Dossier · maximised

The Hypertrophy Protocol

The deliberate opposite of the minimalist dossier. This asks what the ceiling looks like and accepts every cost the answer carries: the hours, the food volume, the social friction, and the aerobic fitness you surrender by keeping interference low. Everything here is built on one requirement — a fibre grows when it is fully recruited and under high mechanical tension at the same instant — and every variable below is downstream of arranging that as often as recovery allows. The trade-offs are stated rather than hidden, because a maximal programme that pretends it is free is how people end up under-recovered and confused.

Research vintage · Q3 2026

The one requirement

What you are actually manipulating

RCT / meta-analysisResearcher, own field

Recruitment × tension, then recovery

Mechanism

The two conditions, and why both are non-negotiable

Motor units are recruited in ascending order of size (Henneman). Full recruitment arrives around 90% of maximal unfatigued isometric force — roughly a 5RM — but effort substitutes for load: take a lighter set close enough to failure and fatigue forces the nervous system to call up the same high-threshold units. That is condition one. Condition two is per-fibre mechanical tension, governed by the force-velocity relationship: during fast shortening myosin heads detach before completing their power stroke, so few crossbridges are bound and per-fibre force is low; during slow shortening they accumulate. Only reps satisfying both count — roughly the last five before failure. Everything else is a fatigue tax paid to arrive there.

Mechanism

How tension becomes protein

Load deforms integrin complexes at the costamere, recruiting focal adhesion kinase. The route onward to mTORC1 is genuinely not closed — the defining finding is that mechanical mTORC1 activation is *insensitive to PI3K/Akt inhibition*, so the growth-factor pathway is not the carrier, and the FAK→TSC2→Rheb chain is a leading proposal rather than settled. In parallel, mechanical stimulation activates diacylglycerol kinase-ζ, generating phosphatidic acid, which binds and activates mTOR directly. Downstream, mTORC1 phosphorylates S6K1 and 4E-BP1, raising translational efficiency, while satellite cells donate myonuclei to support a larger cytoplasmic domain.

So the entire programme reduces to one optimisation: maximise the number of stimulating reps delivered in a recovered state per unit time, without accumulating fatigue faster than you can clear it. Volume, frequency, exercise selection and nutrition are all just levers on that one quantity.

The dose

Volume, frequency and the landmarks

RCT / meta-analysisResearcher, own field

How much, and how it scales

Mechanism

The per-session ceiling, and why weekly volume is meaningless without frequency

Beardsley's reading of the volume literature puts the useful ceiling at roughly 5 hard sets per muscle per session — about 25 stimulating reps. Between 5 and 10 the curve plateaus without obvious harm; past 10 it turns negative, plausibly because muscle damage begins consuming the repair resources growth needs. Two fatigue types set that ceiling: central fatigue accumulating across the session (markedly worse with short rests — which is why trials using three-minute rests find higher optimal volumes than those using ninety seconds), and mechanical damage limiting between-session recovery. It follows that there is no maximum weekly volume stated independently of frequency: the weekly ceiling is the per-session ceiling times how often you train.

Frequency per muscle	Sets per session	Weekly working sets	Use when
2×	4–6	8–12	Recovery is limited — poor sleep, heavy deficit, high life stress
3×	4–5	12–15	The default for maximisation. Best ratio of stimulus to accumulated fatigue
4×	3–4	12–16	Specialisation blocks for a lagging muscle, or small muscles that recover fast
1×	5–8	5–8	Maintenance only. Roughly a third of the volume that built a muscle will hold it

Sets are counted only where the target muscle is the limiting factor. The squat is a poor rectus femoris stimulus — that head is also a hip flexor, shortened at the hip throughout. Bench press loads triceps and pectorals roughly equally, so it counts for both.

Proximity to failure: 0–2 reps in reserve on every working set. The last set of each exercise goes to genuine failure. RIR is systematically over-estimated by trainees, so calibrate periodically by actually failing.
Rest: 2–3 minutes on compounds, 90s on isolations. Short rests do not save time — they cost stimulating reps in the following set by carrying fatigue into it.
Rep range: 5–12 for most work, 12–20 for isolations where joint stress at heavy loads is the limiter. Load barely matters between 5RM and 30RM provided proximity to failure is matched — it matters for joints, for time, and for how much fatigue each set costs.
Progression: add a rep, then add load when the top of the range is hit for all sets. Log every set. Unlogged training is not progressive overload, it is exercise.
Deload every 6–10 weeks, or when performance drops for two consecutive sessions on the same lift. Halve the sets, keep the load. You do not detrain in a week — you clear fatigue and express the fitness that was already there.

The movements

Exercise selection by resistance profile

Not 'best exercises' — exercises whose hardest point matches the muscle's most trainable length.

