Go slow to go fast. Win the long game.
Dance&Walks
MTWTFSS
Morning dance
Midday walk
Dinner dance
Evening walk
Session
Not started
0:00
Monday
▲Upper-body push · knee work
Must finish feeling capable, not crushed.
Alternating pairbench first, a handstand between each · 90 s gaps
Bench press elbow · light
Video＋
Bench opens and closes the pair: five bench sets with four handstands between them.
The 90 s gap is the rest. Gap, handstand, gap puts bench-to-bench at roughly 3½ min, which is at or above what bench needs; the handstand fills time you were standing around in rather than taking time away from the lift.
Watch the bar, not the clock: if working weight drops across sets, the gap was cut too far.
Warm-up sets go in before the first round, not inside the pair.
Grip width is the variable on the last two sets, not load. Drop the weight if the wide grip pinches the shoulder.
The timer names each set. Tap ‘Run the block’ below and it walks you through power grip set one, two, three, then wide grip, then narrow grip, with a handstand between each and a heads-up before both grip changes. Nothing to track yourself.
Handstand elbow · out
Video＋
Four sets now, up from three: the fourth is free, it sits in a gap that already existed.
Wrist prep first, every time. Non-negotiable at this volume.
Stack it: wrists, shoulders, hips, ankles in one line. A stacked handstand costs far less per second than an arched one.
Neck extended, gaze between the hands. Neck flexion impairs handstand performance — measured, and the cue you have been missing. No figure attaches to how much of your balance it accounts for.
40 s is time under work, not one unbroken hold. Enter at 25–30 s, come down before the line breaks, go back up — accumulate to 40 s. Never to failure; test a true max once a week or two.
Batchback to back · 2 min break at the end
Bent-over dumbbell lateral raise
Video＋
Light. This is a positioning exercise, not a loading one.
Chest supported if the lower back complains.
Single-leg standing hip flexion
Left side is the one that matters: psoas was cut and healed back together.
Slow and deliberate. No swinging the knee up with momentum.
Stand tall; don't lean back to lift higher.
Rotationround the three · 45 s rest between
Bicep curls, machine
Video＋
New, and on Monday deliberately: fresh arms, and it keeps bicep fatigue off Thursday where pull-up quality matters.
Machine rather than dumbbells: less stabiliser cost on a day already carrying handstands.
Hip thrust
Video＋
New, and one of the best exercises available to you: glute max runs off a nerve your injury didn't touch, so this is genuinely bilateral.
Ribs down, chin tucked. Finish with the hips, not the lower back.
Pause a beat at the top.
Dips elbow · out
Video＋
Third in the rotation, so the arms are already worked when you get here. Drop reps rather than control.
New. Start with bodyweight, controlled, no bouncing at the bottom.
Shoulders down and back: don't let them shrug up at the bottom.
Stop 2 reps short of failure while the pattern is new.
Backward drag knee
Video＋
What it is for. A way to move the limb under load without provoking the joint, and a warm-up for the knee before the squat and the leg press. Cheap and low-risk.
What it is not. Nobody has published a measurement of knee or kneecap loading during any sled or drag task, in either direction. The reduction seen in backward running comes from landing on the forefoot rather than from going backwards, and in backward walking the ankle does the work — the knee generates almost no propulsion. So do not count this as knee strength training, and do not assume the load is free.
It is still real load, and it has an adjustment. 50–60 kg. The ≤2/10 rule applies here as everywhere else, and because there is no depth to shorten, the only lever is weight: if the knee goes above 2 out of 10, drop 10 kg for that session and write it down. The drag is not a green-week progression variable — it does not compete with load or depth on the press.
How to do it. Short steps, stand tall, do not lean back.
Where it sits. It warms the knee up before the squat and the leg press.
Leg press knee
Video＋
This is the tracked set. First of three exposures, and the one whose working load is logged every week. It sits on Monday because Monday is fresh: Saturday’s press follows a squat, so its load reads squat fatigue as much as knee capacity. Thursday and Saturday are volume around this, not tests.
This is the main knee exercise. Use the full range you have, including the deep part. Going deep is the point of it, not a risk being tolerated.
Why deep is safe here and not on a leg-extension machine. In a leg press the pressure behind the kneecap is lower in the deep positions; on an extension machine it is higher. The two swap over at roughly 48 degrees of knee bend. That is why the press is the deep-range tool and the extension machine is not in the programme at all.
Setting the weight. Pick a load where the deepest position feels 2 out of 10 or less. Start lighter than seems worth doing.
Speed. 3 seconds to lower under control, then press back up at a normal speed. The slow lowering is the part that builds tissue.
Both legs for now — and this is the thinnest part of the whole design. A two-legged press hides how much your right leg is taking, and it is taking most of it. The deep-range bet rests almost entirely on this machine, at a load set by the ≤2/10 rule, split across two legs. Until the right side is loaded on its own its actual dose is a guess — which is what the monthly single-leg measurement below exists to end.
Single-leg right: the monthly measurement. At each monthly review, one set of single-leg right at the same depth. Find the load that sits at ≤2/10 and write it down beside the two-legged load. That pair is the number. The ratio between them is the only read you can get on what share of the bilateral load the right leg is actually carrying, and it costs one set a month. It does not wait for the gate.
Single-leg right: the progression. It enters the session properly after four consecutive green weeks on the two-legged press — 2×10, same depth, same 3 s lowering, at the measured ≤2/10 load, not at an assumed half. Half was a guess at the very quantity the exercise exists to measure. If it has not gone in by 2 November 2026, that is itself the finding and it gets said out loud at the review rather than quietly rolled forward.
If you cannot get to 2 out of 10 at the bottom even with light weight, use less range today only and write it down. It stays a one-day adjustment, not a new shorter range.
Tomorrow morning. Rate the knee before you get out of bed. That number governs the whole rehab.
Seated hamstring curl
Video＋
Two sets. One set versus three made no difference in a 9-week trial in 42 women with patellofemoral pain: this costs almost nothing.
Moved here from Saturday, 11 August 2026. Saturday was running four posterior-chain exposures in one session — kettlebell swing, back squat with a 4–5 s eccentric, dumbbell RDL, and this. Two were hip hinges and this was a third loading of the same tissue. The redundancy is the stronger argument: the RDL is architecturally equivalent to the Nordic — fascicle length +9%, ACSA +10%, every between-group comparison p>0.48 in a 32-person RCT (PMID 40085810) — so a hinge and a curl on the same day were largely doing the same job.
Why Monday, and why here rather than earlier in it. Monday’s only other posterior-chain item is the hip thrust, which is hip extension; this is knee flexion, so it adds rather than duplicates. It sits after the leg press deliberately: that press is the tracked set the whole knee rehab reads from, and loaded knee flexion in front of it would put a variable into the one measurement that cannot carry noise.
Cycling V2
First thing to cut if the session runs long. That's what it's here for.
Tuesday
▲Ground flow · hips, spine, balance
Feel better after than before.
Ground10 min
Lie face down, forehead on your hands, and breathe into the floor
Face down, not on your back. Stomach on the floor, forehead resting on your hands with one palm over the other. Chest, arms and neck loose.
That position is the whole exercise. Lying prone stops the chest from lifting, so the diaphragm has to do the breathing, and the floor tells you whether it is: your belly should press into the ground on every inhale without you pushing it there.
In slowly through the nose, pause, out slowly for longer than you came in, pause again. Exhale fully before starting the next one.
If the shoulders rise toward the ears, you have gone back to breathing with the chest. Let them go and start the breath lower.
Roll onto your back and rock gently along the spine
Knees toward the chest, hands wherever they are comfortable, and rock along the length of the spine.
Let momentum do it. If you are working, you are doing too much. This is massage for the back of the spine, not an abdominal exercise.
Keep it slow enough that you could stop at any point.
