Add weight when the work you planned has become reliably manageable with the same technique and a similar amount of effort. Until then, another well-controlled repetition can be a useful next step. You do not need a personal record every time you train.
Once you have a routine you can sustain, you need a way to adjust it. Awkward dumbbell jumps, a difficult last set, and an interrupted workweek can make that decision less obvious than “lift more every Monday.”
Adding weight is one way to progress
Progressive overload means adjusting the training challenge as your capacity changes. Load is the weight you lift; repetitions are how many times you lift it. Both can change the demand.
Daniel Plotkin and colleagues compared those two approaches in a 2022 randomized trial. Forty-three adults with resistance-training experience entered; 38 completed eight weeks. They trained the lower body twice weekly. One group increased weight while staying within an 8–12-repetition range. The other kept the starting weights and increased repetitions. Most muscle-size outcomes were similar. Squat strength slightly favored increasing weight, but the uncertainty included an advantage in either direction. [1]
This was a short, supervised study in young adults, with sets encouraged to failure. It supports adding repetitions as a legitimate option. It cannot establish that the approaches remain interchangeable indefinitely, or that every population and exercise will respond the same way. [1]
Let the goal influence the next change
Someone trying to build muscle has more room to progress through repetitions than someone whose main goal is lifting a particular heavy weight. That distinction is about the task you want to improve.
In a separate 12-week trial, 49 trained young men completed either higher-repetition, lighter-load or lower-repetition, heavier-load training to volitional failure. Muscle growth was similar. Bench-press maximum increased more with heavier training, by roughly 14 kilograms versus 9 kilograms; the other strength comparisons did not show clear group differences. Both groups also practiced maximum-strength tests during the study. [2]
If heavier lifting is your goal, progressively practicing heavier lifting belongs in the plan. For muscle growth, you have more flexibility in choosing the next change.
Use effort as a guide with room for error
“Repetitions in reserve,” or RIR, means the number of additional repetitions you think you could complete before another repetition becomes impossible with the prescribed technique. Two RIR means you estimate you could do two more. It is an estimate of capacity, rather than a score for how much the set burns.
A study in the Journal of Strength and Conditioning Research tested 24 experienced lifters on the bench press. They called out when they thought they had one or three repetitions left, then continued to failure. Across the trials, the average absolute prediction error was 0.65 repetitions. However, nearly everyone already used RIR, all had worked with a coach, and they had at least three years of training experience. That accuracy should not be promised to a beginner. [3]
An earlier study of 141 gym members, spanning beginners through experienced trainees, found appreciable prediction error and a tendency to underestimate capacity. It asked people to predict their total repetitions before starting, rather than judge remaining repetitions during a set. These are different tasks, which helps explain why the findings cannot be treated as a simple disagreement. [4]
For everyday decisions, combine your effort estimate with what actually happened. Was the movement consistent? Did you finish the planned repetitions without changing the exercise to get them? A precise-looking RIR number should not overrule those observations.
A hard set does not need a failed repetition
In a small eight-week study, 18 trained adults contributed data from leg-press and leg-extension training. Each person trained one leg to failure and the other with an estimated two repetitions left on the leg press and one on the extension. Average quadriceps thickness increased similarly, while the failure condition produced greater losses in repetitions and lifting velocity across sets. [5]
The study was supervised, short, and limited to those muscles and exercises. It supports leaving a small margin as a workable option, without proving an exact universal stopping point. There is no need to turn the progression example below into a maximum-strength test or deliberately fail a repetition to qualify for a heavier weight.
A practical way to decide
The following is an editorial example, not a progression algorithm validated in a trial. Keep the set count and repetition range already appropriate for your program. The example uses three sets of 8–12 seated rows, with an estimated two good repetitions left at the end of each set.
1. Keep the comparison fair. Use the same machine, seat setting, movement range, and reasonably consistent rest between sets. If you shorten the pull or swing your torso farther, a higher number does not answer the same question.
2. Add repetitions while the weight still fits. Suppose 40 pounds produces 10, 9, and 8 repetitions. On later sessions, you might achieve 10, 10, and 9, then 11, 10, and 10, while keeping your intended effort margin. Those are possible steps, not a schedule you must meet.