RCT / meta-analysisPrimary literature

Why the stretched position is the free variable

Mechanism

Long muscle lengths, titin, and sarcomerogenesis

At long muscle lengths, sarcomeres sit past optimal actin-myosin overlap, so the contractile element produces less active force and the deficit is carried by passive tension in titin. Titin is not merely a spring: strain-dependent unfolding of its immunoglobulin domains exposes binding sites for signalling proteins, making it a genuine mechanosensor. Long-length training is also *proposed* to add sarcomeres in series, lengthening fascicles — but state that carefully: Wolf et al. (2025) note no study has yet directly estimated serial sarcomere number from long- versus short-length training, and fascicle-length estimation methods so far have questionable validity. The growth advantage is supported; this particular explanation for it is not yet measured. Meta-analytically, lengthened partials match or beat full range of motion for growth, and with one exception every study comparing muscle lengths found longer produced more.

Mechanism

Resistance profile — where an exercise is actually hard

Every exercise has a resistance profile determined by how the external moment arm changes through range, and every muscle has an active length-tension relationship. Where those overlap decides which region of the muscle is loaded hardest — the mechanistic basis of regional hypertrophy. A preacher curl is hardest near full elbow extension; a cable curl is roughly constant; a concentration curl is hardest mid-range. This is why exercises are not interchangeable at matched effort, and why 'pick anything and take it to failure' is incomplete.

The selection rule: for each muscle, pick one movement that loads it hard in the stretched position and one that loads a different part of the curve. Two exercises per muscle is enough. More is variety for its own sake, and variety costs you the ability to track progression.

Operator, verifiable record

The movement list

Muscle	Primary — stretch-loaded	Secondary	Why this pairing
Quadriceps	Deep squat or hack squat, full depth	Leg extension (lean back to lengthen rectus femoris)	Squats load vasti hard in the stretch; rectus femoris is shortened at the hip and needs the extension to be trained at all
Hamstrings	Romanian deadlift or seated leg curl	Lying leg curl	Seated curl trains the hamstrings at a longer length (hip flexed) — measurably more growth than lying. RDL covers the hip-extension function
Glutes	Deep squat / split squat	Hip thrust	Squat loads the stretch; hip thrust loads peak contraction with the most horizontally-oriented resistance available
Chest	Deep dumbbell press or machine fly with a full stretch	Incline press	Dumbbells and cables allow a genuine stretch the barbell's fixed bar path prevents
Back — lats	Wide-grip pulldown with full overhead stretch	Chest-supported row	Lats are longest overhead; the row trains the same muscle in a shorter, thicker position
Back — upper	Chest-supported row	Face pull / reverse fly	Chest support removes lower-back fatigue as the limiter, which is what caps most rowing volume
Delts — side	Cable lateral raise (behind the body)	Dumbbell lateral raise	The cable keeps tension at the bottom, where a dumbbell has almost none — the stretched position for the medial delt
Delts — rear	Reverse pec-deck	Cable reverse fly	Both keep constant tension; the rear delt has almost no useful stretch position, so tension continuity matters more
Triceps	Overhead cable or dumbbell extension	Pushdown	The long head crosses the shoulder — it is only genuinely stretched overhead. Pushdowns cannot train it in that position
Biceps	Incline dumbbell curl	Preacher or cable curl	Shoulder extension on the incline lengthens the long head. The preacher hits the other end of the curve
Calves	Standing raise, full stretch, pause at bottom	Seated raise	Standing trains gastrocnemius (crosses the knee); seated isolates soleus with the knee bent
Abs	Weighted cable crunch	Hanging leg raise	Abs respond to load like any muscle — bodyweight-only work stops progressing quickly

Machines and cables are not inferior to free weights for hypertrophy. They are often superior, because they let you load the stretched position without the stability limitation deciding the set. Free weights win on transfer and on loading the whole body at once — which matters for strength, not for growth per unit of fatigue.

The week

The programme

Operator, verifiable record

Upper / lower, four days, each muscle 2× — scaling to 3×

Four sessions of 60–75 minutes. Upper/lower alternating, which is the structure that best reconciles the per-session ceiling with the frequency the mechanism rewards. Six days a week (PPL×2) is the higher-frequency version and only worth it if sleep and food are genuinely handled — otherwise it converts stimulus into accumulated fatigue.