Left leg straight up, let the foot fall open, then tuck the knee in
Left side only, and do not skip it. This is the one exercise in the week aimed straight at the psoas, the muscle that was cut and healed on that side.
The movement. On your back. Lift the left leg toward the ceiling with the knee straight. Let the foot rotate outward so the whole leg turns out from the hip. Then bend the knee and draw it in toward your stomach. Lower and repeat.
Why the turn-out matters. Letting the foot fall open rotates the thigh outward, which is the position the psoas works from. Lifting with the foot pointing straight up hands the job to the muscles on the front of the thigh instead.
Breath leads, and it is not decoration. Float the knee up on the inhale, lower on the exhale. The breath is what keeps this slow, and slow is the difference between a psoas input and a leg swing.
No momentum, no snatching the knee up. If the low back lifts off the floor, you have gone too far.
Six a set, and each set gets 30 s. The set used to run 40 s on the assumption of a six-or-seven-second rep. Measured against what you actually do, six reps take twenty to thirty seconds, so the extra ten was dead time rather than work. Six still sits inside the 2 to 3 sets of 6 to 10 that hip-flexor rehab normally uses.
The timer now names each set. It speaks “Second set” and “Third set” at the boundaries, so you break where you want inside the thirty seconds and get called back. Same on a single tap and on the full session run — those two used to disagree.
Why left only, and why that does not unbalance you. The right psoas is intact; only the left was cut. More to the point, the rest of your week already leans right: every two-legged lift you do is quietly a right-leg lift, because the left cannot take an equal share. Eighteen unloaded leg lifts a week do not tip that balance the other way, they take a small bite out of it.
Dead bug, lowering opposite arm and leg with the back flat
The shape. On your back, arms straight up toward the ceiling, knees and hips both bent to a right angle. Lower one arm and the opposite leg together, slowly, then bring them back.
Flat back is the whole rule. The low back stays pressed into the floor for the entire rep. The moment it lifts, shorten the range. That is not a failure, it is the exercise telling you where your control ends.
Why this is in your programme specifically. It is the only place all week where you train hip flexion while the low back is pinned flat. The easy compensation for a weak psoas is to arch the back and let something else swing the leg. Here that option is removed, and you know the moment you take it.
Left twice for every right. Deliberate, and the same reason as the exercise before it.
Hold each end position 3 s rather than cycling straight through. It was 5 s, which made the set arithmetic longer than this exercise needs: the job here is control at the end position, not time under tension. Three seconds is enough to prove the low back stayed flat, and it puts six reps inside 30 s.
The timer names each set — “Second set”, “Third set” at the boundaries.
Left foot against the wall, push and release
3 seconds of push, 3 seconds of release. Nothing moves, so the effort is the whole exercise.
Push through the whole foot, not the toes.
Five a set, and the set is 30 s. Three seconds pushing plus three releasing is six seconds a rep, so five reps fit exactly. It ran at four and four in a 40 s set until 11 August 2026. Nothing was lost in the change: no source anywhere in the evidence base prescribes a contraction duration for this, and the four-second figure was a design choice, never a finding. Shortening it holds the rep count and takes a quarter off the left leg’s time under load.
Why shorter is the right trade here. This is activation, not exertion — the job is neurological input to what the left leg still has, and Tuesday is already the day carrying three consecutive left-sided items. Time under tension is what a strength exercise is scored on. This is not one.
Left only, same reason as the leg lifts. This is activation for what the left leg still has, and the right does not need it: that quadriceps is normal, and it is the right knee that is symptomatic. Loading it here would buy nothing and cost something.
Last clinical shape of the session. After this, nothing else on Tuesday feels medical.
Transitions: face down for the breathing, roll onto your back for the rocking, and you are already in position for the leg work. Knees stay bent for the dead bug, then roll toward the wall for the press.
Hips + Spine9 min
Right ankle across the left thigh, draw the legs in
On your back. Cross the right ankle over the left thigh just above the knee, then reach through and draw the left thigh toward you.
Stretch, not threat. Ease in. If it grips, back off rather than pushing through it.
Roll the legs slowly side to side inside the position rather than holding one fixed spot.
Why this version and not another. Lying down is the cheapest way into this tissue there is. Pigeon and the seated versions reach the same place but ask you to hold yourself up, and pigeon puts the knee under bodyweight. Nothing to gain by trading up.
Left ankle across the right thigh, draw the legs in
Second side. Left ankle over the right thigh, draw the right thigh in.
Compare the two honestly, but do not push the tighter one to match the other.
Sit up and circle the right leg from front to back
Sitting on the floor, right leg straight. Take that leg out to the side and round, front to back, through the biggest circle you can drive yourself.
Active, not passive. You are moving the joint, not being moved through it. Slow enough that you could stop anywhere in the circle and hold it there.
Keep the rest of you still. If the torso starts swinging to help, the circle is bigger than you own.
What this is worth, stated honestly. The idea is that actively controlling end range holds onto it better than passive stretching does. That principle has support; this specific drill does not have a trial behind it. It costs two minutes and the hips are a stated goal, so it stays, but it is the first thing to cut if the session gets squeezed.
Sit up and circle the left leg from front to back
Second side. Left leg straight, same circle, same speed.
Thread the right arm under and through
On hands and knees. Slide the right arm under the body and across, letting the right shoulder travel toward the floor, then unwind and open back up.
Dynamic rather than held. Breathe through the rotation and keep it moving.
Thread the left arm under and through
Second side, same thing with the left arm.
Single-arm bridge, right arm reaching overhead
Skip this one without guilt if the low back feels sensitised this week. It is the one item on Tuesday with an off switch.
Press the hips up and reach the free arm overhead, opening the chest toward the ceiling.
Drive with the glutes and open through the front of the hip. This is not a lower-back exercise.
Single-arm bridge, left arm reaching overhead
Second side, same reach.
Transitions: figure-four on your back, sit up for the hip circles, onto hands and knees for the thread, then up onto the hands for the bridge.
Standing + Balance5 min
Ordered easiest to hardest: a staggered base, then a narrow one, then one leg. Stand within reach of something solid the whole way through.
Move into a split stance, right foot forward
One foot in front of the other, hip width apart side to side, and move your weight slowly from the back foot to the front and back again.
Slow enough to feel where the weight actually goes. This is the widest base in the block, which is why it comes first.
The progression nobody uses. When this gets easy, do it standing on a folded towel rather than making the movement bigger. Softening the surface is the cheapest way to make any of these harder, and you already own one.
Move into a split stance, left foot forward
Second side, right foot back now.
Move into a tandem stance, right foot in front
Heel of the front foot touching the toes of the back one, so the base is narrow side to side and long front to back.
Head moves, body stays. Turn slowly left and right, then up and down. Moving the head is what makes this harder than the split stance: it brings the inner ear and the neck into the task instead of leaving it all to the feet.
If heel to toe is too much today, leave a small gap between the feet and close it over the weeks.
Move into a tandem stance, left foot in front
Second side, left foot leading.
Stand on the right leg, eyes open, then closed
Hardest position in the block, which is why it is last. Stay within arm’s reach of a wall or a counter.
Eyes closed is the half that matters. Closing them takes vision out and forces the inner ear and your sense of joint position to carry the balance on their own. Spend roughly half the minute each way.
The left side will be worse. That is expected, not a failure.
Stand on the left leg, eyes open, then closed
Second side. Expect a clear difference between the two, and do not chase it.
Close5 min
Free movement
No performance. This is where the nervous system integrates the session.
Do not skip.
Wednesday
▲VO₂max intervals · elliptical
This is the highest-priority session of the week.
Warm-up
Build gradually. Don't arrive at the intervals cold.
VMAX intervals knee · tracked
Untouchable. Nothing else in the programme displaces this.
Last interval should be brutal but controlled, not survived.
