Junk volume: how to spot sets that add fatigue but not growth

Push day ends with three sets of cable crossovers for 15 reps. Each one stops with five reps still in the tank, because the session has run long and the next lift is the car park. The sets go in the log and the weekly count for the chest goes up by three. In the app's growth-reps model, those reps add nothing. That is junk volume: sets that cost time and fatigue but sit too far from failure, or too deep into a session, to add much muscle.

The fix is rarely fewer sets across the board. It is finding the few that did nothing and either taking them closer to failure or moving them to another day. The numbers below come from compute.mjs, the script that runs the app's own formulas, the same ones behind the free tools. Every other figure carries a citation.

What counts as junk volume

Junk volume has two common forms, and high weekly volume on its own is not one of them.

The first is a set that ends too far from failure. A set of 15 with five reps left is mostly a long warm-up. It loads the joints and gives the muscle little reason to grow.

The second is a set that comes too late in a session. Once a muscle has had a lot of sets in one sitting, each extra set adds less. A 2025 analysis put the point where more sets in a session stop adding a detectable gain in muscle size at about 11 fractional sets, counting assisting lifts as half [1]. It is a preprint, not yet peer reviewed, and its authors say to read that figure cautiously.

What junk volume is not: a high weekly total. In a trial of 34 trained men over 8 weeks, the higher-volume groups (three and five sets per exercise, against one) grew significantly more in three of the four sites measured [2]. The largest analysis so far, 67 studies and 2,058 participants, found muscle growth kept rising with weekly sets, at a diminishing rate [3]. More sets cost more time and recovery for a smaller extra gain. They do not stop working at a line.

Warm-ups are not junk volume either. They are not meant to grow anything, and neither the app nor the research counts them. The post on whether warm-up sets count as volume covers that rule.

Why sets far from failure count for little

Counting sets only measures training if the sets are hard. A review of 14 studies concluded that the number of sets is a fair measure of volume for muscle growth when the sets are taken to failure or near it, across roughly 6 to 20 or more reps [4]. A set stopped with five or six reps left is not what those studies counted.

The closer to failure, the more growth, up to a point. A 2024 series of meta-regressions found that muscle size increased as sets ended closer to failure, while strength gains were similar across a wide range of reps in reserve [5]. The authors call it exploratory, because reps in reserve had to be estimated from each study's description.

Failure itself is not required. A 2023 meta-analysis of 15 studies found no advantage for training to momentary failure over stopping short of it (effect size 0.12, 95% CI -0.13 to 0.37) [6]. The useful zone is close, not all the way. The RPE chart explains how RPE and reps in reserve map onto each other: RPE 10 is failure, and RPE 8 means two reps left.

Where the app draws the line

The app needs a rule it can apply to every set, so it uses a simple one. A working set at RPE 6 or higher, which is four reps in reserve or fewer, counts every rep as a growth rep. A set below RPE 6 counts none. So 15 reps at RPE 6 give 15 growth reps, and the same 15 reps at RPE 5.5, four and a half reps in reserve, give 0.

That cliff is a simplification. The research shows a slope, not a step, and a set at RPE 5.5 is not worthless in the body the way it is in the model. Treat the line as a flag. A set that lands below it is worth a second look, not an automatic delete.

A push day where 4 of 12 sets did nothing

Here is a chest-focused push day as one lifter might log it, run twice a week. Two light warm-up sets on the bench come first and are left out of everything below.

The session totals 58 growth reps from 12 working sets. Four of those sets, the three crossovers and the last set of dips, added none.

The weekly count does not see that. By the app's set rules, all 12 sets train the chest directly, so the session gives the chest 12 sets. Twice a week, that is 24 weekly chest sets, which the weekly sets calculator reads as Very high. On paper the chest is drowning in work. In the growth-reps column, 4 of the 12 sets are filler.

There is a second problem. Twelve chest sets in one session sit past the 11 or so where the preprint found extra sets stop adding a detectable gain [1]. The app's own rule for that (weekly sets divided by 11, rounded up) says 12 sets want two sessions, not one.