Day	Session	Content
Mon	Upper A	Incline press 3×6–8 · Chest-supported row 3×8–10 · Wide pulldown 3×8–10 · Cable lateral 3×12–15 · Overhead triceps 3×8–12 · Incline curl 2×8–12
Tue	Lower A	Squat 3×5–8 · Romanian deadlift 3×6–10 · Leg extension 3×12–15 · Seated leg curl 3×10–12 · Standing calf 4×8–12
Wed	Off	Walk. Mobility. Eat.
Thu	Upper B	Deep DB press 3×8–10 · Weighted pull-up 3×6–8 · Machine fly 3×10–12 · Reverse pec-deck 3×12–15 · Pushdown 3×10–12 · Preacher curl 2×10–12
Fri	Lower B	Hack squat 3×8–10 · Hip thrust 3×8–10 · Seated leg curl 3×10–12 · Leg extension 3×15–20 · Seated calf 4×10–15
Sat / Sun	Off, or Upper C	Add a third upper session if recovery allows — arms and delts recover fastest and respond best to added frequency

The cost, stated honestly

Running this properly means keeping conditioning to a maintenance dose — one hard interval session weekly and walking, nothing more. Endurance volume activates AMPK, which phosphorylates TSC2 and Raptor and directly inhibits mTORC1. You are choosing to give up aerobic ceiling for anabolic signal. That is a real cost with real mortality implications, and it is the honest price of a maximal hypertrophy block. Do not run this permanently.

Fuel

Nutrition for maximum growth

RCT / meta-analysisPrimary literature

The four variables that matter, in order

Mechanism

Protein and the leucine threshold

Leucine binds Sestrin2, releasing its inhibition of GATOR2, which permits the Rag GTPase heterodimer to recruit mTORC1 to the lysosomal surface where Rheb activates it (Wolfson et al., Science 2016). This is a threshold, not a gradient: roughly 2.5–3 g of leucine per feeding — about 30–45 g of quality protein — fully triggers the response, and more per sitting adds little. So distribution matters as much as total.

Variable	Target	Reasoning
Protein	1.6–2.2 g/kg bodyweight. Going to 2.4–3.0 g/kg only in a deficit	Meta-analytic plateau sits near 1.6; the higher end buys insurance and satiety. In a deficit, higher protein protects lean mass
Distribution	4–5 feedings of 0.4 g/kg, every 3–4 hours	Each feeding is a separate mTORC1 activation event. Two huge meals leave most of the day sub-threshold
Energy	Maintenance +10–15% to gain; −20–25% to cut	Rate matters more than the number — see below
Carbohydrate	4–6 g/kg on training days	Glycogen availability determines how many hard sets you can produce before performance falls, and glycogen-depleted training measurably reduces volume tolerance
Fat	0.8–1.0 g/kg, remainder of calories	Below ~0.5 g/kg starts affecting hormone production; above that it is just an energy source competing with carbohydrate
Peri-workout	30–40 g protein + 40–60 g carbohydrate within a few hours either side	The 'anabolic window' as a 30-minute panic is dead. What survives is that total daily intake and training-day carbohydrate matter, and eating near the session is a convenient way to hit both

What does not matter as much as the internet suggests

Meal timing beyond rough distribution, protein source beyond leucine content and digestibility, fasted versus fed training, and nutrient partitioning tricks. These are real effects with small magnitudes, and chasing them is how people end up with an elaborate diet they cannot sustain and a protein target they still miss.

RCT / meta-analysisPrimary literature

Supplements that survive scrutiny for this goal

What	Dose	Effect size, honestly
Creatine monohydrate	5 g daily	The only supplement with a large, consistently replicated effect on training performance and lean mass. Everything else on this list is smaller
Caffeine	3–6 mg/kg, 45–60 min pre-session	Real acute increase in force output and reps to failure via adenosine receptor antagonism. Tolerance is receptor upregulation — cycle it, or it becomes baseline
Protein powder	As needed to hit the target	Not magic. It is food that is convenient. That is the whole claim, and it is sufficient
Vitamin D3	Titrated to a measured 30–50 ng/mL	Corrects a deficiency that is common and does impair force production. Not ergogenic if you are already replete

Not included, deliberately: beta-alanine (real but tiny and only for 60–240s efforts), HMB (effect sizes that failed to replicate outside one lab), test boosters (nothing), BCAAs (incomplete amino acid profile — they compete for transport without supplying what synthesis needs), and every 'anabolic' compound sold without a prescription.

The arithmetic

Bulking, cutting and lean gaining

The rate is the decision. The calorie number falls out of it.

RCT / meta-analysisPrimary literature

The lean surplus

Mechanism

Why a small surplus, and why the rate is the real variable

Muscle protein accretion has a hard biological ceiling — an experienced trainee gains on the order of 0.25–0.5% of bodyweight per month in lean mass, and no surplus makes that faster. Energy above what that accretion plus training expenditure requires becomes adipose, and the partitioning ratio worsens as body fat rises: higher adiposity means more inflammation, worse insulin sensitivity in muscle, and a larger share of surplus energy stored rather than used. So the surplus is sized to the ceiling, not to appetite.