Knee: tracked, not capped. This is the largest block of repeated knee bending under load in your week, and it is the one session that cannot be cut. So it does not get the ≤2/10 rule and it does not get a ceiling. It gets a number. Rate the knee tomorrow morning as always, and write Wed next to it so the day is attributable.
Why this is here at all. Nobody has ever measured kneecap-joint pressure on an elliptical — there is no figure to cite in either direction. Untracked is not the same as low. Three tagged exercises were carrying the whole rehab record while most of the week’s knee exposure sat outside it.
If Wednesday mornings are the ones that rise, that is a finding, and it changes the resistance, the stride or the machine — not the session, and not before anything tagged steps back.
V2 cool-down
Actually cool down. Don't step off hot.
Supine leg cycling
Low-threat movement for the left leg with no balance demand.
The timer opens on this session by default: 4 min warm-up, 4×(4 min on / 2 min off), 6 min cool-down — 16 min ON, 34 min total. V5 sets the ON time, not the split; adjust the fields if you run it differently.
Thursday
▲Upper-body pull · flexibility
No HIIT today.
Batchback to back · 90 s break at the end
Pull-ups elbow · out
Video＋
Stop the set when bar speed visibly drops, not when you fail. Training deep into fatigue gained nothing in the one trial that tested it.
Goal is 20 strict. 30 stays an aspiration.
Four sets now, up from three. The cable crossover went in between the pull-up and the handstand, which lengthened the round; the fourth set fits inside time that already existed rather than adding a fourth gap.
The 90 s break comes at the end of the round. Pull-up, crossover, handstand, break puts pull-up to pull-up well past what the lift needs. If elbows feel tender two mornings running, cut volume.
Standing cable crossover
Stand between the two stacks, a handle in each hand, one foot slightly forward. Sweep the hands together in front of you along the line the pulleys set, then let them travel back under control until you feel the chest lengthen. Elbows stay softly bent and fixed — the angle at the elbow should not change through the rep.
The pulley height is the variable, not the load. High pulley sweeps down and in. Middle sweeps straight across. Low sweeps up and in. Four sets: one high, two middle, one low.
The timer calls the height. Run the block and it names high, middle, middle, low as each set comes round. Nothing to track.
Why it sits between the pull-up and the handstand. The 90 s gap after each pull-up was time you were standing in. A horizontal press-pattern fills it without competing: the pull-up is a pull, the handstand is a vertical push, and neither is loaded by a cable fly. It is the only horizontal adduction work in the week now that Monday’s bench is the sole pressing lift.
What it is not. This is added on your own read of what the session was missing, not off a finding. No study puts a cable crossover ahead of any other chest movement, and none has been looked for. If Thursday starts running long or pull-up quality drops, this is the first thing out.
Handstand elbow · out
Video＋
Wrist prep first, every time. Non-negotiable at this volume.
Stack it: wrists, shoulders, hips, ankles in one line. A stacked handstand costs far less per second than an arched one.
Neck extended, gaze between the hands. Neck flexion impairs handstand performance — measured, and the cue you have been missing. No figure attaches to how much of your balance it accounts for.
40 s is time under work, not one unbroken hold. Enter at 25–30 s, come down before the line breaks, go back up — accumulate to 40 s. Never to failure; test a true max once a week or two.
Alternating pairone, then the other · 2 min break
Farmer carry
Video＋
The 2 min gap is the rest, shared with the row. Grip is the limiter: if it fails before the distance, cut the carry time rather than the rest.
Ribs down, shoulders packed. This is spinal-erector endurance, not a grip contest.
Walk tall and breathe. Don't hold your breath.
Single-arm dumbbell row, hand on the bench elbow · light
Video＋
Renamed from ‘Barbell row’, which it never was — this is a bench-supported single-arm row: one hand and the same-side knee on the bench, the other foot on the floor, and the working arm rows a dumbbell up to the ribs.
The halfway cue is the side-swap. 8 reps right, then the timer says halfway, then 8 reps left, inside the same 45 s set. This is why the cue matters here and stays on for this exercise.
Rest stays at 2 min: this is already at the evidence floor.
Pairs well with bench if you want to shorten the session.
Batchback to back · 90 s break at the end
Back extension
Video＋
Range before load. Don't hyperextend at the top.
Side-lying plank
Lift the leg. Glute medius: the nerve supply here is intact on both sides, so both sides should work.
Lying leg raise
Adjust to psoas tiredness: the left side fatigues differently and that's the injury, not weakness of will.
Backward drag kneeafter 2 green weeks
Video＋
What it is for. A way to move the limb under load without provoking the joint, and a warm-up for the knee before the squat and the leg press. Cheap and low-risk.
What it is not. Nobody has published a measurement of knee or kneecap loading during any sled or drag task, in either direction. The reduction seen in backward running comes from landing on the forefoot rather than from going backwards, and in backward walking the ankle does the work — the knee generates almost no propulsion. So do not count this as knee strength training, and do not assume the load is free.
It is still real load, and it has an adjustment. 50–60 kg. The ≤2/10 rule applies here as everywhere else, and because there is no depth to shorten, the only lever is weight: if the knee goes above 2 out of 10, drop 10 kg for that session and write it down. The drag is not a green-week progression variable — it does not compete with load or depth on the press.
How to do it. Short steps, stand tall, do not lean back.
Where it sits. It warms the knee up before the squat and the leg press.
Leg press kneeafter 2 green weeks
Video＋
Goes in after two green weeks in a row, and comes out early on a red week — after the Zone 2, before anything else. It goes back in on the same condition it entered on: two green weeks in a row after the red.
This is the main knee exercise. Use the full range you have, including the deep part. Going deep is the point of it, not a risk being tolerated.
Why deep is safe here and not on a leg-extension machine. In a leg press the pressure behind the kneecap is lower in the deep positions; on an extension machine it is higher. The two swap over at roughly 48 degrees of knee bend. That is why the press is the deep-range tool and the extension machine is not in the programme at all.
Setting the weight. Pick a load where the deepest position feels 2 out of 10 or less. Start lighter than seems worth doing.
Speed. 3 seconds to lower under control, then press back up at a normal speed. The slow lowering is the part that builds tissue.
Both legs for now — and this is the thinnest part of the whole design. A two-legged press hides how much your right leg is taking, and it is taking most of it. The deep-range bet rests almost entirely on this machine, at a load set by the ≤2/10 rule, split across two legs. Until the right side is loaded on its own its actual dose is a guess — which is what the monthly single-leg measurement below exists to end.
Single-leg right lives on Monday. The monthly measurement and the progression are both described there, on the tracked set. Nothing to do about them here.
If you cannot get to 2 out of 10 at the bottom even with light weight, use less range today only and write it down. It stays a one-day adjustment, not a new shorter range.
Tomorrow morning. Rate the knee before you get out of bed. That number governs the whole rehab.
Flexibility9 min
Read before you start
Five positions, once through, no rounds. Two are dose: the only work the week gives those tissues. Two are rehearsal for ranges the lifting already builds.
Intensity does not matter: mild stretching produced the same range gains as intense, so there is nothing to win by going to pain. Frequency does not either: only total weekly minutes per target, which is why this block exists rather than a longer Sunday. The worst positions earn the minutes — a design choice, not a finding: nothing in the evidence ranks gains by starting stiffness. And it reverses: in the one trial that measured detraining here, eight weeks of adductor gains were fully lost four weeks after stopping. One study, one target — but nothing suggests other ranges behave better. (Intensity, frequency and the ~10 min/week/target plateau: Ingram 2025, PMID 39614059, 189 studies. Never before lifting: PMID 36525533. Detraining: PMID 35682148.)
