateau flag all run through this one function, so the choice is applied consistently everywhere an e1RM appears.
This is arithmetic, not a trial result. Both formulas are population fits; how far either sits from your tested max depends on your lift and your leverages, and the spread grows with the reps for both. The 1RM calculator shows the two side by side for any set, so you can see how much the choice matters for the reps you actually train in.
In plain terms: keep your estimate sets at 12 reps or fewer and the formula question mostly disappears.
The short version
Brzycki (36 / (37 − reps)) and Epley (1 + reps / 30) agree at 10 reps, differ by under 5 kg per 100 kg anywhere from 1 to 12, and diverge sharply after that. Zyra Workouts uses Brzycki through 12 and Epley from 13 so a long set cannot outscore a heavy one. Neither is "the accurate one" — the estimate from a set of 3–8 reps is close under both, and the trend across sessions is what tells you whether you are getting stronger. The background on what the number means is in what is an estimated 1RM. Zyra Workouts is in development and not yet on the App Store; the rules quoted here are read from its source code and may change before release. Training guidance, never medical advice. Educational overview only — not medical advice.