3. Wait for a repeatable reason to increase. In this example, reach 12 repetitions on all three sets with consistent technique and the intended margin on two comparable sessions. Repeating the result makes a single unusually good day less influential. Two sessions is a practical choice here, not a biological threshold.
4. Try the smallest useful increase. If the machine allows 42.5 pounds, that is a 6.25% increase from 40. Expect that you may complete fewer repetitions. If you can still stay within the chosen range with the intended technique and effort, use that new load as your starting point.
5. Adjust if the jump is too large. If the heavier setting makes even eight controlled repetitions unrealistic, return to the previous load. Smaller approved increments, or equipment with finer steps, may help. Avoid adding improvised attachments that the equipment does not support.
Think of the repetition range as a small staircase between weight increases. You can move up a step before moving the whole staircase higher. The stair spacing will differ by exercise: going from a 20-pound dumbbell to 25 pounds is a 25% jump, much larger than adding five pounds to a 100-pound lift.
When to hold steady or step back
A session that feels harder than usual is a reason to reassess today’s load. It does not, by itself, establish that the program has stopped working. If the planned repetitions require changing your technique or abandoning the intended effort margin, keep the load unchanged or reduce it enough to complete useful work.
After several comparable sessions without progress, review what changed before adding more weight or sets. Has attendance been irregular? Have you shortened rest periods? Are you returning after illness or time away? Did another exercise move ahead of this one? A qualified coach can help separate an equipment problem, a technique issue, and a program that needs adjustment.
Pain that is sharp, worsening, or changing your movement is a reason to stop that exercise and seek appropriate guidance. Do not use a repetition target as permission to push through it. If you have an injury, a medical restriction, or a rehabilitation plan, progression should follow the advice of the professional managing it.
The decision for your next workout can stay small: repeat the load, add a repetition, try a modest increase, or scale back. Make that decision from repeatable performance and your goal. You can leave the gym with a productive session behind you even when the weight on the stack stays exactly where it was.
Evidence and access notes
Sources checked October 19, 2026
1. Plotkin D et al. Progressive overload without progressing load? The effects of load or repetition progression on muscular adaptations. PeerJ. 2022;10:e14142.
DOI and publication record | Read the full paper
Full 19-page published PDF reviewed through an accessible mirror after publisher and PMC access errors. The article’s two-session progression rule is editorial.
2. Morton RW et al. Neither load nor systemic hormones determine resistance training-mediated hypertrophy or strength gains in resistance-trained young men. Journal of Applied Physiology. 2016;121:129–138.
DOI and publication record | Read the full paper
Full published paper reviewed in PMC. Note 49 completers, repeated maximum-strength testing, supervised training, and the bench-press exception. This compares loading ranges; it does not directly test competing progression rules.
3. Refalo MC et al. Accuracy of Intraset Repetitions-in-Reserve Predictions During the Bench Press Exercise in Resistance-Trained Male and Female Subjects. Journal of Strength and Conditioning Research. 2024;38:e78–e85.
DOI and publication record | Read the author-hosted full paper
Full eight-page advance-publication PDF reviewed from an author’s ResearchGate copy; final issue details verified in PubMed. The PDF is dated 2023; the journal issue is 2024. This is a prediction-accuracy experiment, not a long-term progression trial.
4. Steele J et al. Ability to predict repetitions to momentary failure is not perfectly accurate, though improves with resistance training experience. PeerJ. 2017;5:e4105.
DOI and publication record | Read the repository full paper
Full published paper reviewed through the Solent University repository. This was a single-session study using predictions made before each exercise. Its cross-sectional experience groups do not establish that practice caused better accuracy.
5. Refalo MC et al. Similar muscle hypertrophy following eight weeks of resistance training to momentary muscular failure or with repetitions-in-reserve in resistance-trained individuals. Journal of Sports Sciences. 2024;42:85–101.
DOI and publication record | Open the repository and full paper
Full published paper reviewed through Auckland University of Technology. Nineteen trained adults participated; 18 were included in the main analysis. Within-person leg comparison, supervised sessions, a controlled nutrition plan, and a mid-study increase in sets limit generalization.
Related reading: How much strength training to start with.
Explore Fitness Tools or Build Physical Capability.
Reader discussion
Comments are reviewed before appearing publicly. Please keep discussion relevant and respectful.