One Rep Max Calculator

Estimate your one-rep max from a submaximal set under any of seven published equations, see how far apart they are, and get the training loads with the plates for each.

Pounds (lb) Kilograms (kg)

Weight lifted

Reps completed

reps

Formula

Bar and plates — 45 lb

Bar

Estimated 1RM

262.5 lb

Epley (default) estimate

Range across the seven

253.1–267.8 lb

The seven equations disagree by 5.8%.

Training max (90%)

235 lb

What 5/3/1 and most percentage programs want

How reliable

High — ±2–4%

Effective reps — the more you did, the less the estimate is worth

Training loads

% of 1RM	Load	Plates per side	Reps	Use it for
100%	260 lb 262.5 lb	2×45 + 1×10 + 1×5 + 1×2.5	1	Max strength
95%	250 lb 249.4 lb	2×45 + 1×10 + 1×2.5	2	Max strength
90%	235 lb 236.3 lb	2×45 + 1×5	3	Max strength
85%	225 lb 223.1 lb	2×45	5	Max strength
80%	210 lb 210 lb	1×45 + 1×25 + 1×10 + 1×2.5	8	Strength
75%	195 lb 196.9 lb	1×45 + 1×25 + 1×5	10	Strength
70%	185 lb 183.8 lb	1×45 + 1×25	13	Size
65%	170 lb 170.6 lb	1×45 + 1×10 + 1×5 + 1×2.5	16	Size
60%	155 lb 157.5 lb	1×45 + 1×10	20	Endurance

Loads are snapped to the nearest 5 lb — the smallest jump on a bar with 2.5 lb plates. The rep count is what the percentage implies, not what the rounded load implies.

How the seven formulas compare

Formula	Estimate	vs. yours
Epley (default) w × (1 + r/30)	262.5 lb	0%
Brzycki w × 36 / (37 − r)	253.1 lb	-3.6%
Wathen 100w / (48.8 + 53.8·e^(−0.075r))	262.3 lb	-0.1%
Lombardi w × r^0.10	264.3 lb	+0.7%
O'Conner w × (1 + r/40)	253.1 lb	-3.6%
Mayhew 100w / (52.2 + 41.9·e^(−0.055r))	267.8 lb	+2%
Lander 100w / (101.3 − 2.67123r)	255.8 lb	-2.5%

The seven equations LeSuer et al. (1997) tested. At 5 effective reps they disagree by 5.8%.

How this number was reached

5 reps with 0 in reserve = 5 effective reps.
Epley (default): w × (1 + r/30) = 262.5 lb
Estimated 1RM = 262.5 lb
90% of 262.5 lb is 236.3 lb, snapped to the nearest loadable 235 lb

These are estimates, not guarantees. If you do test a true max, warm up properly and use a spotter or safety bars.

Updated: September 8, 2026

One rep max calculator. Your 1RM from one set, with the range across seven formulas.

A one rep max calculator estimates the heaviest single you could lift, worked back from a set you have already finished. This one also asks how close the set was to failure, reports how far apart all seven published equations are on your numbers, and turns the estimate into weights you can load.

What an Estimated One Rep Max Is

A one-repetition maximum is the heaviest load you can lift once through a full range with acceptable form. An estimated 1RM, written e1RM in most training logs, is that number worked back from a set you have already finished, so nobody has to load a bar they might not be able to put back down. Epley's equation reads a 225 lb set of 5 as a 262.5 lb max. Run the same set through the other six published equations and the answers land between 253.1 lb and 267.8 lb — a 14.7 lb argument about one lifter's bench.

The seven equations on this page are the ones LeSuer and colleagues put through measured bench, squat and deadlift maxes in 1997. They were fitted on different lifts, different populations and different rep ranges, which is why they disagree, and why the shape of the disagreement changes as the reps climb. Wikipedia's consolidated list runs to more than a dozen; the seven here are the ones put through one shared data set.

Two inputs move an estimated 1RM more than the choice of equation does. The first is where the set stopped. A 225 lb set of 5 taken to failure estimates 262.5 lb; the same set with two reps still in reserve estimates 277.5 lb, and that 15 lb gap is wider than the one all seven equations opened between themselves. The second is how many reps the set lasted: at 5 effective reps the seven sit 5.79% apart, at 12 they sit 12.64% apart, and at 30 they sit 266% apart. That is why the card labeled How reliable turns amber at 11 effective reps and red at 13, and why the estimate is published next to its own width.