Phase	Surplus	Rate of gain	Expected split
Novice (<1 yr)	+15–20% (~400–500 kcal)	0.5–1.0% bodyweight/month	Roughly 60:40 lean:fat
Intermediate	+10–15% (~250–400 kcal)	0.25–0.5%/month	Roughly 50:50
Advanced (3+ yr)	+5–10% (~150–250 kcal)	0.1–0.25%/month	Roughly 30:70 — this is why advanced lifters bulk slowly or not at all

For a 78 kg intermediate: +300 kcal, expecting ~0.3 kg/month. If the scale is moving faster than that, the extra is fat. Run the block until body fat reaches roughly 15–18%, then cut back to 10–12%. Start a bulk from over ~15% and the partitioning is already working against you.

Weigh daily, same conditions, and use the 7-day rolling average. Daily weight is mostly water, glycogen and gut content — reacting to it is reacting to noise.
Adjust every 2–3 weeks, not weekly. One data point is not a trend.
If the average is flat for two weeks, add ~150 kcal, preferentially as carbohydrate on training days.

RCT / meta-analysisPrimary literature

The cut

Mechanism

What actually protects lean mass in a deficit

Three things, in order of importance. Resistance training provides an mTORC1-activating mechanical stimulus that is largely independent of energy status — the single strongest lean-mass protector. Elevated protein (2.4–3.0 g/kg) supplies substrate when the body is otherwise catabolising to meet amino acid demand. And a moderate rather than aggressive deficit limits the adaptive response: expenditure falls disproportionately during restriction, driven mostly by reduced NEAT (largely subconscious — people fidget and walk measurably less), reduced T4→T3 conversion, and reduced sympathetic tone.

Approach	Deficit	Rate	When
Conservative	−15% (~350 kcal)	0.25–0.5% BW/week	The default. Best lean-mass retention, best training quality
Moderate	−20–25% (~500–600 kcal)	0.5–0.75%/week	When there is more fat to lose and a deadline
Aggressive	−30%+	1%+/week	Rarely justified. Lean-mass loss rises sharply and training performance collapses

Keep training load. Reduce volume by ~20–30% if recovery suffers; do not reduce intensity. Intensity is what tells the body the muscle is still needed.
Diet break of 1–2 weeks at maintenance every 8–12 weeks. It restores leptin and thyroid signalling, and reduces the adaptive expenditure drop.
Cardio is a tool for creating deficit, not the primary one — food is easier to control and does not add recovery cost. Add cardio when food gets uncomfortably low, not before.
Expect strength to hold for weeks and then dip. That is normal and reverses on refeed.

Limited human dataPrimary literature

Lean gaining — the small-surplus version

The compromise, and for most people past the novice stage the correct default: a surplus small enough that fat gain stays near-invisible, run for a long time. +100–200 kcal, roughly 0.1–0.2% bodyweight per month, held for six to twelve months.

The honest maths on why this usually wins: a 6-month aggressive bulk at +500 kcal might add 3 kg lean and 4 kg fat, followed by a 3-month cut that costs some of the lean. A 12-month lean gain at +150 kcal might add 4 kg lean and 1.5 kg fat, with no cut required. Same or better lean outcome, no time spent dieting, and body fat never leaves the range where partitioning is favourable. The aggressive version only wins if you are a true novice or coming back from a layoff.

Body recomposition — who it actually works for

Simultaneous fat loss and muscle gain is real and mechanistically coherent: mTORC1 activation via mechanical tension can proceed during an energy deficit as long as amino acids are available. But the magnitude is strongly moderated by training status — largest in novices, detrained returnees and the overweight, and modest in already-lean trained individuals, where phase-cycling outperforms it. Most content marketing recomposition as universally achievable is selling to the group it works worst for.

The feedback loop

Tracking

Primary literature

What to log, and the spreadsheet

Every working set: exercise, load, reps, RIR. Without this there is no progressive overload, only exercise.
Bodyweight daily, same conditions, read as a 7-day rolling average.
Waist at navel weekly — it separates lean gain from fat gain better than the scale alone.
Session-level: total hard sets per muscle, and whether the session was completed as written.
Sleep hours and a 1–5 recovery rating. Volume tolerance tracks these more tightly than anything else.

A downloadable tracker with all of this pre-built — training log, weekly bodyweight with rolling average, macro targets computed from bodyweight and phase, a bulk/cut calculator, and progression benchmarks — is at /protocol-tracker.xlsx. It is a spreadsheet, not an app, on purpose: it opens anywhere, it will still work in ten years, and nobody can take it away or start charging for it.

Research notes, not medical or financial advice. Every prescription or experimental item named here is named with its mechanism and its risk and without a dose, on purpose — several require physician supervision, several are unregulated, and several are inappropriate for a body that is still developing. Start from your own bloodwork and a doctor, never from someone else’s regimen.

Non invenitur. Fit.