Two ways to fix it

The blunt fix is to cut the four sets. Bench, incline and one hard set of dips make 8 working sets, and the session still totals 58 growth reps. The weekly chest count drops to 16, which reads On target on the calculator. Same growth reps from 8 sets instead of 12, and a shorter session.

The better fix is to keep the crossovers but make them count. Drop the easy set of dips, and take two sets of crossovers for 12 reps to RPE 8 and 8.5. That is 10 working sets and 82 growth reps per session. The weekly chest count lands on 20, which the calculator reads as High: more time and recovery for a smaller extra gain, a fair trade if the chest is the priority.

Neither version makes the session longer. Both get more out of it. For a muscle that needs more than 11 sets a session, the answer is another day, and the post on training frequency shows how to split them.

How to spot junk volume in your own log

Start with RPE, because without it nothing else works. A set logged without RPE gives 0 growth reps in the app, and a session logged with no RPE at all shows the caption "Log RPE to track growth reps" on its summary. That is not a judgment that the set was easy. It is the model refusing to guess. The estimated 1RM makes the opposite assumption and treats a set without RPE as RPE 10, which is why a missing RPE can make a light set look like a max on one screen and like nothing on another.

Once RPE is logged, compare two numbers after each session: working sets and growth reps. The session summary shows them as Growth reps, next to its other figures. If a session's working sets climb and its growth reps do not, the new sets are the suspects.

Then look at where the low-RPE sets sit. Junk volume tends to collect in three places in a log. The last exercise of a long session is one. Isolation work tacked on after compounds that already hit the same muscle is another. The third is the final set of an exercise, when the plan said three sets and the lifter did a fourth for the sake of a round number.

Finally, read the weekly count with care. The weekly sets per muscle count works the same way for every working set: it does not look at RPE. A muscle can read On target with half its sets at RPE 5. The count tells you how much you did. The growth reps tell you how much of it was hard.

Mistakes that turn good sets into junk

Rating RPE generously is the most common one. A lifter who stops a set of crossovers at 15 because the plan said 15, with five reps left, will often log RPE 7 out of habit. The set then counts in full and the log looks better than the session was. Reps in reserve are hard to judge on light, high-rep sets, so if a set of 15 felt easy, test it once: keep going to near failure and see how many reps were really left. The post on reps in reserve for hypertrophy covers how close to go.

Adding sets instead of effort is the next. When a lift stalls, the reflex is a fourth set. If the first three were at RPE 6, the fourth will be too. Taking the existing sets one RPE point harder usually does more than adding another easy one.

Stacking every chest movement into one day is the third. A program can look balanced across the week and still pile 12 or more sets for one muscle into a single session. Check per session, not just per week.

The last mistake runs the other way: logging warm-ups as working sets. A ramp-up set at RPE 4 logged with the normal set role counts toward the weekly total and drags the session's average RPE down. Mark it as a warm-up and it drops out of both, which is where it belongs.

Questions

Is junk volume real or a gym myth?

It is a gym term rather than a research one, but the idea holds up in part. Counting sets only measures training well when the sets are near failure [4], so very easy sets add little. Very high weekly volume is a different matter: growth keeps rising with more weekly sets, just at a slower rate, and no study has found a weekly count past which it stops [3].

How many reps in reserve is too many for a set to count?

There is no hard cut-off in the research, which shows growth rising steadily as sets end closer to failure [5]. The app draws a working line at four reps in reserve, RPE 6. A set at RPE 6 or harder counts every rep as a growth rep, and a set below it counts none. Use the line as a prompt to look at the set again, not as proof the set did nothing.

Should I cut sets or train harder to remove junk volume?

Train harder first. A set at RPE 5 becomes a useful set at RPE 8 without adding any time to the session. Cut sets when a muscle is getting more than about 11 sets in one session, the point where a 2025 preprint found extra sets stopped adding a detectable gain [1]. Move those sets to another day rather than deleting them, if the weekly total is where you want it.