Dumbbell Bench Press 1RM from 20 Reps — Epley and Brzycki Estimate
If you can dumbbell bench press 50 kg for 20 reps, your estimated one-rep max is 94.6 kg — the average of the Epley estimate (83.3 kg) and the Brzycki estimate (105.9 kg). That is the headline number, and the rest of this page shows exactly how it was calculated, what loads to train with at every percentage of it, and how far you should trust it.
The Dumbbell Bench Press works the chest, triceps and shoulders, and it is a dumbbell lift. The worked example uses two 25 kg dumbbells, 50 kg in total — at twenty reps the formulas diverge a lot, and the page is honest about that. The worked example below uses a 20-rep set with 50 kg as the load — that is an assumption for illustration, not a prescription for you. Put your own weight and reps into the One Rep Max Calculator and the same two formulas will estimate your own max in seconds.
An estimated one-rep max matters because almost every serious strength programme is written in percentages of it. "Five sets of five at 80%" only means something if you know what 100% is, and testing a true one-rep max is risky, fatiguing and unnecessary for most lifters. A submaximal set of 20 reps gives you the number without the risk — provided you understand the limits of the formulas, which this page sets out honestly.
Use the one-rep-max calculator for your own numbers
The worked example on this page uses 50 kg for 20 reps. Your numbers will be different, and the calculator does the same maths with them. Use the One Rep Max Calculator for this lift, or the general One Rep Max Calculator for any lift, and you will get the Epley estimate, the Brzycki estimate, their average, and the full training-percentage table for your own max.
For comparison, here are the other one-rep-max pages in this batch — the same two formulas applied to different lifts and rep counts:
- Bench Press 1RM from 3 Reps — Epley and Brzycki Estimate
- Squat 1RM from 5 Reps — Epley and Brzycki Estimate
- Deadlift 1RM from 6 Reps — Epley and Brzycki Estimate
The formulas, shown and cited
Two published formulas estimate a one-rep max from a submaximal set. Both are shown here with their sources, and both are worked through with this page's numbers.
Epley (Epley 1985): 1RM = weight × (1 + reps ÷ 30)
Brzycki (Brzycki 1993): 1RM = weight × 36 ÷ (37 − reps)
Worked through for 50 kg × 20 reps on the dumbbell bench press:
- Epley: 50 × (1 + 20 ÷ 30) = 50 × 1.667 = 83.3 kg
- Brzycki: 50 × 36 ÷ (37 − 20) = 50 × 36 ÷ 17 = 105.9 kg
- Average of the two: (83.3 + 105.9) ÷ 2 = 94.6 kg (209 lb)
The two formulas agree to within 22.5 kg on this set. That agreement is itself useful information: when Epley and Brzycki land close together, the estimate is more trustworthy than when they diverge. Both formulas lose accuracy above ten reps, and this page is built from a 20-rep set, so treat the result as a rough guide. The two formulas disagree by 22.5 kg here, which is the formulas telling you they are being stretched beyond the range they were designed for. Use the estimate to set training loads, not to predict a competition attempt.
Neither formula is magic. Both were fitted to real lifting data, and both assume the set was taken close to failure with good technique — a set of 20 reps with two more reps left in the tank will underestimate your max, because the formulas assume those 20 reps were hard. If you stopped well short of failure, add that to your interpretation: the estimate is a floor, not a ceiling. And if your technique broke down on the last rep — a squat that turned into a good morning, a bench press with the hips off the bench — the set does not count for estimation purposes at all.
Training percentages for a 94.6 kg dumbbell bench press max
Most programmes prescribe loads as a percentage of your one-rep max. The table below gives the load for every percentage from 50% to 100% of the 94.6 kg estimate, so you can read your working weights straight off it without doing the multiplication in your head between sets.
| % of 1RM | Load (kg) | Load (lb) |
|---|---|---|
| 50% | 47.3 kg | 104 lb |
| 55% | 52.0 kg | 115 lb |
| 60% | 56.8 kg | 125 lb |
| 65% | 61.5 kg | 136 lb |
| 70% | 66.2 kg | 146 lb |
| 75% | 71.0 kg | 156 lb |
| 80% | 75.7 kg | 167 lb |
| 85% | 80.4 kg | 177 lb |
| 90% | 85.1 kg | 188 lb |
| 95% | 89.9 kg | 198 lb |
| 100% | 94.6 kg | 209 lb |
How those percentages are used in practice depends on the goal. Strength work lives at 85–95% of 1RM for sets of 1–5 reps, with long rests of three to five minutes. Hypertrophy work lives at 65–85% for sets of 6–12 reps, with rests of 60–120 seconds. Muscular endurance work lives at 50–65% for sets of 15 or more reps, with short rests. The same max produces all three training zones; what changes is the percentage and the rep count, not the lift.
A practical example: if your programme calls for five sets of five at 80% on the dumbbell bench press, that is 75.7 kg for five reps, five times, with three minutes of rest between sets. If it calls for three sets of ten at 70%, that is 66.2 kg for ten reps, three times, with 90 seconds of rest. The table above is the only arithmetic you need; the programme supplies the sets, reps and rest.