Intensity does not matter: mild stretching produced the same range gains as intense, so there is nothing to win by going to pain. Frequency does not either: only total weekly minutes per target, which is why this block exists rather than a longer Sunday. The worst positions earn the minutes — a design choice, not a finding: nothing in the evidence ranks gains by starting stiffness. And it reverses: in the one trial that measured detraining here, eight weeks of adductor gains were fully lost four weeks after stopping. One study, one target — but nothing suggests other ranges behave better. (Intensity, frequency and the ~10 min/week/target plateau: Ingram 2025, PMID 39614059, 189 studies. Never before lifting: PMID 36525533. Detraining: PMID 35682148.)
Front split, right side: supported slide
Dose. Nothing else in the week loads the front of the hip at length. This is the only stimulus that target gets.
In the gym: use a bench, a box, a low rack pin, or the wall bars. Anything solid at roughly hip height that lets the hands carry weight. The support is what keeps the knee out of it.
Tuck the pelvis first, then slide the rear foot back. The tuck is what does the work: in the one trial that measured it, the pelvic tilt carried the entire effect and the stretch position alone did nothing.
Contract–relax, the same way as Sunday. Settle into the position, then push for 10–15 s at about 60–70% effort: front heel presses down into the floor, rear hip presses back. Release, exhale, and let the position sink. Repeat as you go deeper. Three or four cycles across the 90 s, not one long hold.
Rear knee straight. A bent rear knee moves the stretch onto rectus femoris, which you need for knee extension.
Never before lifting: stretching immediately before a lift measurably costs strength.
What progress looks like: the rear hip opening, not the front foot travelling further. If only the front knee is moving, the target has stopped receiving the work.
Front split, left side: supported slide
Dose. Second side, same target.
Contract–relax, the same way as Sunday. Settle into the position, then push for 10–15 s at about 60–70% effort: front heel presses down into the floor, rear hip presses back. Release, exhale, and let the position sink. Repeat as you go deeper. Three or four cycles across the 90 s, not one long hold.
Compare the two sides honestly. If the left feels vaguer rather than tighter, that is sensation, not range. Do not push to match.
Roll the mid-back over the roller, knees bent
Video＋
Dose. Little else in the week takes the mid-back into extension, and it is the fastest-declining spinal motion.
Breathe out and open over it, then move the roller a couple of centimetres and repeat. Work segment by segment rather than parking in one spot.
Thoracic, not lumbar. If it hinges at the low back, you have gone too low.
Where you should feel it: between the shoulder blades. If the sensation sits below them, the lumbar spine is taking the extension. Back off rather than pushing.
Progression, not a swap: reach overhead on the same roller. Once you can hold the roller version with the low back quiet, let the arms reach back overhead as you extend, aiming to bring the backs of the wrists toward the floor. Thoracic extension supplies roughly the last 15° of overhead shoulder reach the shoulder joint cannot provide alone, so this couples the two directly — no ball needed, same roller, same 180 s. If the sensation migrates down toward the belt line or the low back has to arch to get the arms back, the lumbar spine is making up the difference: stop reaching overhead and go back to arms-behind-the-head for that set.
This is now the only overhead-shoulder work in the block, and that is deliberate. A separate standing shoulder-flexion stretch used to sit after this one, justified as rehearsal for the handstand shape. But Thursday already opens with a real handstand hold — not just Monday's — so a weaker standing rehearsal of a shape you loaded under bodyweight an hour earlier in the same session was doing very little. Retired; this progression carries the range instead.
Sit tall into a wide straddle, hands taking weight in front
Dose, not rehearsal. Saturday's adduction machine builds strength but not range. This and Sunday's middle-split block are the whole of the adductor length work — together they sit at about the ten weekly minutes past which nothing more is gained, which is why two minutes is enough here. (Ingram 2025, PMID 39614059.)
Seated here. What the right knee should not pay for is lowering and rising under eccentric quadriceps control, not the standing position itself, so a standing straddle with the hands down on blocks is fine too. That is how Sunday now runs it. Without something to take the weight, sit.
Hinge at the hip with a flat back. Rounding lowers your chest without opening the hip.
Free to move within the position: side to side, forward and back, small circles. That is what the two minutes are for.
Contract–relax works here too: press the heels into the floor for 10–15 s, release, exhale, widen.
If Thursday runs long, cut here first. Not the split slides, not the roller.
When to stop chasing it: if twelve weeks has not moved the width and the end of the range feels like a hard, abrupt stop rather than a stretch, that is bone rather than tissue. Keep what you have and give the minutes to something else.
Finish
Cycling or elliptical
No added weight. The kettlebell is gone: Thursday already carries the farmer carries.
Zone 2. First thing to cut if the session runs long.
Friday
▲Rest, or gentle movement only
Do less than you want to. Friday is what protects Saturday and Sunday. Walking and dancing still count as the day being used well.
Saturday
▲Lower-body strength · hinge + squat
Legs should feel trained, not irritated.
Batchback to back · 90 s between rounds
Dead-stop kettlebell swing elbow · light
Video＋
Hinge, not squat. Hips back, not down.
Dead stop every rep: no rhythm bouncing.
Calf raises
Dropped to 2 sets to make room for the knee block.
Full range: all the way down, all the way up. Pause at the top.
Low knee cost, so this is a safe place to keep working hard.
Backward drag knee
Video＋
What it is for. A way to move the limb under load without provoking the joint, and a warm-up for the knee before the squat and the leg press. Cheap and low-risk.
What it is not. Nobody has published a measurement of knee or kneecap loading during any sled or drag task, in either direction. The reduction seen in backward running comes from landing on the forefoot rather than from going backwards, and in backward walking the ankle does the work — the knee generates almost no propulsion. So do not count this as knee strength training, and do not assume the load is free.
It is still real load, and it has an adjustment. 50–60 kg. The ≤2/10 rule applies here as everywhere else, and because there is no depth to shorten, the only lever is weight: if the knee goes above 2 out of 10, drop 10 kg for that session and write it down. The drag is not a green-week progression variable — it does not compete with load or depth on the press.
How to do it. Short steps, stand tall, do not lean back.
Where it sits. It warms the knee up before the squat and the leg press.
Barbell back squat knee
Video＋
How deep to go. On your first working rep, lower slowly until the knee sensation is about to reach 2 out of 10. Stop there. That depth is your depth for every set today.
Marking the depth. Set the safety bars in the rack at that height. They give you something to touch so you are not judging depth by feel on every rep. Touch them lightly and stand back up — do not lower the bar onto them and rest.
Decide once. Do not go deeper later in the session because an early set felt fine. The depth is set on rep one and does not change until next week.
Speed. Take 4–5 seconds to lower, then stand up at a normal speed. The slow lowering is where the useful loading happens.
Where your knees go. Do not let your knees slide forward past your toes. Forward knee travel measurably raises the pressure behind the kneecap.
Let your chest come forward. Leaning the upper body forward more, not less, takes load off the kneecap. People with knee pain naturally do the opposite and stay too upright, so this one needs deliberate attention.
Keep your weight toward the front of the foot, not back on the heels. In a measured trial this cut the force through the kneecap joint by about 11% at a 90 degree knee bend, and cost nothing.
No plates under the heels. Elevating the heels increases knee bend and forward knee travel — exactly the positions that raise kneecap pressure.
The load is deliberately low. 40 kg. You were fine at 60 and hurt yourself at 70, then avoided squatting for five months.
What to log. Write down the safety-bar height. It should get lower week to week. If it has not moved in a month, say so.
Tomorrow morning. Rate the knee before you get out of bed. That number, not how the set felt, decides what happens next week.
Leg press knee
Video＋
Volume, not the tracked set. Monday carries the logged load — it is the fresh exposure, where this one sits behind a squat. Straight after the squat is still the right place for it: the squat covers the depth your 2 out of 10 limit allows under a bar, this covers everything below that.
