This 1 rep max calculator estimates the heaviest single you could lift from a set of several reps. Enter the load and the reps you completed, pick the lift, and it runs all seven classic prediction equations side by side: Epley, Brzycki, Lander, Lombardi, Mayhew, O'Conner and Wathen. You get the average and the range across them, plus the NSCA training-load chart value, a percentage-of-1RM table and rep-max loads rounded to your plates. It works in pounds or kilograms for the bench press, squat, deadlift, overhead press or any other barbell lift.
A one-rep max (1RM) is the most weight you can move once through a full range of motion with good technique. Testing it directly means working up through several heavy singles with long rests. The alternative is repetitions to fatigue: take a lighter load for as many clean reps as you can, then predict the single from that set. That is what the equations below do. Each one is a different curve for how quickly the reps you can do fall off as the load climbs toward your max.
How to use the 1RM calculator
- Pick the lift. This doesn't change the math, but it changes the accuracy note, because validation studies found different biases for bench, squat and deadlift.
- Enter the total weight on the bar, including the bar itself, in lb or kg.
- Enter the reps you completed with full range of motion. Best is a recent set of 2–10 reps.
- If you stopped before failure, enter how many more reps you honestly had left (RIR). Leave it at 0 if the set went to failure.
- Choose a main formula or keep the average of all seven, and set your smallest plate jump (5 lb or 2.5 kg by default) so that training loads come out as weights you can actually load.
1RM formulas compared
The equations are as published in LeSuer et al. (1997, Table 1), where w is the weight lifted and r is reps to fatigue. Mayhew et al. (2008, Table 2) print the same set in rearranged form. Brzycki's equation is often written as w × 36 ÷ (37 − r), which gives almost the same numbers. The last two columns show what share of your 1RM each equation thinks a 5-rep and a 10-rep load is.
| Formula | Equation | Shape | 185 × 5 | 5 reps = % 1RM | 10 reps = % 1RM |
|---|---|---|---|---|---|
| Epley (1985) | w × (1 + 0.0333 × r) | linear | 215.8 lb | 85.7% | 75% |
| Brzycki (1993) | 100 × w ÷ (102.78 − 2.78 × r) | linear (reciprocal) | 208.1 lb | 88.9% | 75% |
| Lander (1985) | 100 × w ÷ (101.3 − 2.67123 × r) | linear (reciprocal) | 210.4 lb | 87.9% | 74.6% |
| Lombardi (1989) | w × r^0.10 | power | 217.3 lb | 85.1% | 79.4% |
| Mayhew et al. (1992) | 100 × w ÷ (52.2 + 41.9 × e^(−0.055 × r)) | exponential | 220.2 lb | 84% | 76.4% |
| O'Conner et al. (1989) | w × (1 + 0.025 × r) | linear | 208.1 lb | 88.9% | 80% |
| Wathen (1994) | 100 × w ÷ (48.8 + 53.8 × e^(−0.075 × r)) | exponential | 215.7 lb | 85.8% | 74.2% |
| NSCA chart | w ÷ chart % | lookup | 212.6 lb | 87% | 75% |
At 5 reps the equations sit within a few percent of each other. At 10 reps Epley, Brzycki and Lander still roughly agree (about 75%), but Lombardi and O'Conner say a 10-rep load is nearly 80% of max. As a result, those two give lower 1RM estimates from high-rep sets. The reciprocal equations (Brzycki and Lander) climb steeply at high reps and would divide by zero near 37 reps, which is one reason they should not be used with long sets.
Which formula is most accurate, and for how many reps?
No single equation wins everywhere. The best evidence comes from studies that tested people's real 1RM and compared it with each prediction:
- LeSuer et al. (1997). 67 untrained college students (40 men, 27 women) did a 1RM and a set to fatigue of 10 or fewer reps on the bench press, squat and deadlift. Every correlation between predicted and achieved 1RM exceeded r = 0.95, but the average error still mattered. For the bench, only Mayhew and Wathen did not differ significantly from the true max; the other five underestimated it by 2.2–5.4 lb (0.8–6%). For the squat, only Wathen matched it; the rest were 4.6–16.5 lb low (2–8%). For the deadlift, every equation was 22.2–33.7 lb low (9–14%).