How to Use the 1RM Calculator

The page opens on a finished calculation: 225 lb for 5 reps taken to failure, read through Epley, on a 45 lb bar.

1. Set the unit pair first. Pounds and Kilograms sit at the top of the form. Switching converts the weight you typed and leaves the bar alone, because a 45 lb bar converted twice drifts off 20 kg and would corrupt every plate row underneath it.

2. Enter the weight and drag the reps slider. The slider stops at 12, since 12 reps with nothing in reserve already lands in the amber band.

3. Answer Reps left in the tank. Zero means the set ended because the next rep was not going to happen. This is the field most 1RM pages do not have, and getting it wrong is the largest single error available to you.

4. Choose a formula, or choose Average of all seven. Epley is the default here, and it is what StrengthLog and Built With Science publish by name; the tool ranking first for this term names no equation at all.

5. Open Bar and plates only if you are not on a standard bar. The fold is closed and carries the current bar in its own title, so a link that sets a 35 lb bar still says 35 lb while shut. No bar (machine or dumbbell) blanks the plate column and leaves the rest of the table intact.

The answer comes back as one large card with the estimate, three smaller ones — Range across the seven, Training max (90%) and How reliable — and a gauge of effective reps with ticks at 5, 10 and 12. Training loads sits below them and stays open: nine rows from 100% down to 60%, each carrying the load, the plates per side, the reps that percentage implies and what the row is for, with the row nearest the set you entered highlighted. How the seven formulas compare and How this number was reached are folded shut until you want them. A warning block appears under everything at 11 effective reps and above.

The Seven Equations, and the One That Runs by Default

\text{Estimated 1RM} = w \times \left(1 + \frac{r}{30}\right)

$w$ = The weight you lifted for the set, in pounds or kilograms. The default state uses 225 lb.
$r$ = Effective reps: the reps you completed plus the reps you left in the tank. Five reps with two in reserve gives r = 7.
$\text{TM}$ = Training max: 90% of the estimate, snapped to the nearest loadable weight. 90% of 262.5 lb is 236.3 lb, which becomes 235 lb.

Epley is the default here because it is the equation named on the pages that name one at all.

\text{Estimated 1RM} = w \times \left(1 + \frac{r}{30}\right)

At one effective rep this page returns the weight you lifted and stops. Left to themselves, five of the seven overshoot a declared single — Epley by 3.3%, Mayhew by 8.9%, O'Conner by 2.5%, Lander by 1.4%, Wathen by 1.3% — and reporting a 225 lb single back as a 244.9 lb max is the most visible defect in the field. Only Brzycki and Lombardi are already exact there.

Here are all seven, with what each says about 225 lb for 5 reps taken to failure:

Epley — w × (1 + r/30) → 262.5 lb
Brzycki — w × 36 / (37 − r) → 253.1 lb
Wathen — 100w / (48.8 + 53.8·e^(−0.075r)) → 262.3 lb
Lombardi — w × r^0.10 → 264.3 lb
O'Conner — w × (1 + r/40) → 253.1 lb
Mayhew — 100w / (52.2 + 41.9·e^(−0.055r)) → 267.8 lb
Lander — 100w / (101.3 − 2.67123r) → 255.8 lb

Their order is not stable. Brzycki ties O'Conner at the bottom of the seven at 5 effective reps — the two equations return the same number there — and climbs to second highest at 12, where it reads 266.4 lb against Epley's 259.0 lb. Picking the equation that flattered you once means picking it for a rep count you may never use again.

The training max is a second, smaller step, and the one your program will use. The estimate is multiplied by 0.9 and snapped to a weight the bar can hold: 236.3 lb becomes 235 lb. Every row of the loads table takes its percentage from the estimate. A program asking for 85% of your TM therefore wants 85% of 235 lb, which is not the 225 lb sitting on the 85% row.

