Autoregulation of Set Volume: Add or Cut Sets Based on Readiness and Velocity Loss

Autoregulation training volume is a simple idea: the number of sets you finish is decided partly by how the session is going, not only by the number in your spreadsheet. If your first sets are fast and crisp, you can earn an extra one. If reps fall off a cliff, you stop that exercise early and move on. Research supports a speed drop of roughly 20-25% within a set as a sensible stop point, and a similar drop in reps across sets is a reasonable gym translation. It does not show that autoregulation beats a fixed plan for strength.

This sits inside the bigger question of how many sets you need each week, which we cover in our training volume and frequency guide. Here the question is narrower. You walk in feeling wrecked, or feeling unstoppable. What do you do with the set count?

What does autoregulating set volume actually mean?

Plans are written weeks ahead by someone who cannot see how you slept, what you ate, or whether your boss ruined your Tuesday. Autoregulation hands a small part of that decision back to you, using a measurable signal.

Three signals do most of the work:

• Bar speed at a familiar warm-up load.
• Rep performance: how many reps you get at the same weight as last week, and how fast they fall across sets.
• RIR or RPE: how many reps you think you had left.

You still follow a plan. You just write it as a range, say 3-5 sets, and the session picks the number. The mistake I see most often is treating “autoregulation” as a polite word for “do less when it’s hard”. It is not. It works in both directions, and it needs a rule you set before you pick up the bar.

What is velocity loss and why does it set a stop point?

Every rep in a set gets a little slower as fatigue builds. Velocity loss is the percentage drop from your fastest rep to the current one. Researchers use it as a live fatigue meter, ending the set (or the exercise) once the drop hits a threshold such as 10%, 20% or 40%.

A 2023 systematic review and meta-analysis pooled 18 acute and 19 longitudinal studies on this (Jukic et al., 2023). It found a graded relationship: as the allowed velocity loss rises, reps per set, blood lactate and perceived exertion go up, while jump height, sprint times and speed against fixed submaximal loads tend to get worse. Exercise, load and training history modulate the size of those effects.

That graded relationship is the whole logic of volume autoregulation. A set that stops at 20% loss has used less fatigue budget than one that grinds to 40%.

One detail matters for this article. In the training trials below, the velocity loss rule capped the reps within each set, while the number of sets was fixed in advance. Using a performance drop across sets to decide how many sets you do is a practical extension of that idea, not something these trials tested directly.

What do the studies say about 20% versus 40%?

Two trials from the same Seville group are the best-known head-to-heads.

In the first, 22 young men trained the squat for 8 weeks, stopping each set at either 20% or 40% velocity loss (Pareja-Blanco et al., 2017). The 20% group did about 40% fewer repetitions yet gained similar squat strength and improved countermovement jump more (9.5% versus 3.5%). The 40% group did build more muscle in two of the quadriceps muscles measured.

In the second, 64 resistance-trained men were split into 0%, 10%, 20% and 40% groups for 8 weeks of squatting, with sets and relative load matched (Pareja-Blanco et al., 2020). Strength, jump and sprint gains did not differ between groups. The 20% and 40% groups showed significant increases in muscle size, while only the 40% group showed negative neuromuscular changes such as slower force development. The authors suggest moderate thresholds make sense when strength is the priority.

Both trials are small, short and in young men squatting. Treat them as a signal, not a law.

What does the bigger picture show?

Meta-analyses smooth out single-study noise, but they add their own complications.

The 2023 Jukic review concluded that the threshold did not affect strength or muscular endurance gains, that higher thresholds were superior for hypertrophy, and that lower thresholds were superior for jumping, sprinting and speed against submaximal loads.

A 2022 meta-analysis on load and volume autoregulation in resistance-trained people reached a more specific split. Thresholds at or below 25% gave slightly more 1RM strength (about 2.3 kg), while thresholds above 25% gave slightly more muscle cross-sectional area (about 0.6 cm2) (Hickmott et al., 2022). Above 25% did not beat 20-25% for hypertrophy, but it did beat 20% or less.

Then there is the proximity-to-failure meta-analysis. It found no significant hypertrophy difference between high (over 25%) and moderate (20-25%) velocity loss thresholds, and no advantage of momentary failure over stopping short of it (Refalo et al., 2023).