- Mayhew et al. (2008). 103 college women were tested on the bench press before and after 12 weeks of training, using 14 equations. When sets ran anywhere from 2 to 30 reps, Brzycki and Lander broke down (intraclass correlations of just 0.24 and 0.40 before training). Restricted to sets of 10 reps or fewer, the equations used here were all within 4% of the true max on average before training: Brzycki −2.0%, Lander −1.1%, Lombardi −0.9%, Mayhew +1.6%, O'Conner −3.7%, Wathen +0.7%, and +0.5% for the Epley-form equation (listed there as Welday).
- Reynolds et al. (2006). 70 adults aged 18–69 were tested at 1, 5, 10 and 20RM. A 5RM gave the most accurate prediction, and the authors advised using no more than 10 reps in linear equations.
| Formula | Rep range with published support | Notes from validation studies |
|---|---|---|
| Mayhew et al. | 2–10 (also held up to 30 in women) | Among the most accurate for bench press in both studies |
| Wathen | 2–10 | Only equation to match the squat 1RM in LeSuer 1997; accurate for bench |
| Epley | 2–10 | Slight underestimate for bench (2.2 lb, about 1%) and for squat |
| Brzycki | 2–10 only | Good at low reps; large errors when sets run past 10 |
| Lander | 2–10 only | Behaves like Brzycki; unreliable at high reps |
| Lombardi | 2–10 | Flattest curve; gives the lowest estimates from long sets |
| O'Conner et al. | 2–10 | Largest underestimate of the seven for bench and squat in LeSuer 1997 |
That is why this calculator defaults to the average of all seven and also shows the full range. If the range is narrow, the equations agree and your estimate is solid. If it is wide, usually because the set was long, treat the estimate with more caution.
Bench press max calculator: worked example
Say you benched 185 lb for 5 clean reps to failure. Epley gives 185 × (1 + 0.0333 × 5) = 215.8 lb. Brzycki gives 100 × 185 ÷ (102.78 − 13.9) = 208.1 lb, and Mayhew gives 220.2 lb. The seven-formula average is 213.7 lb, and the NSCA chart (5 reps = 87%) gives 212.6 lb. If you stopped that set with one rep left (RIR 1), the calculator treats it as a 6-rep set and the average rises to 219.3 lb. For a program calling for 3 × 5 at 80%, you would load 170.9 lb, which rounds to 170 lb with 5 lb jumps.
Squat, deadlift and overhead press estimates
The same arithmetic applies to every lift, but the errors differ. In LeSuer's group of untrained students, the equations predicted the bench most accurately, then the squat, then the deadlift, where they fell short by roughly a tenth. A 405 lb × 3 deadlift averages 442.7 lb across the formulas. If the study's bias applies to you, your true max could be tens of pounds higher. The overhead press was not part of the validation studies cited here, so use those estimates as a planning figure. The same goes for rows, lunges and machine lifts. Keep the rep count low, and compare one estimate with another only for the same lift.
1RM percentage chart
Programs usually prescribe loads as a percentage of your max. The table multiplies common 1RMs by each percentage, before rounding. The calculator does the same for your own estimate and rounds each load to your plate increment.
