Should I ever test a true 1RM?

Only if you compete in a sport that requires it, and then with a spotter, a proper warm-up and a plan. For everyone else the estimate carries the same information at a fraction of the risk and fatigue.

Why do the two formulas disagree?

They were fitted to different groups of lifters. Below 10 reps Epley reads a few percent higher; at exactly 10 they agree; above 10 Brzycki climbs faster and the two quickly drift apart. Pick one and stay with it: consistency matters far more than which formula is nearer the truth.

My estimate dropped. Am I getting weaker?

Probably not, in the short term. Sleep, food, stress and where you are in the week all move a single session by a few percent. A single lower estimate is noise; a downward trend over a month is worth acting on.

Estimated One-Rep Max: How to Read the Number

Your estimated one-rep max, and what it is worth

An estimated one-rep max, or 1RM, is a prediction of the heaviest weight you could lift once, calculated from a set you actually performed. Lift 80 kg for 5 reps and the common formulas put your 1RM somewhere around 90 to 93 kg.

It is an estimate, not a measurement. Its value is as a trend line you can watch without ever testing a true maximum, which is the part most people should skip.

The two formulas worth knowing

Dozens exist; two cover almost every practical case. Both take the weight you lifted and the number of reps you got.

Epley: 1RM = weight × (1 + reps ÷ 30). Reads a few percent higher than Brzycki below 10 reps.
Brzycki: 1RM = weight × 36 ÷ (37 − reps). Reads a little lower below 10 reps, matches Epley exactly at 10, then climbs faster and reads higher.

A worked example

Take a set of 5 reps at 80 kg.

Epley: 80 × (1 + 5 ÷ 30) = 80 × 1.167 ≈ 93.3 kg.

Brzycki: 80 × 36 ÷ (37 − 5) = 2880 ÷ 32 = 90.0 kg.

A three-kilo spread on a set of five. Push the same maths out to 15 reps and they disagree by about 11 kg, with Brzycki now the higher of the two, which is the clearest warning about where they stop working.

When the estimate is reliable, and when it is not

Estimates are at their best between 1 and 5 reps, still useful up to about 8, and increasingly speculative beyond 10. At high rep counts you are measuring endurance and pain tolerance as much as strength, and the formulas were never built for that.

Two people can also produce the same 5-rep set and genuinely different true maxima. Muscle fibre makeup, technique and experience with heavy singles all matter. Compare your estimate against your own history, not against someone else’s.

The set must be close to failure. An easy 5 reps tells the formula nothing useful.
Use a compound lift. An estimate on a cable fly says little.
Keep the conditions similar: same exercise, same rest, same depth or range.
Treat a jump of a kilo or two as noise. The trend over months is the signal.

What to do with it

The practical use is programming. Many plans prescribe percentages ("work up to 5 sets of 3 at 80%"), and an estimated 1RM gives you that number without spending a session testing a true single, which is fatiguing and, without a spotter, best avoided.

It is also a fair way to compare progress across rep ranges. Five reps at 80 kg and eight reps at 72.5 kg look unrelated in a logbook; as estimated maxima they are almost the same lift, which tells you your training is holding steady rather than stalling.

PushBeyond records every set, so the estimate updates from work you already did, with no test day required.