Every set does not need to end in failure
You can build muscle without taking every working set to failure. In the experiments below, stopping close to failure produced growth, and reaching failure did not consistently improve every outcome. The useful question is how effort fits the exercise, the number of sets, and the work still to come.
Our practical interpretation is to make the stopping point deliberate and record it. Leaving about one or two estimated reps in reserve is a reasonable option for many working sets; it is not a proven universal optimum. The studies do not establish that very easy sets are interchangeable with hard ones, or that failure is always unnecessary.
Define what ended the set
Momentary muscular failure means you cannot complete another repetition through the intended movement under the set’s conditions. Reps in reserve (RIR) is your estimate of how many more reps you could have completed. Two RIR means roughly two left, not two unfinished reps you must attempt.
On the lifting-specific RPE scale, RPE 8 corresponds to about two RIR, RPE 9 to one, and RPE 10 to zero. The original RIR-based RPE study examined perceived effort in novice and experienced squatters. Treat the rating as feedback, not a precise measurement of unused capacity.
Stopping because technique changes, grip slips, or pain appears is a different observation from failing a repetition. Record the reason instead of calling every difficult stop “failure.” Keep range of motion and technique standards consistent; otherwise a higher rep count can describe a different task.
Two studies, two useful pieces of context
Close to failure can be sufficient
Refalo and colleagues (2024) studied 18 trained adults for eight weeks, with each person’s legs assigned to different conditions. Leg presses and extensions went either to failure or to an estimated two and one RIR, respectively. Average quadriceps thickness increased similarly, while failure produced greater repetition and lifting-velocity loss. Results differed somewhat by measured muscle region. This supports stopping close to failure in that setting; it cannot establish the best effort for every muscle or program.
Low-volume training adds a counterpoint
Hermann and colleagues (2025) assigned 42 trained young adults to failure or two RIR. They performed one set of nine exercises twice weekly for eight weeks. Both groups improved; several muscle-size measures tended to favor failure, with generally modest differences. Strength and local endurance changes were similar. RIR estimates also varied by exercise and improved with practice. A single-set routine is a different context from repeated sets with more work to follow.
These are relatively small, short studies in trained adults. Their results are evidence about specific training conditions, not a guarantee of your response. Neither a hard final rep nor one unusually good workout proves that a program will produce more growth.
A worked log: separate the plan from the result
This is a hypothetical machine-row entry, not participant data or a prescribed workout. The lifter plans three working sets with about two RIR, using the same machine and two minutes between sets. The final column records what actually happened.
| Set | Load | Reps | Estimated RIR | Note |
|---|---|---|---|---|
| 1 | 50 kg | 12 | 2 | Same range of motion |
| 2 | 50 kg | 11 | 2 | Two-minute rest |
| 3 | 50 kg | 10 | 0 estimated | Last clean rep felt maximal; no failed attempt |
The log contains 33 completed reps. The final set exceeded the planned effort even though no repetition was attempted and failed. That distinction is worth preserving. Next time, compare the same exercise and rest interval, then decide whether the load or rep target needs adjusting to match the intended effort. Do not rewrite the recorded RIR to match the plan.
Review effort alongside the rest of the session
- Record immediately: exercise, load, completed reps, estimated RIR, rest, and why you stopped.
- Keep comparisons consistent: changing equipment, technique, or exercise order can explain a different result.
- Look beyond the first set: note whether later sets and the next comparable session still fit the plan.
- Change one clear variable: explain any effort, rest, or volume change in your notes so you can interpret the next session.
Use the free workout log for the individual entries and the weekly-set tally to review the amount of work. Set count alone cannot tell you how hard those sets were. If short rests are driving the drop in reps, review rest intervals before treating every lower number as lost strength.
Common questions
Is zero RIR the same as a failed rep?
Zero RIR is an estimate that no further clean rep was available. You can record that after completing a final rep without attempting and failing another. Distinguish your estimate from an observed failed attempt.
Should I add failure sets when progress slows?
A plateau alone does not identify the cause. First review consistency, progression, rest, effort, and the number of working sets. These studies do not justify adding failure work automatically.
Do I need to attempt failure to learn RIR?
The rating is an estimate that can improve with practice and consistent feedback. This guide does not require failed attempts. Exercise choice, experience, and a suitable setup matter; a qualified coach can help you assess technique and effort.
Sources & method
Written by the Brolic team. These references explain the evidence and assumptions; this page does not replace individualized professional advice.
- Refalo et al. — Failure versus repetitions in reserve (2024) — Eight-week, within-person leg-training experiment in 18 trained adults.
- Hermann et al. — Without Fail (2025) — Eight-week, single-set training trial in 42 trained adults.
- Zourdos et al. — RIR-based rating of perceived exertion (2016) — Primary study of the lifting-specific RPE scale in experienced and novice squatters.