How to find your 1RM safely
Your one-rep max (1RM) is useful for working out program percentages and seeing your strength. But you don't have to find it by grinding out a single every time — and most people don't need to.
Why you want a 1RM
A one-rep max does two jobs. One — working out weights in a program: "5×5 at 80%", "triples at 85%". Two — seeing whether you're getting stronger. Both work without a real max: percentages and progress both run fine off an estimate.
A real 1RM test is for powerlifters before a meet and people who just want to know the number. For everyone else it's risk without much payoff: an all-out attempt loads your joints and technique harder than any working set, and you recover from testing more slowly.
There's a subtlety too: a real max depends on the day. Slept well, ate, got fired up — a few percent more; drained from work — less. One number every six months says less than an estimated max worked out from every workout: a trend of twenty points is more honest than one.
How an estimated max is calculated
The formulas take the weight and the reps of a set done close to failure and estimate what you could lift once. The most common:
- Epley: weight × (1 + reps ÷ 30). 100 kg for 5 → about 117 kg.
- Brzycki: weight × 36 ÷ (37 − reps). 100 kg for 5 → about 113 kg.
At 2–6 reps the formulas nearly agree; beyond that they drift apart. Neither knows you personally: some people have more endurance and get lots of reps off their max, some fewer. So an estimate carries an error of a few percent — but use one formula on similar sets and the trend stays honest. To work it out from your numbers, use the 1RM calculator, which also has a percentage table.
Which sets work
- 3–8 reps — the best window: heavy enough for an accurate estimate, light enough not to take risks.
- Close to failure. If you left 4 reps in reserve, the formula will underestimate — unless you account for RPE (below).
- No longer than 10–12 reps. On long sets endurance decides and the formulas fall apart. In my diary, sets longer than 12 reps don't go into the estimated max at all.
- Clean technique. A set with cheating gives an estimate you don't have.
- One formula for good. Switch from Epley to Brzycki and the whole chart shifts. In my diary you pick the formula in settings, and I recalculate the whole history with it so you never compare apples and oranges.
RPE makes the estimate more accurate
Working sets rarely go to failure. If you did 100 kg for 5 and felt you had two more, that's RPE 8 — strength-wise, like 7 reps to failure. An estimate that counts the reserve gives about 123 kg instead of 117 — closer to the truth. The RPE scale where each point is one rep in reserve was described by Zourdos et al. (2016); experienced lifters rate more accurately, beginners more roughly.
My diary counts it the same way: give an RPE for a set and I'll add the reserve to the reps, but never more than five. To work out "what to load for a triple at RPE 9", use the RPE chart.
If you test it live
Decided to find your real max? Do it on a good day, fresh, without a heavy session the day before.
- A warm-up ladder. The empty bar, then roughly 40%, 60%, 75%, 85% and 90% of your expected max for 5, 3, 2, 1 and 1 reps. Get ready, don't get tired. The warm-up calculator will build the ladder for your weight.
- Attempts. The first around 95% of expected, the second for the record, a third if the second moved easily. More than 3–4 attempts after the warm-up is no longer a test, just fatigue. Jump 2.5–5 kg between attempts on the bench and 5–10 kg on the squat and deadlift — no just-in-case leaps.
- Rest 3–5 minutes between attempts.
- A spotter is a must. Bench with a spotter or in a rack with safety pins; squat in a power rack with pins set just below your bottom position. A deadlift you can drop if it goes wrong.
- The standard is a clean rep. If the bar went up but your technique fell apart, count the previous attempt.
An alternative: a heavy single at RPE 8–9
The middle ground between estimating and testing is warming up to a heavy single that feels like RPE 8–9: "one or two more were there". That gives you almost the same information as a max, without the risk of missing. Do a single like that at the start of a workout every few weeks and watch its weight go up.
Want to check yourself with no risk at all? Do a set of 3–5 reps with one or two in reserve and log it with an RPE. It's a working set you'd do anyway, and the estimated max from it will be accurate.
For your log you need neither: I work out the estimated max from every set you log and show its trend on a chart. If it climbs month after month, you're stronger — even if you've never tested a 1RM. Want to know when you'll hit a round number? The goal forecast works from your estimated max.
Log your sets — I'll work out the estimated max for every exercise myself.