All articles
Training science4 min read

What your estimated 1RM is actually telling you

Your e1RM is not a prediction of what you'd lift on a max attempt. Here is what the number really measures, and how to read it properly.

By Max van Strydonck

Founder of VERTEXPublished

Your estimated 1RM is the number VERTEX shows climbing on your strength chart. It is also the number most people misread.

It is not a promise that you could walk into the gym tomorrow and hit that weight for a single. It is a common unit that lets you compare a set of 5 in March against a set of 8 in July, and tell whether you actually got stronger in between. That is a genuinely useful thing. It is just not the thing most people think it is.

1–8 reps

The range where the estimate is worth trusting

14 kg

How far two respectable formulas disagree at 15 reps

Where the number comes from

You lift a weight for a number of reps. A formula converts that pair into the single-rep load it corresponds to. The two that almost every app and spreadsheet uses were published decades ago and fitted to real lifting data:

  • Epley: 1RM = weight × (1 + reps / 30)
  • Brzycki: 1RM = weight × 36 / (37 − reps)

Both take the same input and both give you a number. They do not give you the same number.

The formulas disagree, and the gap tells you something

Here is what each formula returns for a 100 kg lift at different rep counts:

RepsEpleyBrzycki
1103 kg100 kg
3110 kg106 kg
5117 kg113 kg
8127 kg124 kg
10133 kg133 kg
12140 kg144 kg
15150 kg164 kg

Two things stand out.

They agree exactly at 10 reps — a coincidence of how the two curves were fitted, not a sign that 10 reps is the accurate zone. And they diverge in opposite directions on either side of it: below 10 reps Brzycki is the conservative one, above 10 reps it runs away, predicting 164 kg from a set of 15 where Epley says 150 kg.

Notice also that Epley at one rep returns 103 kg for a 100 kg single. The formula does not know that a single is the max. That is your first clue about how much precision is really on offer here.

Epley vs Brzycki, 100 kg lift

The gap between the two formulas.

How far apart the two standard estimates land at each rep count. They cross at ten and diverge in opposite directions either side of it.

3

1 rep

3 kg apart

4

5 reps

4 kg apart

0

10 reps

they agree

4

12 reps

4 kg apart

14

15 reps

14 kg apart

Calculated from the Epley and Brzycki formulas above for a 100 kg working weight. The agreement at ten reps is an artefact of how the two curves were fitted, not evidence of accuracy.

The disagreement between two respectable formulas at 15 reps is 14 kg. That is the error bar. Any number you read off a high-rep set is carrying it.

The rep range you can trust

The rule is simple: the further the set is from a single, the more the formula is guessing.

How much to trust it

Four zones, by rep count.

1–5 reps

Tight

Both formulas land within a couple of percent of each other and of reality. If you need a number you can act on, take it from here.

Trust it

6–8 reps

Solid

The sweet spot for most people — heavy enough to be informative, not so heavy that you need a spotter and a good night's sleep to run one.

Trust it

9–12 reps

Getting noisy

Usable for tracking a trend, not for planning attempts. Fine as long as you compare it only against other sets in the same range.

Trend only

13+ reps

A fitness marker, not a strength one

At this range you are measuring muscular endurance as much as maximal force, and the formulas were never good here.

Careful
Reliability falls as the set moves further from a single, because the formulas are extrapolating further from the data they were fitted to.

The caveat that matters more than the formula

Every one of these formulas assumes the set was taken close to failure.

This is the part that quietly wrecks most people's data. Ten reps with two left in the tank and ten reps where the last one took six seconds and a lot of noise are the same input to the formula and completely different events in your body. The formula cannot see the difference. It will hand you an e1RM either way, and one of them will be wrong by a wide margin.

If you stop your sets well short of failure — which is often the correct training decision — your e1RM will systematically understate you. That is fine, as long as you are consistent. A number that is reliably 5% low still shows you a real trend. A number that is 5% low some weeks and accurate on others shows you nothing but noise.

Consistency

A number that is reliably 5% low still shows you a real trend. A number that is 5% low some weeks and accurate on others shows you nothing but noise. Log your sets the same way, take them to a similar proximity to failure, and the trend line becomes trustworthy even if any individual point is not.

What to actually do with it

  1. Read the slope, not the point. One session's e1RM is a data point with an error bar. Six weeks of them is a signal. The question worth asking is never "what is my e1RM today" — it is "is this line going up."
  2. Compare like with like. Your bench e1RM from a set of 5 and from a set of 12 are not interchangeable readings. Where you can, track the estimate from a consistent rep range.
  3. Use it to pick loads. This is what it is genuinely good at. If your e1RM says 120 kg and your program calls for a set at 75%, you have a defensible starting point at 90 kg instead of a guess.
  4. Don't test a real max to validate it. A true 1RM attempt is a high-fatigue, non-trivial-risk event that buys you one data point. It does not make your training better. If you are not competing in a sport that requires one, you almost certainly do not need one.
  5. Let plateaus mean something. If the line has been flat for six weeks, that is real information about your programming, recovery, or nutrition. It is not a reason to test your max — it is a reason to change something upstream.

"I should test my real max to find out if the estimate is right."

Costs more than it tells you

A max attempt buys one data point, several days of fatigue and some risk. The estimate was never meant to be accurate in absolute terms — it was meant to be consistent.

The short version

Your e1RM is a well-calibrated ruler with a slightly fuzzy zero point. It measures change extremely well and absolute strength only roughly. Track it from sets of 5 to 8 taken near failure, log honestly, and watch the trend over months rather than sessions.

That is the whole game. The number climbing is the point. The number itself is just the unit it climbs in.

Keep reading