| % 1RM | 1RM 100 | 1RM 135 | 1RM 185 | 1RM 225 | 1RM 275 | 1RM 315 | 1RM 405 |
|---|---|---|---|---|---|---|---|
| 95% | 95 | 128.3 | 175.8 | 213.8 | 261.3 | 299.3 | 384.8 |
| 90% | 90 | 121.5 | 166.5 | 202.5 | 247.5 | 283.5 | 364.5 |
| 85% | 85 | 114.8 | 157.3 | 191.3 | 233.8 | 267.8 | 344.3 |
| 80% | 80 | 108 | 148 | 180 | 220 | 252 | 324 |
| 75% | 75 | 101.3 | 138.8 | 168.8 | 206.3 | 236.3 | 303.8 |
| 70% | 70 | 94.5 | 129.5 | 157.5 | 192.5 | 220.5 | 283.5 |
| 65% | 65 | 87.8 | 120.3 | 146.3 | 178.8 | 204.8 | 263.3 |
| 60% | 60 | 81 | 111 | 135 | 165 | 189 | 243 |
| 50% | 50 | 67.5 | 92.5 | 112.5 | 137.5 | 157.5 | 202.5 |
Reps at a percentage of 1RM
The NSCA training load chart, adapted from Landers (1984), links each rep max to a percentage of 1RM. Read it one way to estimate a max: a set of 8 at 160 lb means a max of about 200 lb, the example the NSCA itself gives. Read it the other way to choose loads: with a 200 lb squat, you should manage about 10 reps at 150 lb (75%).
| Max reps | % 1RM | Load if 1RM = 100 | Load if 1RM = 300 |
|---|---|---|---|
| 1 | 100% | 100 | 300 |
| 2 | 95% | 95 | 285 |
| 3 | 93% | 93 | 279 |
| 4 | 90% | 90 | 270 |
| 5 | 87% | 87 | 261 |
| 6 | 85% | 85 | 255 |
| 7 | 83% | 83 | 249 |
| 8 | 80% | 80 | 240 |
| 9 | 77% | 77 | 231 |
| 10 | 75% | 75 | 225 |
| 12 | 70% | 70 | 210 |
These are group averages. The rep count at a given percentage varies with the lift, training status and muscle-fiber make-up. In the Mayhew 2008 study, the same women doing the same percentage of their max changed by anywhere from −15 to +17 reps after training, even though the group average barely moved. If you consistently beat the chart, your true max is probably higher than your last estimate.
Estimating from sets that were not to failure (RIR and RPE)
The equations were derived from sets taken to momentary failure, but most working sets stop short of it. The repetitions-in-reserve RPE scale (Zourdos et al. 2016; Helms et al. 2016) lets you record how close you were: RPE 10 means no reps left, RPE 9 means one rep left, RPE 8 two, and so on. Enter that number as RIR, and the calculator adds it to your completed reps before applying the formulas. Helms and colleagues note that experienced lifters judge RIR more consistently as they get closer to failure. A set rated with 1–2 reps in reserve therefore gives a more reliable estimate than one where you guessed you had 5 left.
Common mistakes that skew a 1RM estimate
- Using a very long set. A 20-rep set measures endurance as much as strength. Beyond 10 reps, the equations disagree by more than 10%.
- Counting partial or assisted reps. A rep a spotter helped with, or one cut short, isn't a completed rep. Count only full-range reps you did yourself.
- Leaving out the bar. A standard men's barbell is part of the load, and forgetting it understates your max.
- Mixing up lifts or variations. A paused bench, a box squat and a sumo deadlift each have their own max. Estimate each separately.
- Estimating when fatigued. A set at the end of a long session underestimates you. The studies tested fresh lifters after a warm-up and with several minutes of rest between heavy attempts.
When to test a true max
An estimate is enough for planning training blocks and tracking progress, as long as you retest the same way each time. A measured 1RM is better when the exact number matters. Examples are a powerlifting meet, where you can compare totals with the DOTS score calculator, or a test that scores a heavy deadlift, such as the 3-repetition maximum deadlift event in the Army Fitness Test calculator. Mayhew et al. (2008) note that a properly supervised 1RM test is considered safe. They also note that rep testing has its own risk: form tends to break down as fatigue builds late in a long set. Either way, warm up, use collars and safety bars or spotters, and stop the set when bar speed or technique falls apart. If strength work is part of a body-composition plan, the body recomposition calculator covers the nutrition side.
Frequently asked questions
How do I calculate my 1 rep max?