Source	What it found	Plain read
Jukic 2023 (37 studies)	Threshold did not change strength; higher thresholds favoured hypertrophy	More fatigue, slightly more growth
Hickmott 2022 (15 studies)	25% or less: a little more strength. Over 25%: a little more size	A trade-off, not a winner
Refalo 2023 (15 studies)	No significant gap between over 25% and 20-25%	The curve flattens

Why the apparent disagreement? Different studies lump thresholds differently, use different exercises, and measure muscle size with different tools. The cautious takeaway: stopping at 20-25% is a good default. If a muscle is a priority, you can let a few sets run a little further, but the extra growth on offer is small.

How to do it without a bar-speed sensor

Velocity trackers are accurate but nobody needs one to apply the idea. Two free substitutes carry most of the signal.

Rep drop at the same load. If your first work set at 60 kg is 10 reps and the third is 7-8, you have lost roughly 20-25% of your output. That is a stop sign for that exercise. This is an extrapolation from bar-speed studies, not a validated rule, so adjust by feel.

RIR as the gate. Recent work suggests lifters can estimate reps in reserve well. In 24 resistance-trained men and women on the bench press, 1 RIR and 3 RIR predictions had a mean absolute error of 0.65 reps, with a small tendency to underpredict (Refalo et al., 2024). Another study of trained and untrained men and women on machine biceps curls, triceps pushdowns and seated rows found accuracy improved closer to failure and in later sets, with no significant effect of sex or experience (Remmert et al., 2023). Use our RIR/RPE calculator to anchor loads.

The takeaway: you are not as bad at guessing as you think, especially near the end of a set.

A simple session-day decision rule

Here is a clean way to run it. Write each exercise as a range, for example 3-5 sets, then:

1. Warm up properly. Do not judge your day off a cold first rep.
2. Check set 1 against last week. Same load, similar reps, similar effort?
3. After each set, check the drop. If reps or speed are down about 20-25% from set 1, stop that exercise.
4. Add a set only when the set before met your target with reps to spare, and you are still under your weekly ceiling.
5. Cut a set when set 1 is clearly below last week even after warming up.

If you run a push, pull, legs split, the same rule works on any lift. Try it with our push day and see where your set counts land.

When “cutting sets short” is just being lazy

This is the part that gets skipped. Autoregulation is not a free pass.

Warm-up first. A rep that feels heavy at 7 am in a cold gym often feels normal after two ramp-up sets. Judge by the numbers, not by your mood walking in.

Then ask whether this is one flat day or a pattern. One flat day: trim a set or two, keep the session. Three or four in a row, plus poor sleep, low appetite or nagging joint pain: that is a signal to reduce volume across the week, which we cover in overtraining versus non-functional overreaching and deloads and volume resensitization.

If you feel unwell, dizzy, have chest pain or unusual shortness of breath, or fatigue that lasts for weeks, that is a doctor question, not a programming one.

What if you feel on fire?

The opposite problem. First set flies, your PRs are suddenly easy, and the temptation is to add six sets.

Add one, maybe two, on a muscle you are prioritising. Then stop. Weekly volume creeps, and what you can recover from tomorrow is not what you feel today. The per-session ceiling is covered in our guide to set volume per session, and the weekly range in volume landmarks.

Do women need different thresholds?

Possibly, but the data are thin. In an 8-week trial comparing 20% and 40% velocity loss in young men and women, everyone improved and there were no statistically significant differences between groups, though women showed larger effect sizes with 40% in strength and power measures (Rissanen et al., 2022). That is one short study. A reasonable approach is to treat 20-25% as the default and let women experiment toward the upper end if sets feel too easy.

Keeping it sustainable across a mesocycle

Autoregulation changes the daily set count, not the direction of the block. Aim for a weekly set range, start toward the low end, and let good days nudge you up. Weeks where your sessions keep coming in under target are information: either recovery needs attention (sleep, calories, protein) or your plan is too ambitious.

In an Indian context this often means the basics. A hot, humid gym, a long commute, a breakfast of two parathas and chai, then a 6 pm session on four hours of sleep. Speed and reps will tell you the truth long before a diary will. Check the plate before you blame the program, and for how much protein you are actually eating, use our protein calculator.

For a deeper look at rep-based progression rules, see autoregulated progressive overload (APRE), and for the failure question see training to failure versus RIR. To total up your week, use the training volume calculator.

The bottom line

Autoregulating volume is a rule, not a mood. Warm up, compare set 1 with last week, stop an exercise when output has dropped about 20-25%, add a set only when you have earned it, and keep a weekly ceiling and floor. The evidence says the 20-25% zone is a sensible compromise between strength and size, and that autoregulation matches fixed plans rather than beating them. That is enough to make it worth using, and honest enough to not oversell it.