How to Test Your One Rep Max Safely in 2026 (Step by Step)
Quick answer
Test a one-rep max on a rested day, in a rack with the safeties set just below your bottom position, with a spotter for bench. Warm up in a ladder: empty bar, then 50, 70, 80, 90 and 95 percent of your estimated max for falling reps, resting longer as the weight rises. Take 2 to 3 attempts above that, adding 2 to 5 percent each time, and stop at the first miss or the first ugly rep. With no rack, no spotter, or under six months on the lift, estimate from a hard set of 3 to 5 instead.
A true one-rep max is the only number in strength training that cannot be argued with. It is also the only set you do where a small mistake in setup has a large cost. This guide covers how to run a test so that it produces a real number and nothing else: the day before, the rack, the spotter, the warm-up ladder, the attempts, and the honest cases where you should skip the test and let the one rep max calculator do the work instead.
Before the day
A max test is a snapshot of your nervous system. Set it up to be a good one.
- Rest first. Take at least two easy days before the test, with no heavy work on the lift you are testing. A deload week before a test day is standard for a reason.
- Sleep and eat normally. Nothing special is required, but a short night or a skipped meal can take 5 percent off a max, which is more than a month of training adds.
- Know your target. Run your most recent hard set through the calculator and write down the estimate. The warm-up ladder below is built from that number, and going in with a plan stops you jumping in weight because the bar felt light.
- Book the equipment. You need a power rack with adjustable safeties for squat and bench, and a clear platform for deadlift. If the gym is busy at your usual time, test at a quieter one.
Rack safeties and spotters, stated plainly
For squat and bench press, the test happens inside a rack with the safety bars set, or it does not happen.
Set the safeties one notch below the lowest point of your rep. For the squat, that is just under your hips at the bottom, low enough that a full-depth rep clears them and high enough that a failed rep lands on steel, not on you. For the bench, set them just below your chest height with your back arched as it will be during the lift. Check with the empty bar, then with the first warm-up weight, that a full rep does not touch them. If the rack's notches do not fit, stack a plate under the safeties or choose a different rack.
A spotter for bench stands behind the bar with hands ready but not touching it, lifts off if you ask, and takes the bar only when it stops moving upward or you say the word. Agree the plan out loud before the set: lift-off or not, what you will say if you need help. A spotter who grabs the bar during a grinding rep turns a made lift into a missed one; a spotter who is not there turns a missed lift into an injury. Safeties alone are enough for a bench test if the pins are set correctly, but a spotter is better.
A spotter for squat is optional if the safeties are set, and not a substitute for them. A single spotter behind you can help with a slow rep by lifting under the arms; they cannot catch a dropped bar. If you have no safeties, do not attempt a max squat.
Deadlift needs neither. A missed deadlift is set back down. Keep the area behind you clear.
Collars go on for every attempt above the warm-up. A plate sliding off one side at 95 percent is how shoulders get hurt.
The warm-up ladder
The ladder brings your body up to speed without using up the strength you are about to test. Percentages are of the estimate you wrote down. Rest 1 to 2 minutes on the early rungs and 3 to 4 minutes once you pass 80 percent.
| Step | Load | Reps | Rest after |
|---|---|---|---|
| 1 | Empty bar | 8 to 10 | 1 minute |
| 2 | 50% | 5 | 1 to 2 minutes |
| 3 | 70% | 3 | 2 minutes |
| 4 | 80% | 2 | 3 minutes |
| 5 | 90% | 1 | 3 to 4 minutes |
| 6 | 95% | 1 | 4 minutes |
With a 140 kg estimate, that is 20, 70, 100, 112, 126 and 133 kg, each rounded to the plates you have. The single at 90 percent is the one to watch. If it moved fast and looked like your training reps, the estimate is sound and you can go on. If it moved slowly, treat 95 percent as your first real attempt and adjust the targets down.
Two things this ladder deliberately avoids. It does not do sets of 5 at 80 percent, which tire you; the reps drop as the weight rises so that every rung feels heavier than the last but costs less. And it does not skip rungs. A jump from 70 percent straight to 95 percent feels efficient and produces a first attempt with a nervous system that has not seen the load.
The attempts
Plan for three, expect two, and be content with one.
First attempt: 100 percent of the estimate. If the calculator said 140 kg, load 140 kg. It should feel heavy and move. This is a good result on its own; anything after it is a bonus.
Second attempt: add 2 to 5 percent. After a clean first attempt, go up by 2.5 to 7 kg depending on how it moved. A fast first attempt earns the larger jump; a slow one earns the smallest plate you have.
Third attempt: only if the second was clean. Add another 2 to 3 percent. If the second attempt was a grind, the second attempt is your max and the test is over. A grinding third attempt on a tired nervous system is where technique breaks down and where most testing injuries happen.