How to use the result, and how to progress safely
Use the 94.6 kg figure as a planning number, not as a weight to attempt tomorrow. The safest way to use an estimated max is to base your training percentages on it for four to six weeks, then re-test with a fresh submaximal set and update the estimate. If the new set gives a higher estimate, your training weights go up; if it gives the same number, you have maintained; if it gives a lower number, you may be under-recovered and need a lighter week.
Progression should be small and regular. For the dumbbell bench press, adding 2.5 kg to the bar when you complete all prescribed sets and reps is a sensible rule for most lifters, and 1.25 kg steps are better for smaller lifts like the overhead press and the curl, where 2.5 kg is a large percentage jump. Doubling the jump because the estimate went up is a mistake — the estimate is an average of two formulas, and it carries error bars of several kilograms either way.
Warm up properly before any heavy set. For a dumbbell bench press session, that means the empty bar, then 50% of your working weight for a few reps, then 70%, then your working sets — never jumping straight from the warm-up to a heavy load. And never test a true one-rep max alone, without spotters or safety bars. The whole point of estimating from a 20-rep set is that you never have to find out what your true max is by attempting it.
Recovery is part of the programme, not a break from it. The dumbbell bench press stresses the chest, triceps and shoulders, and those muscles need 48–72 hours before they are trained hard again. Sleep, food and rest days are where the strength is actually built; the training session is just the stimulus. A lifter who adds weight to the bar every week while sleeping five hours a night will stall and then regress, and no formula can fix that.
What the number does NOT tell you
An estimated one-rep max is a useful number with clear limits. First, it is an estimate, not a measurement. Your true max can only be found by attempting a maximum lift, and the estimate can be off by 5–10% in either direction even from a good set. That is why the percentage table above should be treated as a guide: if 80% feels like 90% on the day, trust the feeling, not the table.
Second, the formulas assume good technique. A dumbbell bench press performed with a shortened range of motion, bouncing, or momentum will produce a higher rep count than a strict set, and the formulas will reward that with an inflated max. Compare estimates only between sets done the same way — same depth, same pause, same standard. A stricter set that gives a lower estimate is more useful than a sloppy set that gives a higher one.
Third, a one-rep max says nothing about your health, your body composition or your overall fitness. It measures one thing — the most weight you can lift once in this movement — and a big number in one lift does not imply a big number in another, let alone general health. It also says nothing about injury risk: a lifter with a large estimated max and poor technique is closer to an injury than a lifter with a modest max and excellent technique.
Finally, the formulas are population tools. They were fitted to groups of lifters, and individuals vary. Some lifters are naturally better at high reps than at singles, and the formulas will underestimate their max; others are the opposite, and the formulas will overestimate. Over time you will learn your own pattern — whether your true max tends to sit above or below the estimate — and you can adjust your training percentages accordingly.
Frequently asked questions
What is my dumbbell bench press 1RM if I lift 50 kg for 20 reps?
Using the worked example on this page: Epley gives 83.3 kg and Brzycki gives 105.9 kg, so the averaged estimate is 94.6 kg. That figure assumes the 20-rep set was taken close to failure with good technique. For your own weight and rep count, use the One Rep Max Calculator.
Which formula is better, Epley or Brzycki?
Neither is clearly better in general — that is why this page averages them. Epley (1985) tends to give slightly higher estimates at higher rep counts, while Brzycki (1993) is often a little more conservative. From sets of ten reps or fewer the two usually agree within a few kilograms, and the average is a sensible compromise. Above ten reps both become less reliable, and the gap between them is a warning sign rather than a detail to average away.
How accurate is a 1RM estimated from 20 reps?
Both formulas lose accuracy above ten reps, and this page is built from a 20-rep set, so treat the result as a rough guide. The two formulas disagree by 22.5 kg here, which is the formulas telling you they are being stretched beyond the range they were designed for. Use the estimate to set training loads, not to predict a competition attempt. The estimate is best used to set training loads for a four-to-six-week block, then re-tested. It is not a substitute for a properly supervised max attempt if you genuinely need to know your true maximum, for example before a competition.
What weight should I train with for strength on the dumbbell bench press?
Strength work is usually 85–95% of your estimated max. For a 94.6 kg max, that is 80.4 kg to 89.9 kg, for sets of 1–5 reps with three to five minutes of rest. For muscle growth, use 65–85% (61.5–80.4 kg) for sets of 6–12 reps. The full percentage table above covers every step from 50% to 100%.
How often should I re-test my dumbbell bench press max?
Every four to eight weeks is enough for most lifters. Testing more often wastes training time and adds fatigue without new information, because strength changes slowly. Re-test with the same rep count and the same technique standard as last time, so the two estimates are genuinely comparable, and update your training percentages from the new figure.
Related pages
Estimates only, not medical advice. Minimum safe intakes: 1,200 kcal/day women, 1,500 kcal/day men — consult a healthcare professional before a very-low-calorie diet.