Take a set you did close to failure, note the load and the reps, and put them into a prediction equation. Epley is 1RM = weight × (1 + 0.0333 × reps). For 185 lb × 5 that gives 215.8 lb. This calculator runs seven published equations at once and averages them (213.7 lb for the same set).
What is the most accurate 1RM formula?
It depends on the lift. LeSuer et al. (1997) found that, for the bench press, only the Mayhew and Wathen equations did not differ significantly from the tested 1RM. For the squat, only Wathen did. For the deadlift, all seven equations underestimated the tested max by 9–14%. Mayhew et al. (2008) tested college women on the bench press and found the exponential Mayhew equation among the best. Most equations performed well when sets had 10 reps or fewer.
How many reps should I use to estimate my max?
Use 2–10 reps. Mayhew et al. (2008) reported that predictions were more accurate with fewer than 10 reps to fatigue. Reynolds et al. (2006) concluded that no more than 10 reps should be used in linear equations. Past 10 reps, the formulas spread apart quickly.
Is this a bench press max calculator too?
Yes. Choose Bench press, then enter the bar-plus-plate load and your reps. The bench press is the lift most 1RM equations were built and tested on, so it is also where the predictions have held up best in validation studies.
Does the calculator work for squat and deadlift?
Yes, but read the result differently. In the same study of untrained lifters, most equations underestimated the back-squat 1RM by 2–8%, and every equation underestimated the deadlift 1RM by 9–14%. Your true deadlift max may therefore be higher than the estimate.
What does 5 reps at a weight mean as a percentage of 1RM?
On the NSCA training load chart, a 5-rep max is 87% of 1RM, so 1RM ≈ 5RM ÷ 0.87. The equations put a 5-rep load at 84% (Mayhew) to 88.9% (Brzycki) of 1RM.
What is RIR and why does the calculator ask for it?
RIR means repetitions in reserve: how many more reps you could have done. On the RIR-based RPE scale from Zourdos et al. (2016), RPE 10 = 0 RIR, RPE 9 = 1 RIR, and so on. The calculator adds RIR to your completed reps to estimate reps to failure, because the equations were built from sets taken to fatigue.
Should I test a true one-rep max instead?
If you need a precise number, yes. Examples are a competitive lifter peaking for a meet or a coach placing athletes. Mayhew et al. (2008) point out that a properly run 1RM test is not shown to be more dangerous than rep testing, and that fatigue during long rep sets can itself break down technique. For general training, an estimate from a hard 3–8 rep set is usually enough.
Should I include the bar weight?
Yes. Enter the total load: bar plus all plates and collars. A 45 lb bar with a 45 lb plate on each side is 135 lb; a 20 kg bar with two 20 kg plates is 60 kg.
How do I use my 1RM to pick training weights?
Multiply your 1RM by the percentage your program calls for, then round to a load you can build with your plates. The NSCA chart lists about 10 reps at 75%, 8 reps at 80%, 6 reps at 85% and 4 reps at 90%. The calculator lays out every percentage from 50% to 100% and rounds each load to your plate increment.
Sources & method
- LeSuer, McCormick, Mayhew, Wasserstein & Arnold (1997), J Strength Cond Res 11(4):211–213 — accuracy of 7 prediction equations for bench press, squat and deadlift
- Mayhew, Johnson, LaMonte, Lauber & Kemmler (2008), J Strength Cond Res 22(5):1570–1577 — 14 prediction equations for 1RM bench press in women (PubMed 18714230)
- Reynolds, Gordon & Robergs (2006), J Strength Cond Res 20(3):584–592 — predicting 1RM from multiple-RM testing (PubMed 16937972)
- National Strength and Conditioning Association — Training Load Chart (%1RM vs max reps, PDF)
- Helms, Cronin, Storey & Zourdos (2016), Strength Cond J 38(4):42–49 — the repetitions-in-reserve RPE scale
- Zourdos et al. (2016), J Strength Cond Res 30(1):267–275 — novel resistance-training RPE scale measuring repetitions in reserve (PubMed 26049792)
Results are estimates for general information. Found an error? It helps everyone — see our methodology.