Two sets, and it stays at two. The third set is gone rather than deferred. One set against three made no difference to pain or function in a 9-week trial in 42 women with patellofemoral pain, so a green week spent adding one is a green week not spent on load or depth — and depth and load are the two things the deep-range design actually depends on. It also keeps Saturday from getting heavier for nothing.
What a green week buys instead. Load, or depth, or a rep. One of them, never two. Set count is no longer a progression variable anywhere in the knee work.
This is the main knee exercise. Use the full range you have, including the deep part. Going deep is the point of it, not a risk being tolerated.
Why deep is safe here and not on a leg-extension machine. In a leg press the pressure behind the kneecap is lower in the deep positions; on an extension machine it is higher. The two swap over at roughly 48 degrees of knee bend. That is why the press is the deep-range tool and the extension machine is not in the programme at all.
Setting the weight. Pick a load where the deepest position feels 2 out of 10 or less. Start lighter than seems worth doing.
Speed. 3 seconds to lower under control, then press back up at a normal speed. The slow lowering is the part that builds tissue.
Both legs for now — and this is the thinnest part of the whole design. A two-legged press hides how much your right leg is taking, and it is taking most of it. The deep-range bet rests almost entirely on this machine, at a load set by the ≤2/10 rule, split across two legs. Until the right side is loaded on its own its actual dose is a guess — which is what the monthly single-leg measurement below exists to end.
Single-leg right lives on Monday. The monthly measurement and the progression are both described there, on the tracked set. Nothing to do about them here.
If you cannot get to 2 out of 10 at the bottom even with light weight, use less range today only and write it down. It stays a one-day adjustment, not a new shorter range.
Tomorrow morning. Rate the knee before you get out of bed. That number governs the whole rehab.
Dumbbell Romanian deadlift
Video＋
This is a HINGE, not a sumo squat: that's the correction. A squat pattern is roughly six times the knee-extensor demand of a hinge.
Hips travel back, knees stay nearly still, load stays close to the legs.
This is the one pattern where your left leg can genuinely contribute: hamstrings and glutes are outside the nerve injury.
Start at 20 kg per hand, 40 kg total, and build in 2.5 kg-per-hand steps. The knee is not what limits this lift — a hinge is a fraction of the knee-extensor demand of a squat pattern — so going lighter buys nothing and makes it too light to do its job. The low back and the left side are the limits.
Stop before the left side starts compensating through the lower back.
Push-ups elbow · out
Ribs down, no sag through the middle.
The slow strict ones are the point: the side-to-side ones are the warm-up.
Moved here from Monday, 11 August 2026. Monday was carrying four elbow-extension exposures in one session — bench, handstand, push-ups and dips — and a triceps strain arrived in week one. Saturday had none, which puts three days between these and the bench instead of none. His call, on his own read that Saturday does not run near its ceiling because it is paced for the left leg.
The only upper body on a lower-body day, and that is the point. It sits after the hinge so nothing pressing precedes the squat or the leg press. If Saturday ever does start running to its limit, this is the first thing out — systemic fatigue outranks everything, and adding to the heaviest day is where that rule bites hardest.
Monday’s remaining three exposures are not resolved by this. Whether Monday should carry bench, handstand and dips in one session is open, and sits with the Week 10 deload.
Rotationround the two · 45 s rest between
Hip abduction machine
Video＋
Technique only, as the adduction machine is. The left-versus-right comparison is retired: nothing collected it, nothing acted on it, and the laterality question it belonged to is closed. This is strength work now.
Hip adduction machine
Video＋
The laterality question here is answered. Left roughly matches right. The adductors run off a different nerve than the vasti, so an intact left side puts the injury inside femoral-nerve territory rather than higher up — which is why the hinge, glute and carry work all load normally on both sides. No ongoing left-versus-right tracking is needed; this is strength work now.
Slow and controlled in both directions. No swinging the pads together.
Replaced the Copenhagen adduction, which loaded the adductors at full length, and was the only thing in the week doing that. The machine is safer for the left side but does not build range. Range now has to come from the Sunday middle split instead.
Bike
Only if fresh. Never a duty.
Sunday
▲Deep flexibility · splits + spine
Worked, never threatened.
Nervous-system preparation8 min
Lie on your back, knees bent, and long slow breaths
4 s in, 6–8 s out. Longer exhale downshifts the system.
The position. Flat on your back, knees bent, feet flat on the floor about hip-width apart, arms wherever they are comfortable. Bent knees let the low back settle onto the floor and take the psoas out of the picture — which is the point of bending them.
If the low back will not settle, put a rolled mat or bolster under the knees and let the legs rest on it instead. Same effect, no work to hold the position.
1½ min, down from 2. At 10–12 s a cycle that is seven or eight full breaths, which is enough to downshift. Nothing measures this — the number is a judgement about getting the block done, not a finding.
Stay down and bring your attention to the left leg
No movement. Just attention. This is real work even though nothing happens.
Stay where you are. Same position as the breathing — do not sit up between the two.
This one keeps its full 2 min while the breathing drops to 1½. It is the only pure attention input the left leg gets all week; Tuesday's left-side work is all movement. Shortening it would be cutting the dose rather than tightening the block.
Roll onto hands and knees for cat and cow
Video＋
Slow cat-cow. Spinal extension is the fastest-declining spinal motion past 40: more than flexion, more than side-bending.
Set up on hands and knees. Hands under the shoulders, knees under the hips, spine flat to start.
Cow is the sag. Breathe in and let the belly drop toward the floor: tailbone tips up, chest opens forward, gaze lifts a little. That is the extension half, and it is the half that matters here.
Cat is the arch. Breathe out and push the floor away: the middle of the back rounds toward the ceiling, tailbone tucks under, head drops between the arms.
One breath per direction, no faster. You are moving the spine segment by segment, not swinging between two end positions.
Stand up for hip circles, then leg swings
Standing. One hand on a wall, a rack or a door frame. Supported means holding on, not lying down — this is the one item in the block that is on your feet.
Warming, not stretching. Swing to the range that arrives, not the range you want.
Circles both directions, swings front-to-back and side-to-side, both legs. Split the 2 min however it falls.
Transitions: breathing → stay exactly where you are for the left leg → roll onto hands and knees for cat-cow → stand up for the hip circles. The block climbs from lying to standing on purpose, so nothing gets set up twice.
Front split block10 min
One leg all the way through, then the other. Four steps per side, five minutes a side. Primary target: front-leg hamstring and rear-hip flexor, worked as one movement. Supported, controlled, no bouncing.
Right leg straight in front, fold into the half split
The shape. Right leg straight out in front, heel on the floor, toes up. Left knee on the mat directly under the left hip, top of that foot flat. Hips shift back over the kneeling knee. Fold forward over the straight front leg with a long spine — chest toward the shin, not chin toward the knee.
The front leg stays straight. That is what makes this a half split rather than a lunge. A soft micro-bend at the knee is fine; a working bend is not.
Where you should feel it: the back of the front thigh. This half is hamstring.
Hands on the floor, on blocks, or on a low bench — whatever lets the spine stay long.
Right leg forward, tuck the pelvis and slide toward the split
Tuck before you slide. Tip the pelvis under first, then send the back knee backwards. The tuck is the part that does the work: in the one trial that measured it, the pelvic tilt carried the entire effect and the stretch position alone did nothing.
Both legs straighten. The front leg was already straight and stays straight; the back leg is what changes, going from bent under the hip to extended behind you. This is the move from half split toward full split.
Rear knee straight, not bent. A bent rear knee moves the stretch onto rectus femoris — a muscle your left leg still has and needs for knee extension. Keep it out of this.
Now you are working both halves at once. Front-leg hamstring and rear-hip flexor, one movement. That is why the hip flexor is not trained as a separate exercise anywhere in the week.
Go only as far as the hips allow with the pelvis still tucked. Depth that costs the tuck is not depth.
