Estimated 1RM: the formula, why reps in reserve belong in it, and how Method calculates yours

The problem an estimated 1RM solves

A training log is full of sets that cannot be compared directly. 100 kg for 8, 110 kg for 5, 120 kg for 3. Which is the strongest? An estimated one rep max, usually written e1RM, maps each set to the single-rep load it implies, so they sit on one scale.

The classic formula is Epley's.

e1RM = weight × (1 + reps / 30)

100 kg for 8 gives 126.7 kg. 110 kg for 5 gives 128.3 kg. 120 kg for 3 gives 132 kg. Now the three sets can be ranked.

The approach has a long record. In a study of 70 adults performing 1, 5, 10, and 20 rep maximum tests on the bench press and leg press, one rep max could be estimated accurately from multiple-rep tests, the 5RM gave the most accurate prediction, and the authors concluded that no more than 10 repetitions should be used in linear prediction equations [1]. A separate study of 103 women found prediction equations for the bench press were more accurate when fewer than 10 repetitions to failure were used, both before and after 12 weeks of training [2].

Why reps in reserve belong in the formula

Epley assumes the set went to failure. Most training sets do not. 100 kg for 5 with 3 reps in reserve is closer to a set of 8 than a set of 5, and an estimate that ignores that will undercount every submaximal set and overstate every grinder.

The fix is to count the reps you could still have done. Method calls these effective reps.

effective reps = logged reps + reps in reserve
e1RM = weight × (1 + effective reps / 30)

So 100 kg for 5 at 3 RIR becomes 100 × (1 + 8 / 30) = 126.7 kg, the same as 100 for 8 to failure. If you log in RPE, the conversion is 10 minus RPE, so RPE 7 is 3 RIR; that mapping is the resistance-training RPE scale introduced by Zourdos and colleagues, where RPE 10 means 0 RIR and RPE 9 means 1 RIR [3], and its use in programming is laid out by Helms and colleagues [4]. Method stores one canonical value and shows whichever scale you prefer. See RIR vs RPE.

If no reps in reserve were logged, Method treats the set as having zero. The estimate is then plain Epley.

Which sets count

Method computes an estimate for a set only when all of these hold.

• It has a weight above zero and reps above zero.
• It is a working, backoff, failure, or custom set. Warm-up sets and drop sets are excluded.
• It was completed, inside a completed workout.

Warm-ups are excluded because they are not a strength expression. Drop sets are excluded because the reps were done fatigued and the formula would read them as weakness.

One point per training day

A session might hold four working sets of the same lift. Plotting all four adds noise without information. Method keeps the best estimate from each session, then the best session from each calendar day in your timezone, and plots one point per training day. The chart therefore answers "what was the strongest set-implied max that day", which is the question a strength series should answer.

How Method describes the change

Method does not fit a trend line through those points. It reports the literal change between the first and last point in the window you pick, along with the minimum and maximum, and states how many sessions it was based on. A sentence such as "Estimated 1RM increased by 7.5 kg over last 3 months, based on 12 logged Bench Press sessions" is a description of the data, not a verdict on your training. With fewer than two training days in the window, Method says there is not enough data.

Windows are 1 month, 3 months, 6 months, 1 year, and all time.

Where else the estimate is used

The same formula appears in three other places, so every number agrees.

• Suggested loads. Method estimates a 1RM from your most recent eligible set and inverts it for the target reps and RIR of the next set. See how Method suggests loads.
• Records. A strength PR is a completed set whose estimated 1RM, or reps at a weight, or load, exceeds every earlier set for that exercise.
• History PDF export. Per-exercise charts use the same estimates.

Reading the number honestly

An estimate built from a set of 3 is better than one built from a set of 12. Prediction error grows with reps, which is why the studies above recommend keeping estimates to sets of 10 or fewer [1][2]. Technique, range of motion, fatigue, and how precisely you judged your reps in reserve all move the estimate. Method shows the sets behind every point so you can see what produced it. If one point looks wrong, open the session and look at the set.