Strength · 11 min read

How to Estimate Your 1RM Without Testing a Max

Written by Coach Ohm · reviewed 2026-08-30

Try it: 1RM and Working Weight Calculator — Estimate a 1RM from a real set, then get working weights and a warm-up ladder.

You need a one-rep max number to program percentages. You do not need to actually lift a one-rep max to get one.

Why not just test it

A true maximum attempt means maximal load with a technique that is degrading under it, usually with a spotter, followed by several days of compromised training. For a competing powerlifter that trade is part of the job. For everyone else it costs a week of quality training and carries injury risk to produce a number that an estimate would have given within a few per cent.

The two formulas

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

They disagree slightly and in opposite directions depending on rep count. The working weight calculator averages them, which is not more scientific — it just avoids inheriting the specific bias of either one.

Three sets, run through both formulas

The clearest way to see the two formulas disagree is to run the same sets through each and look at the size of the gap.

In every row here the two formulas land within a handful of kilograms of each other, never dramatically apart and never exactly the same either. Epley consistently reads a touch higher than Brzycki across all three of these examples. That is not a coincidence of these particular numbers — it is the two formulas built on slightly different assumptions about how strength falls off as reps climb, and the gap between them is the honest, visible size of that disagreement.

Neither formula is "the correct one" being approximated by the other. They are two reasonable models of the same underlying relationship, and the true number for any given person on any given lift sits somewhere in the neighbourhood both of them are pointing at, not necessarily exactly on either line.

What the gap means for picking a weight

A gap of two to four kilograms sounds small, and for most training decisions it is. Percentages in the middle of a rep range — a set of eight, a set of ten — are forgiving of a small error either way; the difference between a working set at 90 kilograms and 93 rarely changes whether the set gets completed as programmed.

The gap matters far more at the top of the percentage table. A single at what one formula calls 100 per cent and a single at what the other formula calls 100 per cent can be the difference between a clean lift and a grinding, technique-breaking one. This is the practical reason to treat the lower of the two numbers as the one to load for anything near a genuine top single or a heavy triple, and to save the higher, more optimistic number for lighter, higher-rep work where a few kilograms either way barely registers. See picking your working weight for how this plays out across the whole percentage table rather than just at the top of it.

The input that decides accuracy

Everything rests on the set you enter being genuinely close to failure. This is where the estimate usually goes wrong, and it goes wrong in the same direction every time: people enter a comfortable set.

A set of 5 you could have taken to 8 produces an estimate that is significantly too low. Every working weight derived from it is then too light, and you spend a training block under-loading. If you are unsure, assume you had more in reserve than it felt like, and treat the output as a floor to be tested rather than a ceiling.

Use 3 to 6 reps

Below three, you are close enough to a true max that the risk you were avoiding comes back. Above about ten, the estimate drifts upward because the limiting factor shifts from maximal force production toward muscular endurance — and how long someone can hold a submaximal load varies between people far more than maximal strength does.

Which lifts this works for

Best on compound barbell movements: squat, bench press, deadlift, overhead press. These have a well-understood load–rep relationship and enough stability for the pattern to hold.

Less reliable on isolation work and machines, where the resistance curve, the stabiliser demand and the machine's own leverage all interfere. You can still use it for choosing weights; just do not treat the number as meaningful in itself.

Somewhere between those two extremes sit unilateral and less standardised barbell variations — a front squat, a close-grip bench, a paused deadlift. The formulas still apply the same way mechanically, but the number they produce describes that specific variation, not the standard version of the lift. A front squat estimate is not a proxy for a back squat estimate, even though both are barbell squats in a loose sense, because the limiting factor at the top of each movement is genuinely different — bar position and torso angle in one case, hip and lower back demand in the other. Keep estimates attached to the exact variation the set was performed on, and do not assume they carry across to a cousin movement that merely looks similar from the outside.

Why an estimate beats a guess but loses to a real test at the extremes

Framed one way, this whole guide is about accepting a small, known amount of imprecision in exchange for avoiding a much larger, less predictable cost. An estimate from a well-executed set of five is typically within a small margin of a true tested max — close enough that every percentage in the working-weight table still lands somewhere sensible. That margin is perfectly acceptable for programming months of ordinary training.

It stops being acceptable at the extremes where a few kilograms of error actually changes the outcome: a maximum attempt in competition, a single load that has to be exactly right for a specific test, a rehab context where being slightly wrong in the wrong direction carries real risk. Those are narrow, specific situations, and they are the only ones where the cost of a true test is worth paying. Almost nothing about ordinary strength training happens in that narrow band, which is exactly why the estimate is the right default tool and the true test is the specialist exception.

Why not just test it, revisited

It is worth returning to the opening question with more nuance, because "never test a true max" is not actually the advice here — there are good reasons to do it occasionally, just not as the default way to find a training number. A competitive powerlifter needs to know their real max under real competition conditions, and no formula substitutes for that specific, narrow purpose. Someone curious about their absolute max once or twice a year, done fresh and well-prepared rather than at the end of an already hard session, is not doing anything unreasonable.

What does not make sense is testing a true max every few weeks as a way of tracking progress, which is where most of the real cost shows up. The recovery hit from one maximal attempt is manageable in isolation. Taken repeatedly, month after month, as the default way of answering "how strong am I now", it turns a measurement tool into a recurring tax on the training it is meant to be informing.