How Far Apart the Seven Formulas Are, by Effective Reps

Effective reps	Spread across the seven	Reliability shown
1	0.00%	High — ±2–4%
2	8.34%	High — ±2–4%
3	7.66%	High — ±2–4%
4	6.81%	High — ±2–4%
5	5.79%	High — ±2–4%
6	5.63%	Good — ±5%
7	5.56%	Good — ±5%
8	6.39%	Good — ±5%
9	7.14%	Good — ±5%
10	7.80%	Good — ±5%
11	9.40%	Rough — about ±10%
12	12.64%	Rough — about ±10%
15	24.82%	Unreliable
20	56.95%	Unreliable
30	266.01%	Unreliable

Worked One Rep Max Examples

225 lb for 5, nothing left in the tank

Two plates a side on a 45 lb bar, five reps, the last one slow. Epley returns 262.5 lb. The Range card reads 253.1–267.8 lb and notes that the seven equations disagree by 5.8%. Training max comes back as 235 lb, from an unrounded 236.3 lb. The highlighted row of the loads table is the 85% one, and it is worth reading twice: 225 lb, plates 2×45 per side, 5 reps. The table has handed back the same set you typed in, which is what a percentage table looks like when it agrees with itself.

The same set with two reps still in you

Change nothing but Reps left in the tank, from 0 to 2. The estimate moves to 277.5 lb — 15 lb above the failure reading, a wider gap in pounds than the 14.7 lb between the highest and lowest of the seven equations on the same set, so the field you skipped mattered as much as the formula you argued about. A line under the cards now says the set is being treated as a 7-rep max, the reliability card steps down from High — ±2–4% to Good — ±5%, and the training max rises from 235 lb to 250 lb. Every load moves with it: the 85% row is 235 lb instead of 225 lb, and the 80% row is 220 lb.

185 lb for 12, where one number stops being enough

Twelve reps to failure on 185 lb gives 259.0 lb through Epley, 237.2 lb through Lombardi and 267.2 lb through Lander. The Range card spans 30 lb, 12.6% apart, and the reliability card is amber: Rough — about ±10%. The warning block opens under the tables and says the same thing in the calculator's own words. Two rows deserve a look: the 70% row asks for 180 lb at 13 reps and the 60% row for 155 lb at 20, and both rep counts are tinted amber because they sit past where the equations still agree.

100 kg for 8 on a 20 kg bar

Set the pair to Kilograms, type 100 kg and drag the slider to 8. Epley gives 126.7 kg, the seven span 120.0–127.7 kg at 6.4% apart, and the training max is 115 kg. The plate column earns its place here: the 100% row asks for 127.5 kg, which is 2×25 + 1×2.5 + 1×1.25 per side, and the 85% row asks for 107.5 kg, which is 1×25 + 1×15 + 1×2.5 + 1×1.25. Loads snap to the nearest 2.5 kg in metric and the nearest 5 lb in pounds, with the unrounded figure printed underneath so you can see what was given up.

Average of all seven, and the rep column that moves with it

Leave the set at 225 lb for 5 and switch the formula to Average of all seven. The estimate becomes 259.9 lb, 2.6 lb under Epley's, while the Range card does not move at all — the range is always the seven, whichever one you are reading. What does move is the Reps column, because those counts come from inverting the formula you picked: under Epley the 90% row says 3 reps and the 75% row says 10, under the average they say 4 and 11. Training max lands on 235 lb either way.

Mistakes That Move Your 1RM

Calling a set failure when it was not — on 225 lb for 5, two reps left in the tank are worth 15 lb, or 5.7%. Most calculators that rank for this term have nowhere to say the set had anything left, and the few that do ask for it as an RPE rating rather than a rep count.
Trying to load the exact percentage — 85% of 262.5 lb is 223.1 lb, and no combination of plates gets you there. The row snaps to 225 lb and keeps 223.1 lb underneath so the 1.9 lb is visible. A tool that prints 223.1 lb as the prescription is handing you a number you have to round yourself in the middle of a session.
Reading the estimate as a measurement — it is one number pulled from seven that disagree. At 5 effective reps they span 14.7 lb on this set; at 12 they span 30 lb. The Range card exists to keep that width on screen.
Picking the formula that gives the biggest number — Mayhew is the highest of the seven at 5 effective reps and Lombardi is the lowest at 12. Neither ordering is a fact about your strength. Average of all seven is in the dropdown for anyone who catches themselves shopping.
Estimating from the last back-off set of the session — the equations were fitted where the reps are few. Twelve reps to failure puts the seven 12.64% apart and the reliability card into amber; fifteen would put them 24.82% apart. Feed it the heaviest hard set you did, not the final one.
Treating the Reps column as a target — those counts are what each percentage implies under the formula you selected, so they change when the formula does, and they are tinted amber once they climb past 12. They describe what a row is for; they are not a number of reps to chase to failure.