Right leg forward, press then release and sink deeper
You do not move between positions. You stay exactly where the slide left you for the whole two minutes. The only thing that changes is tension on and tension off.
Isometric means press without moving. The muscle works, nothing travels. Front heel presses down into the floor; rear hip presses back. About 60–70% effort, 10–15 s.
Then release and exhale. Let the position sink into whatever new range has opened. Stay there 20–30 s, breathing.
Three rounds across the two minutes. Press, release, sink. Press, release, sink. Press, release, sink. Each one should go a little further than the last.
Why it works. The contraction briefly quiets the muscle’s protective stretch reflex, which opens a short window where the tissue will lengthen further than it would passively. That is the whole mechanism.
Right leg forward, settle and breathe
Stay where you are. No new position, no pushing. This is the passive minute that follows the pressing.
Not a target depth. This is not a cue to force a flat, both-legs-straight split. Let the exhale do it and take whatever arrives today.
If the left side feels vaguer rather than tighter, that is sensation, not range. Do not push to match the right.
Left leg straight in front, fold into the half split
The shape. Left leg straight out in front, heel on the floor, toes up. Right knee on the mat directly under the right hip, top of that foot flat. Hips shift back over the kneeling knee. Fold forward over the straight front leg with a long spine — chest toward the shin, not chin toward the knee.
The front leg stays straight. That is what makes this a half split rather than a lunge. A soft micro-bend at the knee is fine; a working bend is not.
Where you should feel it: the back of the front thigh. This half is hamstring.
Hands on the floor, on blocks, or on a low bench — whatever lets the spine stay long.
Left leg forward, tuck the pelvis and slide toward the split
Tuck before you slide. Tip the pelvis under first, then send the back knee backwards. The tuck is the part that does the work: in the one trial that measured it, the pelvic tilt carried the entire effect and the stretch position alone did nothing.
Both legs straighten. The front leg was already straight and stays straight; the back leg is what changes, going from bent under the hip to extended behind you. This is the move from half split toward full split.
Rear knee straight, not bent. A bent rear knee moves the stretch onto rectus femoris — a muscle your left leg still has and needs for knee extension. Keep it out of this.
Now you are working both halves at once. Front-leg hamstring and rear-hip flexor, one movement. That is why the hip flexor is not trained as a separate exercise anywhere in the week.
Go only as far as the hips allow with the pelvis still tucked. Depth that costs the tuck is not depth.
Left leg forward, press then release and sink deeper
You do not move between positions. You stay exactly where the slide left you for the whole two minutes. The only thing that changes is tension on and tension off.
Isometric means press without moving. The muscle works, nothing travels. Front heel presses down into the floor; rear hip presses back. About 60–70% effort, 10–15 s.
Then release and exhale. Let the position sink into whatever new range has opened. Stay there 20–30 s, breathing.
Three rounds across the two minutes. Press, release, sink. Press, release, sink. Press, release, sink. Each one should go a little further than the last.
Why it works. The contraction briefly quiets the muscle’s protective stretch reflex, which opens a short window where the tissue will lengthen further than it would passively. That is the whole mechanism.
Left leg forward, settle and breathe
Stay where you are. No new position, no pushing. This is the passive minute that follows the pressing.
Not a target depth. This is not a cue to force a flat, both-legs-straight split. Let the exhale do it and take whatever arrives today.
If the left side feels vaguer rather than tighter, that is sensation, not range. Do not push to match the right.
Transitions: the four steps are one position getting progressively deeper — you set up once per leg and stay on the floor until the side is done. Swap legs only at the halfway point.
Middle split block8 min
One round. Primary target: adductors. High fall-risk relevance.
Wide stance, shallow horse-stance pulses
Shallow is the instruction. Feet wide, toes turned out, and only sink a little way. This is a hip position, not a squat.
Keep the weight on both feet evenly. Standing on one leg at a time would load the kneecap two and a half times harder, and there is nothing to gain from it here.
Step wide, slow side lunges, staying shallow
Shallow is the whole instruction. Sink deep here and it loads the kneecap harder than a forward lunge does. Staying shallow is what keeps this an inner-thigh position instead of a knee exposure.
Standing wide straddle, hands down onto blocks
Standing, with the hands supported. Feet wide, toes forward or slightly out, kneecaps pointing the same way as the toes. Hinge forward from the hips with a flat back and put the hands down onto yoga blocks.
Why this is allowed now, when the rule used to say seated only. The objection was never to standing. It was to lowering yourself down and pushing back up under eccentric quadriceps control, which the right knee should not be paying for. Hands on blocks remove that: the arms carry the descent and the return. Your call, tested 9 August 2026.
Blocks at whatever height lets the back stay flat, and lower them as the hips open. Block height is the progression here, not how wide the feet go.
Free to move around in it: side to side, forward and back, small circles. That is what the two minutes are for.
It also sets up the next two. You finish already hinged forward over wide legs, so you sit straight down into the press and stay there for the reach.
Sit down into the straddle, press your heels 15 seconds, then release and widen
Same method as the front split, different target. Stay sitting in the straddle the whole time. Press the heels down into the floor for 15 s at about 60–70% effort. Press only, nothing moves.
Then release, exhale, and let the legs widen into the range that just opened. Settle there before the next press.
Three rounds across the minute: press, release, widen. Press, release, widen. Press, release, widen.
When to stop chasing it: if twelve weeks has not moved the width and the end of the range feels like a hard, abrupt stop rather than a stretch, that is bone rather than tissue. Keep what you have and give the minutes to something else.
Still sitting, reach forward along the floor with a flat back
Why it is here and not in the spine block. This is adductor and hamstring work, not spine work. It sits with the middle split because that is the same target, deepened: the straddle opens the hips with the torso upright, this adds the forward hinge and the hamstrings with it.
It matters more than it used to. Saturday's Copenhagen adduction was replaced by the adduction machine, which builds strength but not length. These minutes and the straddle are now the only adductor length work in the week.
Pull yourself forward, do not fall forward. Reach actively along the floor with a flat back; letting the chest sink rounds the spine and takes the stretch off the hips.
Spine block6 min
Spine adapts quickly but regresses easily. This is why it lives here.
Loaded flexion first, then the bridge built in three steps: wake the glutes on the floor, walk down the wall, then open the shoulders once you are there.
Stand on the box and roll down one vertebra at a time
Roll, do not hinge. Chin tucks first, then round down through the spine segment by segment, letting the weight hang. Reverse it the same way, stacking from the bottom up.
Very light. This is the loaded counterpart to everything else in the session — the point is control through flexion, not load.
Legs stay straight. Standing on a box is what gives the weight somewhere to travel.
Lie on your back, lift the hips and squeeze the glutes
This is the glute activation, and it earns its place here. The bridge is a glute and thoracic movement before it is a lower-back movement. Waking the glutes first is what keeps the low back from doing the whole job when you go over.
On your back, knees bent, feet flat and close to your hips. Drive through the heels, lift the hips, squeeze at the top, lower slowly.
One exercise, both feet down. Not a single-leg version, and nothing to lift off the floor. This is 90 s of waking the glutes up before you go backwards.
Ribs down. If the low back is doing the lifting rather than the glutes, you have gone too high.
Cheapest goal you have. The bridge involves no lesioned tissue — thoracic spine, shoulders, glutes and erectors are all intact, and it needs no loaded knee flexion.
Back to the wall, walk your hands down into a bridge elbow · out
This is the actual bridge practice. Stand an arm’s length from the wall, back to it, feet shoulder-width. Reach both arms overhead, squeeze the glutes, and lean back until your palms find the wall.
Then walk down, one hand at a time, stepping your feet a little further from the wall as you descend. Go as low as is comfortable — the floor is the eventual destination, not today’s requirement.
Walk back up the same way. Hand over hand, glutes engaged the whole time. Do not fold in the middle and drop.