Programming a whole cycle off one estimate

A single estimate is usually asked to serve an entire training block — often four to six weeks of programmed percentages — and it is worth understanding what that implies. Early in the block, the estimate is likely to undersell you slightly, because strength is still building and the number was set at the start. Late in the block, particularly if progress has been good, the same estimate can start to feel easy at percentages that were challenging in week one. Neither of these is the formula being wrong. It is the formula describing a single point in time being asked to hold steady across several weeks of change.

The practical response is to let the reps-in-reserve target do the correcting work that the static percentage number cannot. If a prescribed set starts finishing with more reps in reserve than planned as the weeks go by, that is normal progress inside the block, not a sign the whole programme needs recalculating mid-cycle. Save the actual recalculation — a fresh set fed back into the formulas — for the natural break between blocks, where a stale number gets replaced with a current one rather than patched repeatedly in the middle of a plan that is still running.

Beginners versus experienced lifters

The rate at which an estimate goes stale is not the same for everyone, and assuming otherwise causes different mistakes at different stages. Someone new to a lift can add real strength week to week, which means an estimate calculated in week one can be noticeably out of date by week four — working weights based on it start to feel too light far sooner than the general four-to-six-week re-test guideline suggests, and there is nothing wrong with retesting sooner for a beginner if the working sets clearly stop matching their intended difficulty.

An experienced lifter with years of training history moves the other direction. Strength changes far more slowly once most of the easy progress has already happened, and an estimate from two months ago is often still close to accurate. Retesting on a fixed short schedule regardless of actual signal wastes effort for this group; the four-to-six-week window is a reasonable default, not a number that overrides what the training itself is reporting back.

Re-test rather than recalculate

An estimate goes stale. A beginner outgrows it in weeks; an experienced lifter in a few months. Rather than adjusting the number on paper, put a real set back into the calculator every four to six weeks and let the working weights move with it.

Recording the input, not just the output

One habit worth building alongside the estimate itself: write down the exact set that produced it — the weight, the reps, how many were genuinely left in reserve — not just the resulting number. A bare estimate on its own is easy to trust blindly months later. The same estimate with its source set attached is easy to re-evaluate honestly, because it is obvious at a glance whether the set behind it was a confident, near-failure effort or a cautious one entered on a day that did not feel great. That context is what makes the difference between updating a number with judgement and simply replacing one guess with another.

What to do with the number

Percentages, not guesses. A 5×5 at 80 per cent, a top single at 90, back-off sets at 70 — these are only meaningful if the number they refer to is roughly right. See picking your working weight for how the percentages map to rep ranges.

A failure case: the set that lies

The most common way an estimate goes badly wrong is not a maths error — it is a bad read of the set that went into it. Two versions of this show up constantly.

The first is the comfortable set logged as a hard one. A set of five that could genuinely have continued to seven or eight, entered as if it were close to failure, produces an estimate that is meaningfully too low. Every working weight built from that estimate is then too light for months, and the training block quietly under-delivers without anyone noticing why — the numbers on paper look fine, the sessions just never feel like they are asking enough.

The second, less common but more dangerous, is the opposite: a set pushed past genuine technique breakdown, with bar speed collapsing on the last one or two reps and form changing noticeably to complete them. Feeding that set into the calculator produces an estimate that looks impressively high and is not trustworthy, because the reps that made it up were not the same quality of rep as the ones the formulas assume. The honest fix in both directions is the same: if there is any doubt about how close to failure a set really was, treat the resulting number as approximate and confirm it against how the next few weeks of training actually feel, rather than trusting the first number the calculator returns.

Two weeks of real training data

An estimate is a hypothesis, and two weeks of programming off it is enough to start testing that hypothesis against reality. If every prescribed set at every prescribed percentage is being completed with several reps still in reserve well before the target rep count is reached, the estimate is probably too conservative and can be nudged upward slightly for the next block. If sets are grinding to failure noticeably before the prescribed reps in reserve should be used up, the estimate was likely too generous, and the safer response is to bring the working weights down rather than push through it hoping the numbers catch up.

What does not make sense is re-running the formula itself after two weeks without a new set behind it. The formula did its job the day the set was entered; two weeks of training data tells you whether that particular set was a good read of where you actually were, not whether the arithmetic needs revisiting.

What the number does not tell you

An estimated max describes one lift, on one day, from one set. It says nothing about which specific day you will be strong — day-to-day readiness swings from sleep, stress and how the rest of the week's training has gone move actual performance more than a percentage table can capture. It says nothing about technique under genuine heavy load either; a lifter can have a mathematically accurate estimate and still grind badly under a percentage of it because the bar path or bracing breaks down under weight the lighter sets never tested. And it does not transfer between lifts — a well-calibrated estimate on the bench press tells you nothing reliable about the overhead press, because the two movements fail for different reasons at the limit even though both respond to the same formulas on paper.

A note if you are in a deficit

Expect the estimate to stall or slip late in a fat loss phase. That is not a failure of the programme. It is what happens when you are recovering on less energy, and holding your numbers roughly steady through a deficit is a good result rather than a disappointing one. If an estimate does slip during a deficit, resist the urge to chase it back up with a harder test — a fresh set fed into the calculator once food comes back is enough, and it will very often show the number returning close to where it started once recovery capacity does too.

Read next

• Picking Your Working Weight for Any Rep Scheme How percentages of your max map to rep ranges, why reps in reserve beats training to failure, and how to warm up to a top set without wasting it.
• Strength Training While in a Calorie Deficit What to expect from your lifts during a fat loss phase, what to change in the programme, and which numbers you should stop chasing entirely.
• How Much Protein You Actually Need The protein range that covers almost everyone, why a calorie deficit raises it, and the one-week test that shows whether you are actually hitting it.