What This Estimate Does Not Tell You

An estimated 1RM says nothing about whether the number is any good. Bodyweight, sex, age and training history decide that, and this page holds none of them. There is no exercise selector either: the same seven equations run whether the weight came off a bench, a squat rack or a leg press.

The exercise does matter, and this page cannot use it. Marzagao's 2026 analysis fitted a weight-dependent equation on 303,494 near-failure sets across 388 exercises and found that the conversion factor moves with the load itself: at light weights each extra rep implies a larger fraction of your maximum than it does at heavy ones. It is cited in the sources below and not implemented, for a reason worth stating: the equation is unit-sensitive by construction. The same 100 kg lift run through it in kilograms and in pounds disagrees with itself: 117.5 kg read in kilograms against 252.8 lb — 114.7 kg — read in pounds, 2.5% on a set of 100 kg for 5, and a calculator whose answer depends on which button you pressed is worse than one that admits it is an approximation.

The Use it for column is this page's rounding, not a published boundary. Only the 85% line has a standard behind it, as the floor the NSCA puts under maximal-strength work. ACSM's 2009 position stand splits the same territory at 80-100%, 70-85% and below 70%, and the NSCA continuum breaks at 67% rather than 65. The four labels here are round numbers chosen to line up with the nine rows of the table.

The spread does not widen step by step with the reps, which matters if you read the reference table above as a trend line. At one effective rep it is zero: the page hands back the weight you lifted and never calls the equations. It jumps to 8.34% at 2 effective reps, falls to its narrowest point of 5.56% at 7, and climbs without interruption only from 8 upward: 7.80% at 10, 12.64% at 12, 24.82% at 15, 266.01% at 30. Two effective reps is a wider disagreement than ten. The last two rows are past anything this form will produce — 12 reps with 5 in reserve is 17 effective reps, and that is the ceiling; they are printed because other calculators will hand you a number at 30 reps without a word about it.

Lifting Terms on This Page

One-rep max (1RM)

The heaviest weight you can lift once through a full range with acceptable form. Measured by attempting it under a spotter or safety bars; estimated by everything on this page.

Estimated 1RM (e1RM)

The output of a prediction equation, not of a lift. Training logs write it e1RM to keep it apart from a max somebody has stood up under.

Effective reps

What this page feeds the equations: the reps you completed plus the reps you left in the tank. Five reps with two in reserve is a 7-rep max, and the gauge under the cards is scaled in these rather than in the reps you typed.

Reps in reserve (RIR)

How many more reps you could have done when you racked it. RPE counts the same thing from the other end, since RPE = 10 − RIR, so RPE 8 and 2 RIR describe the same set.

Training max (TM)

90% of the estimate, snapped to a loadable weight — 235 lb from a 262.5 lb estimate. Wendler's 5/3/1 and most percentage programs run every prescribed weight off this number instead of the max itself, and the 10% haircut is what makes a bad day survivable.

AMRAP set

As many reps as possible: a set taken to or near failure, most often the last one of a lift. It is the set worth feeding into a 1RM calculator, with an honest RIR attached.

Plates per side

What hangs on each end of the bar, not the total. Two plates a side is 225 lb on a 45 lb bar in a US gym, which is where this page's default set comes from.

One Rep Max Questions Lifters Ask

How do I calculate my one rep max without testing it?

Take one hard set, note the weight, the reps and how many reps you had left, then run it through a prediction equation. Epley multiplies the weight by (1 + reps/30), so 225 lb for 5 reps gives 262.5 lb. Every number on this page starts there.

Which 1RM formula is the most accurate?

