Strength training

What this subject is

Skeletal muscle adapts to being made to produce force against resistance it cannot easily overcome. Mechanical tension sensed by the muscle fibre is the primary stimulus for growth; the nervous system's improving ability to recruit and drive motor units is the primary source of early strength gain; and the adaptation is specific to what was trained — the movement, the range, the speed and the load. Everything else in the subject is programming: how heavy, how many hard sets, how often, how close to failure, how long between sets, which exercises, and through what range. Those variables have been tested extensively, and the results are mostly less dramatic than the arguments about them. On top of the training sits the recovery interface — total daily protein, energy availability, sleep — where the same pattern holds: a few variables carry most of the effect and a large industry sells the rest. The package teaches the mechanisms, the programming evidence with its real uncertainties, the health case across the lifespan stated as the associations it is, and the discipline of distinguishing a tested claim from an inherited one.

What the package holds

Curated scaffolding your agent loads before it researches, so it starts from vetted ground rather than a cold search.

• Tier 1: 24
• Tier 2: 36
• Tier 3: 5

The questions and claims below are quoted from the package files.

Where the field disagrees

Each one carries real proponents on more than one side, so your agent cannot quietly pick a winner.

• Does hypertrophy keep increasing with weekly hard sets, and if so, up to what point?

  0 named positions · Strong evidence for a positive volume-hypertrophy relationship across the commonly trained range; genuinely unresolved at the top of that range and genuinely disputed as to whether volume or effort is the dominant term. The defensible teaching is that weekly hard sets is the best available dial, that more is better within limits set by recovery, and that anyone quoting a specific optimal number is reporting a preference as a finding.

• Is the muscle growth produced by training a cause of the strength gained from it?

  0 named positions · Contested hypothesis, and a model of how disagreement should look. Both papers are argued positions rather than pooled evidence, both were published in the same peer-reviewed venue, and neither has been settled by subsequent work. What is not in dispute: strength and size are dissociable, early strength gain is largely neural, and architectural changes break any simple size-to-force mapping (folland-williams-2007). Teach the pair together and resist the urge to award the point.

• Does planned variation of training variables beat simply progressing a sensible programme?

  0 named positions · Reasonable interpretation. Periodisation is a useful organising device with a modest average advantage for maximal strength, most of which may be systematic progression wearing a more impressive name. The honest finding — and the one most often omitted when a model is being sold — is that periodisation models differ from each other far less than their advocates claim.

• How hard does a set have to be?

  0 named positions · Reasonable interpretation on both halves. Sets have to be reasonably hard and do not have to end in failure; where exactly the stimulus starts to fall off is not known, and specific repetitions-in-reserve prescriptions are practical conventions rather than established thresholds. Note also that estimating proximity to failure is itself a trained skill that inexperienced lifters perform poorly, which is a real limit on any prescription written in those units.

• Is full range of motion always best, and is there something special about training at long muscle lengths?

  0 named positions · Strong evidence for full range as the sensible default; the lengthened- position question is genuinely open and moving. This is the part of the package most likely to be out of date first, and an agent using it should search for work published after the package version rather than treating the default as final.

Myths the package corrects

Widely held claims with the evidence that settles or bounds them.

• “You have to lift heavy — low repetitions with high loads — to build muscle. Higher-repetition work with lighter loads is for endurance or toning and does not grow anything.”

  debunked-as-stated

  The intuition mistakes load for tension. Motor units are recruited in ascending order of size, so a heavy load recruits high-threshold units immediately while a light load recruits them too — as the smaller units fatigue and the nervous system calls on its reserves. That is why a set taken close to failure at roughly thirty percent of a one-repetition maximum produced hypertrophy comparable to eighty percent over ten weeks (mitchell-2012), why the same held in already-trained men over twelve weeks (morton-2016), and why the pooled analysis across loads above and below roughly sixty percent found similar hypertrophy between conditions (schoenfeld-load-2017). The claim is not simply false, though: the same pooled analysis found clearly greater gains in maximal strength with heavier loads, the advantage largest on a one-repetition maximum test. The accurate sentence is that muscle size is relatively tolerant of load while maximal strength is load-specific, and that what growth actually tracks is proximity to failure and accumulated hard sets.

• “You must constantly surprise the muscle. Change the exercises often, never let the body adapt, and keep it guessing — otherwise progress stops.”

  debunked

  Adaptation slowing is an argument for increasing the demand, not for changing it. Progressive overload is the variable that stalls when a programme stalls, and novelty does nothing to it: a randomly varied exercise selection in trained men, with volume equated against a fixed selection, produced no advantage in muscle thickness or maximal strength, and the varied group reported lower motivation (baz-valle-2019). Constant variation also actively costs something. Strength is partly a skill in the trained movement (rutherford-jones-1986), so changing exercises resets part of the learning; and the repeated-bout effect means unfamiliar movements produce soreness that is often mistaken for evidence that the change worked (cheung-2003). Consistency plus progression is the mechanism. Muscles do not have expectations to violate.

• “Protein must be consumed within about thirty minutes of finishing a session or the training is wasted.”

  boundary-correction

  The window exists and it is not thirty minutes. The review that reopened the question found the supporting evidence far weaker than the prescription: a pre-exercise meal is still being digested and absorbed during and after training, and the elevation in muscle protein synthesis after a resistance session persists for roughly a day, so the practical window is measured in hours (aragon-schoenfeld-2013). The pooled test is more decisive still: an apparent benefit of protein timing did not survive statistical control for total daily protein intake, because studies with timed protein also had higher totals, and the total accounted for the effect (schoenfeld-timing-2013). Total daily intake is the variable that carries the evidence, with the pooled benefit plateauing around 1.6 grams per kilogram of body mass per day (morton-2018).

• “Women who lift weights, especially heavy ones, will become bulky or masculine-looking.”

  debunked

  The evidence points the other way on both halves. Pooling training studies that included both sexes, relative gains in hypertrophy were similar between men and women and relative upper-body strength gains tended to favour women, while absolute changes were larger in men because they start from a larger base (roberts-sex-2020) — a finding already reported in direct comparison more than thirty years earlier (cureton-1988). Similar relative growth from a much smaller starting muscle mass, at typically lower total body mass, produces a visibly different outcome by arithmetic rather than by any different physiology. Circulating testosterone differs substantially between sexes and is part of the explanation for the different starting point, but it is worth noting that the acute post-exercise hormone story used to explain growth does not hold up in either sex: training in a high-hormone or low-hormone systemic environment produced no difference in hypertrophy or strength (west-2010), and post-exercise testosterone and growth hormone did not predict individual gains (morton-2016). The bulk that is feared takes years of deliberate work that nobody arrives at accidentally.

• “If you are not sore the next day, the session did not work. Soreness is how you know the muscle was stimulated.”

  debunked-as-stated

  Delayed-onset muscle soreness does not track the thing it is believed to measure. Soreness ratings did not correspond to markers of muscle damage or to the loss and recovery of force after eccentric exercise (nosaka-2002), and soreness after an unfamiliar movement falls sharply on the second exposure through the repeated-bout effect, so it indexes novelty at least as much as stimulus (cheung-2003). Damage itself is not the mechanism either: early in a programme, elevated myofibrillar protein synthesis reflects repair and does not predict later hypertrophy, and only once damage subsides does the synthetic response correlate with growth (damas-2016). Mechanical tension accumulated over hard sets is the stimulus; soreness is a sensation produced most reliably by doing something new. A well-run programme in a trained lifter frequently produces little soreness and considerable growth.

Related topics

• Creatine
• Sleep science
• Habit formation
• Intermittent fasting