Open through the shoulders and the upper back rather than hinging at the low back. If it pinches at the belt line, you are bending where you should be opening.
Shoulder range is not trained here, and does not need to be. It is already dosed three times over: the handstand holds on Monday and Thursday load exactly this range under bodyweight, and both the Thursday flexibility block and the handstand practice carry a 90 s shoulder-flexion position. This block spends its minutes on the shape, not on the range that feeds it.
What progress looks like, and why there is no drill for it. Eventually the shoulders travel out over the wrists and the bridge holds itself. Rocking into that deliberately is an advanced drill and it is not in this block: with shoulder flexion still the limiter, practising it now would only teach the low back to make up the difference. Notice how far down the wall you get. That is the measure.
Parasympathetic close5 min
Lie back with the legs supported and let everything settle
Not optional. Long exhales. Let the system settle.
Pull-up bar
▲Home · whenever you pass it
Evening pull-ups toward 20
Grease-the-groove sets elbow · outWalk past the bar, three easy pull-ups, walk away. Again an hour later.
Picture it. You pass the bar on the way to the kitchen. Jump up, three easy reps, chin over the bar, drop off while it still feels light. An hour later, again.
Half your current max, roughly. If your best set is 7, do 3–4.
Never to failure, never even close. The whole point is that it costs nothing to recover from.
This is where the extra volume comes from: 80–120 quality reps a week without adding a minute of gym time.
If elbows are tender two mornings running, stop for a few days.
Remember
Half your best set. Best is seven, do three.
Chin over the bar every rep.
Drop off while it is still easy.
Negatives elbow · outJump or step to the top of a pull-up, then fight the whole way down.
Picture it. Jump so your chin starts above the bar. Then lower for eight to ten seconds, fighting all the way to straight arms.
Replaces the band set: you do not have bands, and this needs nothing but the bar.
Jump or step to the top, then lower as slowly as you can hold the position. Eight to ten seconds is the target; when you can do twelve, add a rep instead.
Costs more recovery than grease-the-groove. Once a week, not more, and not the night before Thursday.
Remember
Count the seconds out loud.
Twelve seconds easy? Add a rep, not seconds.
Once a week. Never the night before Thursday.
Dead hangHang from the bar, arms straight, shoulders switched on, until the grip goes.
Picture it. Straight arms, feet clear of the floor, shoulders pulled down a little so your ears are not buried. Then relax everything except your hands.
About 1 min now, 2 min is the goal. This is its own target, not a pull-up accessory.
Shoulders active, not fully collapsed: a passive hang all the way down loads the capsule rather than the muscle.
Grip is the limiter and it recovers fast. Safe to do most evenings.
Chalk or a towel over the bar if the skin gives out before the grip does.
Remember
Shoulders active, not collapsed.
Let the grip be what fails.
Chalk or a towel if the skin goes first.
Thursday carries the strict work. This is volume the gym week has no room for, in pieces small enough that nothing has to recover from them.
Elbow rehab
▲Temporary · until the elbow exit gate
Right triceps MTJ strain, day zero 12 August 2026. Four items, each with its own entry condition. The Elbow Governor (reference sheets, below) holds the ladder, the weekly gate and the exit gate; this block is the work itself. Everything here retires at the exit gate.
Unloaded elbow range of motion
Comfortable range only, both directions, a few times a day. Pain peaking in the first days is expected; this is motion, not loading.
Formalised here 12 August so it stops being an unwritten instruction. Retires at full return.
Isometric elbow extension holds
Press the palm down into a table or the other hand with the elbow bent about 90°, hold, breathe. Pain-free is the whole prescription: find the effort where nothing complains and stay under it.
Entry step of early loading (principle: PMID 28953439; consensus: PMID 36650032). No acute-strain isometric trial exists, so the dose is labelled inference — and the tendinopathy dosing (80% MVIC) deliberately does not transfer to a fresh grade-2 strain.
Cable pressdown
The tracked vehicle — the elbow’s leg press. Elbow-isolated, smallest increments the stack allows, and the shoulder-neutral position keeps the long head short, which is the lowest-strain place to start loading the MTJ.
Slow tempo: heavy-slow-resistance matched eccentric-only work in tendinopathy with better compliance (PMID 26018970; transfer inference). Mention the load in the debrief when you have it; it is logged from what you say, not chased.
Retires — or converts to the overhead extension — at the dips rung.
Overhead triceps extension
The lengthened progression. The long head crosses the shoulder, so the overhead position loads it at length — and it grew 1.5-fold more trained overhead than neutral, at lower absolute loads (Maeo 2023, PMID 35819335; healthy-adult evidence, rehab transfer inference).
This is the bridge to dips and handstand pressing: the injured structure loaded at length, under control, with small weights. Succeeds the pressdown in the same Tue/Sat slots.
Whether it keeps a permanent slot after the rehab is decided at the Week 10 deload, with Monday’s pressing question — not before.
Handstand practice
▲Home · 10–15 min · eight phases
Pick a phase
Warm-up, every session 3 min
Wrist rangeOn all fours, rock the hands forward and back and circle them — palms down, then fingertips, then backs of the hands.
Kneel on all fours, hands flat under your shoulders, fingers forward. Rock slowly forward over your hands until the inside of your forearms stretches, then back. Repeat with the fingers turned to your knees, then on the backs of the hands.
Palm rocks, fingertip pulses, wrist circles loaded and unloaded. Work the extension range specifically: that is the one the handstand needs.
Do this for range, not as insurance — a survey of handstand practitioners found no link between warming up and lower wrist-pain rates.
Remember
The forward rock is the one that matters.
No bouncing at the end of the range.
Every session, before anything goes upside down.
▶Daily Wrist Warm Up Tom MorrisonShoulder flexionHands on a bench, walk the feet back, let the chest sink toward the floor with the ribs down.
Hands on a bench at hip height. Walk your feet back until your arms are straight and your body makes an L. Let your chest sink toward the floor between your arms.
Here because you say it is what limits you. If the arms cannot reach in line with the torso, the line closes by arching the low back or pushing the hips forward instead — and an arched handstand costs far more per second.
Ribs down. If the low back arches to get the arms further, you are training the compensation.
Do it before the holds, and on days you are not doing handstands too.
Remember
Ribs down. An arched back is not more range.
Hips stay back over your heels.
Breathe out at the bottom and let it settle.
Phase 1 · Find out what is wrong12 min
Gate outThree numbers written down, and you have watched the video back.
Film it: wall-facingFace the wall, walk the feet up until chest, front thigh and toes all touch it.
Picture it. Hands on the floor a metre from the wall. Walk your feet up it behind you and your hands in, until you finish flat against the wall: chest, front of the thighs and tops of the feet all touching.
That position takes balance out of the picture, which is the only way to see the line by itself.
Blocks If your wrists will not reach vertical, put the BalanceBlocks under the heel of each hand. It tells you whether the wrist is what limits the line.
Remember
Hands shoulder-width, fingers spread, pointing away from the wall.
All three contacts touching, or the frame does not count.
Phone on the floor, side on, three metres back.
Film it: freestandingKick up in open floor, hold, come down. Camera stays where it was.
Picture it. Same camera, open floor. Kick up the way you normally do, hold what you hold, come down how you come down.
Remember
Film the handstand you do, not the one you can produce.
Three attempts, then stop.
Watch it back later, not between attempts.
Read the five checkpointsPause on the best frame and check five joints, bottom to top.
Picture it. Sitting down with the video, not upside down. Freeze the steadiest frame and read yourself from the floor up.
Read it from the wrist up. A fault low down forces a compensation higher up, so the first thing that is wrong is the thing to fix. The Movement Athlete, Ultimate Guide to a Freestanding Handstand.