No single one wins. Brzycki reads lower than Epley at low reps and higher at twelve: 253.1 lb against 262.5 lb on a 225 lb set of 5, then 266.4 lb against 259.0 lb on 185 lb for 12, and the seven together span 14.7 lb. If you have no reason to prefer one, Average of all seven is in the dropdown, and the Range card tells you what the choice was worth on your set.

How many reps should I use for the most accurate estimate?

Between three and eight, taken close to failure. That is where the seven equations sit closest together — 5.56% apart at 7 effective reps, their narrowest point — which is why the reliability card reads High up to 5 effective reps and Good up to 10. At 12 they are 12.64% apart, and above 12 the page stops treating the answer as a load and starts treating it as a ballpark.

What do reps in reserve do to the estimate?

They are added to your reps before the equation runs. Five reps with two in reserve is treated as a 7-rep max, and on 225 lb that moves the estimate from 262.5 lb to 277.5 lb — 15 lb, or 5.7%. With a calculator that has no RIR field the workaround is to add the reps you had left and type the total; this page does the addition and prints the effective rep count it used.

Is RIR the same as RPE?

They are the same information counted from opposite ends: RPE = 10 − RIR. RPE 8 means two reps left, RPE 10 means none. This calculator asks for reps in reserve because a rep is a thing you can count.

What is a training max, and why is it 90%?

A training max is the number your program runs its percentages off, set below your real max on purpose so that a bad day still leaves you with clean reps. Wendler's 5/3/1 made 90% the convention, and that is what the Training max (90%) card shows: 235 lb from a 262.5 lb estimate, snapped down from 236.3 lb to something the bar can hold. The loads table works off the estimate, not the training max, so a program asking for 85% of TM wants 85% of 235 lb.

Does this work for the bench press, squat and deadlift?

Yes, and the same seven run for all three, because there is no exercise selector. LeSuer's team tested every one of them against measured bench, squat and deadlift maxes, and each takes a weight and a rep count with no idea which lift produced them. Marzagao's 2026 analysis of 303,494 near-failure sets across 388 exercises found the conversion factor moves with the load, and cites earlier work for the reps-to-percentage relation moving by exercise too, so read a machine or dumbbell estimate as looser than a barbell one.

Can I use a 35 lb women's bar, or no bar at all?

Yes. The Bar and plates fold offers 45, 35, 25 and 15 lb bars, their metric equivalents at 20, 15, 10 and 7 kg, and No bar (machine or dumbbell). On a 35 lb bar, 95 lb for 3 reps with 1 in reserve reads 103.6 lb through Brzycki, the seven span 103.6–110.7 lb, and the 100% row loads 105 lb as 1×25 + 1×10 per side. With No bar selected the plate column drops to dashes and every other column stays.

Can I trust an estimated 1RM enough to program from it?

To set training weights, yes; to claim a number, no. The arithmetic is exact and the prediction is not — Epley on 225 lb for 5 reps returns 262.5 lb here and on any calculator that names Epley, including StrengthLog, which prints the same equation. What varies is which equation you ask for. The seven disagree by 5.79% on that set and by 12.64% on a 12-rep one, and the reliability card puts a figure on it, from ±2–4% at five effective reps or fewer to Unreliable at 13 and above.

Why does the reps slider stop at 12?

Because the equations stop agreeing there. At 12 effective reps the seven are 12.64% apart and the reliability card is already amber; at 15 they are 24.82% apart, and at 30 — where StrengthLevel's percentage table still hands out a number, with no formula named and no warning attached — they are 266% apart. Reps in reserve can still push the effective count to 17, which is where the card turns red.

Is the calculator free, and does it save my sets?

It is free and there is no account. Nothing is stored either, so there is no history and no export, but the address bar carries whatever you changed from the defaults, which means a bookmark or a shared link reopens the same set.

How do I turn this number into next week's training?

Write the block off the training max rather than the estimate. The loads table is already in weights you can load: at a 262.5 lb estimate that is 235 lb for triples, 225 lb for fives and 210 lb for eights, with the plates for each printed beside them. Then leave the number alone for a training block. Re-estimating after every session turns a figure meant to steer six weeks into a mood ring.

Sources & References

Content verified by the Smart Calculators Team