Most lifters want their one-rep max for a practical reason: to set percentages in a program. A few want it for the pleasure of lifting something heavy. The first group rarely needs to test at all. The second group should do it with a plan.
First question: do you need a real test?
An estimate from a set of 3 to 6 reps is usually enough to set training weights. Our one-rep max calculator turns a set into a number with three formulas. The trade-off is accuracy. Research on prediction equations shows strong correlations with tested maxes, yet individual errors of 10 percent or more still occur, according to the sources we reviewed. On a 225 lb bench, 10 percent is 22 lb.
A direct test makes sense when the number itself matters: a competition, a rehab milestone you are supervised for, or a program that calls for it. If you are not sure why you want it, estimate.
What the research says about safety
A 2020 systematic review pooled 32 studies of direct 1RM testing. Test and retest results were closely matched: typical day-to-day variation was about 4 percent, and the range across studies ran from 0.5 to 12.1 percent. The authors note the test has been shown safe across different populations, including children, older adults and people with clinical conditions, though they were describing supervised studies with set protocols. Those conditions are not the same as testing alone in a garage.
Study protocols commonly used a warm-up, a series of progressively heavier attempts, and rests of two to five minutes. Many allowed three to five attempts to find the max. We could not find a single standard protocol in what we read, so what follows is a sensible composite and not an official procedure.
Safe setup
Choose the exercise carefully. The bench press and a machine-supported leg press are easier to bail out of than a back squat or a deadlift. Anything you cannot drop or dump safely deserves extra caution.
Use safeties. For bench, that means a power rack with the pins set just below chest height, or a competent spotter, or ideally both. For squats, set the pins below your lowest depth. Do not collar the bar if you may need to tip the plates off.
Test when fresh. Do it early in a session, on a day when sleep and food have been normal. Avoid testing the day after a hard session for the same muscles.
Have a person nearby who knows what to do, even if the lift does not need a spotter.
The warm-up and the attempts
A common approach works up in stages. Start with an empty bar for 8 to 10 easy reps. Add weight to reach roughly half of your estimated max for 5 reps, then 60 to 70 percent for 3, then about 80 percent for 2 reps, and about 90 percent for a single. Rest between sets for a couple of minutes, longer as the weights climb. Those percentages are our own illustration, not figures from a source.
Then make single attempts, each one a small jump. For upper body lifts, a typical jump is 2.5 to 5 percent. Take two to five minutes between attempts, in line with the rests the studies used. Stop when a rep slows noticeably or your form changes. The last successful clean rep is your max. A miss ends the test; do not try again that day.
Counting what counts
Use the same standards each time. On the bench, decide whether you pause on the chest and whether the butt stays down. For a squat, decide on depth. A max that gets lighter-looking over time because the standards crept is not progress.
Estimating instead
If you decide against a test, use a submaximal set. Choose a weight you can lift 3 to 6 times with good form, finish a set close to failure with a spotter or safeties, and enter the result into a calculator. Fewer reps give better accuracy than many, and a review in our sources reports a 5-rep set predicted bench max more closely than 10 or 20 rep sets.
You can also estimate by reversing the process. If your plan calls for 5 reps at 85 percent and you are unsure of your max, start lighter and add weight across a few weeks until 5 reps feel hard but clean. That avoids a max test altogether.
Who should not test
People recovering from injury, with joint problems, with heart or blood pressure issues, who are pregnant, or who are new to lifting should get professional advice first, and many should skip testing altogether. Teenagers and older adults can be tested in studies, but we advise supervised conditions for them. If you are tired, ill or in pain, postpone.
Estimates and tests are both snapshots. Pick one method and keep it consistent, and let the trend, not any single day, guide your training.