Strength Science
Why Strength Transfers Poorly Between Similar Movements
Getting stronger at one exercise improves performance in another far less than expected, because much of what improves is specific to the position, speed and pattern practised.

A lifter who doubles their leg press rarely doubles their squat. The gap between related movements is larger than intuition suggests, and its causes are reasonably well described.
Adaptation is specific to the pattern practised
Strength gains include a substantial coordination component, and coordination is learned for a particular sequence of joint actions under particular conditions.
Change the stability requirements, the joint angles or the timing, and part of that learned component no longer applies. The muscle is the same; the task is not.
This is why the transfer is asymmetric. A free-weight movement usually transfers better to a machine version than the reverse, because the harder task contains the easier one.
Force capacity varies with joint angle
A muscle's ability to produce force depends on its length, and each exercise loads the muscle most heavily at a specific point in its range.
Training that emphasises one portion of the range produces the greatest gains near that portion. Positions outside it improve less.
Two exercises that look similar can therefore have their hardest points at different joint angles, and the strength built at one point does not fully carry to the other.
Stabilisation demands differ
A supported movement removes the requirement to control the load in space. The prime movers may work equally hard while the surrounding demand disappears entirely.
When the lifter returns to the free version, that stabilising requirement reappears and limits how much force can be expressed, regardless of what the prime movers can produce.
This is the usual explanation for a lifter whose machine numbers climb steadily while their barbell equivalent moves very little.
Velocity is its own specificity
Training at slow speeds against heavy loads produces different adaptations to training at high speeds against light ones, and the transfer between them is limited.
Some of this reflects how rapidly the nervous system can drive the muscle, which improves most at the speeds actually practised.
Anyone training for a task with a strong speed component therefore needs work at that speed, since heavy slow work alone leaves part of the requirement untouched.
What this implies about exercise selection
The lift you want to improve has to be trained, not approximated. Accessories build capacity but do not substitute for practice of the target movement.
The useful role of variations is to load a position or a weakness the main lift underloads, which is a supplement to the main lift rather than a replacement.
Judging an accessory by how much it transfers, rather than by how much can be lifted on it, changes which exercises look worthwhile.
Also by Hiro Tanabe
- What we still do not knowStrength Science
- Making weight without wrecking your performanceFuelling
- Who to trust: evaluating fitness information sourcesStrength Science
- How long a muscle actually needs between sessionsRecovery





