Training in the stretched position: what the evidence shows

The advice doing the rounds in gyms right now is that muscle grows fastest when you load it deep in the stretched position, and that partial reps in the bottom half therefore beat full reps. Plenty of lifters have already binned the top half of every set. Does the evidence actually back that?

This is one of the few training questions where a genuine pile of research has appeared in a short space of time — and one where the conclusions ran well ahead of the data. Here is what is solid, what is merely likely, and what gets mangled in the retelling.

What is reliably established

Full range of motion is the safe default. When training through the whole range is compared with training through half of it, the full range wins or ties far more often than it loses. The evidence here is consistent and has not shifted for years.

If you are doing half a rep, the bottom half is the better half. Comparisons of partials at long muscle lengths against partials at short lengths point the same way almost every time: the stretched portion produces more growth. The gap is not cosmetic, and it has shown up across several different muscle groups.

Exercise choice changes the outcome even when set counts match. Two movements for the same muscle do not load it at the same length. Variations that keep the muscle long precisely where the load is heaviest — overhead triceps extensions, curls with the shoulder behind the body, Romanian deadlifts, deep squats, seated leg curls — tend on average to build more than versions that peak at the shortened end.

Muscles do not grow evenly along their length. The same exercise often produces more thickening in one region than another, which is why two exercises for one muscle complement each other rather than cancel out.

What is likely, but not settled

How big the advantage really is. Most comparisons run 6 to 12 weeks, use 10 to 30 people per group, and measure muscle thickness by ultrasound at one or two sites. That design shows direction well and magnitude poorly. A few percent over a few months is a fairer expectation than anything resembling double the growth.

Whether lengthened partials beat full reps. Results here are mixed — sometimes equal, sometimes a mild edge to the partials. The honest summary is that they are a legitimate option, not a replacement.

Why it works at all. Explanations range from higher passive tension in a stretched muscle, to a different growth signal, to adding sarcomeres in series. None is resolved, and you do not need the answer to make a decision in the gym.

Carryover to strength. Strength is fairly specific to the range you train. If you want to be strong through a full rep, train full reps. That part is well supported; how long any range-specific advantage lasts once you stop is much less studied.

Whether the edge compounds over years. There are no long trials in experienced lifters. The question sits inside the broader argument about where the ceiling on muscle growth actually is.

Loaded stretching between sets and very long holds in the bottom position. This is poorly researched in humans so far, with small samples and protocols most people would not tolerate. It is not a foundation to build a programme on.

What gets reported wrongly

That full range training is obsolete. It is not. The comparisons said to have buried it mostly had it drawing level with lengthened partials, not losing to them.

That stretching by itself builds muscle. Ordinary unloaded static stretching improves flexibility, but as a growth stimulus in humans the effect is negligible. It is the load applied at long lengths that matters, not the stretch on its own.

That deeper is always better. The benefit only holds while the joint and tendon tolerate the position. Depth forced at the cost of technique adds risk without adding growth.

That more soreness confirms more growth. Deep positions genuinely hurt more the first time, then settle after two or three sessions thanks to the repeated bout effect. Soreness was never a progress metric.

That everyone responds the same. The spread in individual response to the same training is wide, and it applies here as much as anywhere else.

What this means for your training

The basics do not change: 10 to 20 hard sets per muscle group per week, reps anywhere from 5 to 30, and sets taken close enough to failure, usually zero to three reps in reserve. Training at long lengths is a way of doing that same work more intelligently, not an excuse to do less of it.

• Full range is the default. Shorten the range only when you have a specific reason.
• Pick at least one exercise per muscle by length. Overhead work for triceps, incline curls for biceps, Romanian deadlifts or seated curls for hamstrings, flyes or a deep press for chest, full depth for calves.
• Lower over 2 to 3 seconds and stop bouncing out of the bottom. A half-second pause at the deepest point is enough; longer holds have no demonstrated extra benefit.
• Use partials as a top-up, not a programme. When full reps run out, add 3 to 6 reps in the bottom half, on isolation work, once or twice a week per muscle.
• Phase it in. Drop the load by 10 to 20 per cent for the first two or three weeks when you switch to a deeper version of a lift.

On squats and deadlifts, depth depends on what your knees and back tolerate that day. If you get knee pain at the bottom of a squat, depth is the last variable you should be pushing.

Nutrition needs no special adjustment. Protein at 1.6 to 2.2 g per kilogram of body weight, with calories at or slightly above maintenance, covers everything this style of training asks for. No supplement specifically amplifies growth from the stretched position.

When to see a doctor

Get checked if you feel a sharp, stabbing pain deep in a movement that forces you to stop the set, if pain is still there outside the gym after two or three weeks, if there is swelling, numbness or pins and needles, or if you felt a sudden pop followed by a loss of strength. A joint whose range stays reduced and does not come back also deserves an assessment rather than another week of deep reps.

In short

• Full range of motion remains the safest default, and the evidence there is consistent.
• If you use partials, use them in the stretched bottom half — that is clearly better than the top.
• Choosing exercises that load the muscle while it is long is the most practical takeaway.
• The size of the advantage is probably modest, measured in percentages rather than multiples.
• Depth your joints cannot handle is not an advantage, it is a risk.