Loading fundamentals 08: Warm-up

Short answer. Ascending warm-up sets cost repetitions on isolation exercises in one trial and made no difference on bench and leg press in another. Our position is that the specific warm-up is a tool for getting familiar with the load: several ascending sets before a heavy compound lift, one or none before isolation work. No trial has tested whether warm-ups affect muscle growth.

---

What it is

What you do before the working sets. The standard prescription has three parts: raise body temperature with light general activity, mobilize with dynamic movement, then perform ascending-load sets of the exercise itself.

The third part is usually called a specific warm-up, and it is the part with the most evidence and the most confusion attached to it.

The proposed mechanism

Four accounts, and they are not equally supported.

Temperature. Warmer muscle has faster contractile kinetics and lower viscous resistance, so force production and rate of force development improve.

Neural priming. Rehearsing the movement pattern at submaximal load is proposed to prepare the motor units and coordination required for the heavier set.

Load familiarization. Ascending sets let you feel the bar and calibrate effort before the set that matters. This is the least physiological account and, we will argue, the most defensible one.

Injury prevention. The most commonly cited reason and the one with the least direct support for resistance training specifically.

What has actually been tested

Specific warm-up made things worse in the one trial that found an effect at all. Fifty-seven resistance-trained older women completed a crossover comparing a specific warm-up, static stretching, and a control condition before four exercises taken to concentric failure. Static stretching improved total repetitions on the two lower-limb exercises compared with both the specific warm-up and the control, while the specific warm-up impaired performance on the two upper-limb exercises, triceps pushdown and preacher curl (Lisboa et al., 2025). All four were isolation movements.

And the trial that varied the warm-up itself found nothing. Twenty-nine resistance-trained participants performed one warm-up set, two ascending warm-up sets, or no warm-up at all, before multiple sets to concentric failure on bench press and leg press. Varying set number or load neither appreciably enhanced nor impaired session performance (Enes et al., 2025).

Neither result supports the standard prescription, and they fail it differently. One found the ramp costs repetitions. The other found it buys none.

High-load dynamic warm-ups do have support, at least for the upper body. A systematic review of thirty-one randomized trials of upper-body warm-ups, covering twenty-five warm-up modes, concluded that high-load dynamic warm-ups enhance power and strength performance. It found that short-duration static stretching has no effect on power outcomes, and that passive heating or cooling is largely ineffective as a warm-up mode (McCrary et al., 2015). The same review found no studies examining upper-body warm-up effects on injury prevention.

One randomized trial did find a warm-up changing a post-exercise outcome. Participants performing loaded front lunges were randomized to warm-up, cool-down or control. Pressure pain threshold at the central muscle belly fell in the control group over the following two days and did not fall in the warm-up group, while loss of maximal isometric force was equivalent across all three (Olsen et al., 2012). The effect was regional and did not extend to force.

And nobody has shown a warm-up affects muscle growth. No controlled trial has randomized warm-up protocol and measured hypertrophy. The question has not been asked.

Our reading

The contradiction resolves once you notice which exercises produced which result.

The trial where specific warm-up impaired performance used only isolation exercises. The trial where it made no difference used bench press and leg press. The review where high-load preparation helped measured upper-body power and strength performance.

That difference may be most of it. On an isolation exercise at moderate load, a specific warm-up is simply additional fatigue with nothing to prepare. The movement is single-joint, the pattern is trivial, the load is not near maximal, and the first few repetitions of the working set do whatever priming is needed. Adding sets beforehand spends capacity and buys nothing.

On a heavy compound lift the calculus inverts, because the thing being prepared is not the muscle. It is the coordination of a multi-joint pattern under a load close to your limit, and your calibration of what that load feels like today. (→ Loading fundamentals 02: Reps in reserve) Walking up to a near-maximal squat cold is a technical and a judgment problem before it is a physiological one.

So our position is that the specific warm-up is a load-familiarization tool, not a tissue-preparation tool, and it should be scaled to how much familiarization the movement and load require. Several ascending sets before a heavy squat. One light set, or none, before a cable curl. The prescription that applies the same ramp to every exercise is spending fatigue on movements that do not need it.

The ramp is also not free: those sets draw on the same session capacity as the working sets. (→ Loading fundamentals 01: Rest between sets; Loading fundamentals 07: Weekly volume)

One practice sits outside this argument entirely. A set performed with the intent to accelerate maximally is not preparing tissue and is not familiarizing you with load. It is doing something else.

In trained men performing squats at matched load, volume and rest, trying to lift maximally fast produced significantly greater muscle activation, force and power than deliberately lifting at half speed (Cornejo-Daza et al., 2024). Whether that acute difference translates into anything over a training block is a separate question.

This method belongs to velocity-based training and the dynamic effort approach, where it is the point of the session rather than a preliminary. (→ Loading fundamentals 09: Movement velocity and intent) But nothing stops it being borrowed. A single maximal-intent set placed in the warm-up of a session built around a different goal is an application of that method, not a warm-up in the sense the rest of this article uses. It is how OMNID's Powerbuilding programs open their main compound lifts: an explosive primer of two fast repetitions at 60% of the working weight.

We are describing a design choice, not reporting a finding. No trial has tested a priming set of this kind against its absence.

On the general warm-up, the evidence is thinner than the confidence around it. Raising temperature is real physiology. That it needs a dedicated ten-minute protocol before lifting is not something we could find evidence for, and passive heating specifically was rated ineffective.

On stretching, we would stop worrying about it. The differences reported are small, the directions are inconsistent, and one of the trials here found static stretching outperforming the alternative that is usually recommended over it. That is not a case for static stretching. It is a case for the question being less important than the argument about it suggests.

Where the goals diverge

Training heavy compound lifts. Ascending sets earn their place, and the heavier the top set the more of them are worth doing.

Isolation and machine work. The evidence here suggests the ramp is a cost. The first repetitions of the working set do the job.

Training for power or sport. High-load dynamic preparation has the clearest support here, and it is a different objective from preparing to lift. (→ Loading fundamentals sub-hub)

Older trainees or people with joint restrictions. The one well-powered trial here was in older women and found the standard prescription counterproductive for repetition performance. That is not a reason to skip preparation, but it is a reason to stop assuming the standard protocol is protective.

The case against

Repetition count is not the only outcome that matters. These trials measured performance within a session. A warm-up that costs repetitions today could still be worth doing for reasons the studies did not measure.

Our reconciliation is an interpretation. Nobody has run the trial that crosses warm-up protocol with exercise type and load. We noticed a pattern across studies designed for other purposes.

The impairment finding is a single trial in one population. Fifty-seven older women, four isolation exercises, one crossover.

And the two trials differ in more than exercise type. Lisboa's participants were resistance-trained older women; the null trial used younger trained adults. Population moved together with exercise selection, so the pattern we are reading off them could be either variable, or neither.

Our injury claim needs a boundary around it. The review found no studies of upper-body warm-up effects on injury prevention. That is not the same as saying warm-ups do not prevent injury: structured neuromuscular warm-up programs in field sports, of which the FIFA 11+ is best known, have randomized evidence behind them. Different intervention, different context.

What is not known

Whether warm-up affects anything over a training block, in either direction.

Whether the ascending-set benefit on compounds is physiological, technical, or psychological. Our account leans on the last two and cannot separate them.

Whether injury risk is affected at all in resistance training specifically. The review we cite found no study of it for upper-body warm-ups, which is different from having been studied and found null.

How much general warm-up, if any, adds once ascending sets are in place.

The practical warning

A joint that feels wrong during preparation is giving you information about the session, not an obstacle to warm through. The ascending sets are the cheapest place in a session to find that out.