Between attempts, rest three to five minutes, sit down, and decide honestly whether the last rep counted: full depth, bar touched the chest, lockout held. A rep you would not accept from someone else does not go in the log.
Stop rules. Stop at the first miss. Stop at the first rep where form changed, even if it went up: a squat that turned into a good morning, a bench where the hips left the pad, a deadlift that hitched. Stop if anything hurts in a way that is not the normal weight of the bar. The number you already have is the number.
Logging what you did
Write down the attempt that counted, the equipment you wore, the date and how it moved. "140 kg, belt, moved slow" is a better record than "140." It sets the baseline for next block's percentages, and it tells you in three months whether 145 kg is a sane target or a hopeful one. In Strongbro the set is logged like any other and the app marks it as a record on the personal records screen, then keeps estimating from your training sets between tests so the number stays current.
When an estimate is better than a test
The calculator exists because most of the time you do not need to test. An estimate from a hard set of 3 to 5 reps sits within about 5 percent of a tested max on bench and squat for most lifters, costs nothing in recovery, and can be repeated every week. Use the estimate instead of a test when any of the following is true:
- You have under six months on the lift. Your technique is still changing. A max attempt tests the weakest link in a chain you have not finished building, and the number you get will be out of date in a month anyway.
- You have no rack with safeties, or no spotter for bench. Heavy triples in the same conditions carry a fraction of the risk and give the calculator everything it needs.
- You are mid-block. Testing costs a session and a few days. Unless the block is designed to end in a test, your top sets already tell you where the max is.
- You are tired, unwell or carrying a niggle. A max on a bad day tells you about the day. The estimate from your next good session is closer to your real strength.
- You want the number for programming, not for its own sake. Percentages built on a fresh estimate work as well as percentages built on a test, and refresh themselves every week. The 1RM percentage chart shows how.
Test when the number itself matters: a competition, the end of a long block you want to close with a hard figure, or a milestone you would like to have actually lifted rather than projected. Two to four times a year covers all of that for almost everyone.
The bottom line
A safe max test is a rested day, a rack with the safeties set just below your bottom position, a spotter for bench, collars on, and a warm-up ladder that climbs from the empty bar through 50, 70, 80, 90 and 95 percent with falling reps and growing rest. Take two or three attempts from your estimate upward in 2 to 5 percent steps, stop at the first miss or the first rep that changed shape, and log exactly what happened. The rest of the year, a hard set of 3 to 5 reps and the calculator will tell you the same number, without the cost.
Get Strongbro on the App StoreFree for 3 days. $19.99 once for lifetime.
Frequently asked questions
How often should I test my one rep max?
Two to four times a year is plenty for most lifters, usually at the end of a training block. A max test costs a session and a few days of recovery. Between tests, estimate from your top sets each week; the estimate tracks the trend, and the test confirms it.
How long should I rest between max attempts?
Three to five minutes between attempts above 90 percent. Heavy singles are limited by the nervous system more than by the muscles, and it needs the time. Shorter rest makes the next attempt feel heavier than it is and turns a strength test into a fatigue test.
Should I test all three lifts on the same day?
Better not. If you must, use the powerlifting order, squat, bench, deadlift, with a long break between lifts, and expect the deadlift to suffer. Spreading the tests across a week, one lift per session, gives each one a fresh nervous system and a truer number.
What if I miss my first attempt?
Rack it or bail onto the safeties, rest fully, and decide honestly whether the miss was technical or a true limit. A technical miss earns one retry at the same weight. A grinding miss where the bar stopped means the previous successful attempt is your max for today, and that is a fine result.
Do I need a belt, sleeves or wraps to test?
Only if you train with them and plan to compare to numbers lifted with them. A belt can add a few percent to squat and deadlift, so a belted max and a beltless max are two different records. Whatever you use, use the same setup each time you test so the numbers compare.
Is a 3-rep max a good substitute?
For most purposes, yes. A hard triple carries far less risk than a single, runs through the same warm-up ladder, and the calculator turns it into a one-rep estimate within a few percent. Many coaches program from a 3RM or 5RM and never test a true single outside competition.
What this is based on
- American College of Sports Medicine — Resistance Training Guidelines Update (2026)
- LeSuer, D. A. et al. (1997). The Accuracy of Prediction Equations for Estimating 1-RM Performance in the Bench Press, Squat, and Deadlift. Journal of Strength and Conditioning Research, 11(4), 211–213.
- Brzycki, M. (1993). Strength Testing — Predicting a One-Rep Max from Reps-to-Fatigue. Journal of Physical Education, Recreation & Dance, 64(1), 88–90.
Get Strongbro on the App StoreFree for 3 days. $19.99 once for lifetime.