Work upward, because a restriction low down forces the compensation higher up. A wrist that cannot reach vertical pushes the shoulders forward, and the low back arches to put the feet back over the hands.
Remember
Wrist — forearm vertical, or leaning?
Elbow — locked, or bent under load?
Shoulder — upper arm in line with your torso, or in front of it?
Low back — flat, or arched?
Hips — stacked, or pushed forward?
Can you breathe without the shape changing?
Three numbersNothing to do but write. Three figures, one line.
Picture it. Phone, a note, today’s date, three numbers. That is the whole exercise.
These replace the 40 s figure that used to sit in this file. It was written here without a measurement behind it.
Remember
One. Best freestanding hold, out of three honest attempts.
Two. Longest wall-facing hold with all three contacts and a quiet low back.
Three. Out of ten kick-ups, how many held three seconds without a save.
Redo this page monthly.
Watch the whole thing onceOnce, not every sessionSit down and watch one tutorial start to finish. Then stop watching handstand videos.
Picture it. Not in the session. One coherent tutorial, start to finish, once.
Remember
Watch once, then practise.
If you want to rewatch something, rewatch your own footage.
Phase 2 · The line13 min
Gate out3 × 30 s wall-facing with all three contacts and no arch, and a 45 s hollow hold with the arms overhead.
Wall-facing hold, three contactsChest, front thigh and toes touching the wall, hands a step back from it.
Picture it. Upside down facing the wall, belly to the paint. Hands a forearm’s length out from the skirting board. Feet walked down and hands walked in until chest, thighs and toes all touch. Head between your arms, eyes on the floor. One straight line, flat to the wall.
You cannot cheat it. If your low back arches, your chest peels off the wall and you feel it happen.
Why hollowing works, in one picture. A flat sheet of paper falls over on its edge; curve it and it stands. Rounding the upper back does the same thing to your handstand — it widens the base in the direction the line of force actually runs. George 1980, reproduced in Uzunov 2008, Figure 8.
Remember
All three contacts, the whole set.
Chest off the wall means the back arched. Come down.
Ribs down first, then push the floor away.
Progress by walking the feet closer. Never the hands.
Handstand shrugs, wall-facingIn the wall handstand, push the floor away until the ears vanish, then let the head sink between the arms.
Picture it. In that same wall handstand, push the floor away until your shoulders climb to your ears and your head disappears between your arms. Then let it all sink back down, elbows locked throughout. Eight slow ones.
That travel is scapular elevation — the range the top of a handstand sits in. Trapezius and anterior deltoid together carry about 40% of the muscular contribution in a handstand. (Kochanowicz 2018, PMID 29471194.)
Remember
Elbows locked. If they bend, the set is over.
Two seconds up, two seconds down.
The movement is the shoulder blades, not the neck.
Hollow hold, arms overheadOn your back, arms by the ears, legs straight, everything a few inches off the floor.
Picture it. On your back, arms stretched out beyond your head, legs straight, toes pointed. Press your low back flat, then lift arms, head and legs a few inches. A shallow curve, bending the opposite way to a banana.
The handstand shape lying down, where you can feel whether it is real.
Remember
Low back glued to the floor. The only rule here.
If it lifts, bend the knees and hold that instead.
Toes pointed, legs squeezed.
Posterior pelvic tilt at the wallBack to the wall, tuck the tailbone until the low back flattens against it.
Picture it. Back to a wall, heels a few inches out. A hand fits in the gap behind your low back. Tuck your tailbone until the gap closes on it. Hold, release, find it again.
You are learning to find it on demand, so it is available upside down.
Remember
The pelvis rolls. The ribs do not crunch down.
Find it, lose it, find it again. The search is the training.
This is the fix for the banana, not more core work.
Phase 3 · The entry11 min
Gate out7 of 10 kick-ups arrive and hold 3 s on the strong leg. 4 of 10 on the other.
Lunge entry, no kickArms up by the ears, step forward, hands down, shoulders travel over — stop just before you tip.
Picture it. Arms straight up by your ears. Step forward, reach for the floor with your fingers, and let your shoulders travel past your hands as they land. Back leg lifts behind you. Stop where one more centimetre would tip you over, and come back.
The shoulders travel over the hands, they do not get thrown there. The kick supplies the last few degrees, not the movement. Uzunov 2008, Figure 2 — the lunge entry.
Reach with the fingers, not the heel of the hand. That puts your weight over the fingertips from the first instant, which is where it has to be to balance.
Remember
Arms up before the foot moves.
Shoulders travel over the hands. They are not thrown there.
Stop before you tip.
Kick-up accuracy: strong legTen kick-ups in open floor. Count how many arrive and stay put.
Picture it. Same entry, but let it go all the way over. Kick just hard enough to arrive and stop — not hard enough to sail past. Deliberately too soft for the first two, then creep the power up.
Remember
An arrival is three seconds with no visible save.
Ten attempts, then stop, whatever the score.
Write the number down.
Kick-up accuracy: other legSame ten, leading with the other leg.
Picture it. The same drill, other foot forward, other leg kicking. It will feel foreign and score badly. That is the information.
The left leg is worth watching. The quadriceps deficit is left-side only, so a left-leg push-off has less to push with. (Laterality.md, v1.4.) Whether it actually limits the kick-up is open — the push comes largely from the hip, and the hip is spared.
Remember
Same rules: three seconds, no save.
Expect a much lower score. Track it, do not fix it tonight.
If the left leg cannot generate the push, say so at the debrief.
Over-kick and under-kickEvery 2nd sessionKick too hard and step out the far side. Kick too soft and come back down. On purpose, both.
Picture it. Two deliberate mistakes. Kick far too hard, sail past vertical, step out the far side. Then far too soft, never arrive, come straight back down. Five of each. You are drawing the map of the edges.
This doubles as bail practice, which is why it sits here rather than in its own phase.
Remember
Both directions in the same set. The pair is the drill.
Step out on your feet. Do not collapse.
Soft floor — a hard landing makes the next attempt tentative.
Phase 4 · Balance range12 min
Gate out30 s freestanding on more than half of attempts, and one save from a visible over-balance.
Back-to-wall heel pullsAlternateBack to the wall, peel the heels off a centimetre at a time until the wall is doing nothing.
Picture it. Back to the wall, heels resting on it, belly out to the room. Peel one heel off, then the other, pulling your hips over your hands a centimetre at a time, until the wall carries nothing at all. Hover there. When it goes, let the heels drop back.
The wrist is the ankle of the handstand. The three figures on the left are counter-balancing — big corrections made by moving the hips and shoulders. The one on the right is on-line balancing, everything stacked and the correction made at the wrist alone. That is the one to train for. George 1980, reproduced in Uzunov 2008, Figure 3.
Remember
A centimetre at a time, not in one go.
Hips travel over the hands. Do not just lift the feet.
When it goes, drop back to the wall. The wall is the design.
Wall-facing toe pullsAlternateFacing the wall, the toes leave it and you come back toward vertical.
Picture it. Chest to wall. Lift your toes off it and let your weight come back toward vertical, clawing the floor with your fingers to stop yourself going over the top. The same balance point as the heel pulls, approached from the other side.
One or the other, not both. Alternate them session to session. Running both is what pushed this phase past twenty minutes.
Remember
The fingers do the braking. Claw the floor.
Toes leave the wall; the chest stays put.
Alternate with the heel pulls.
Fingertip pressureOn hands and knees, rock forward and stop it with the fingers, back and stop it with the palm.
Picture it. Hands and knees, shoulders over your wrists. Rock forward and stop yourself through your fingertips. Rock back and stop yourself through the heels of your hands. Then take the same sensation upside down.
Gymnasts hold a handstand with a wrist strategy, only recruiting the shoulders, hips and elbows once the wrist has run out. The wrist flexors carry the highest muscular demand of any group in the position.
Remember
Going over the top: press the fingers.
Falling back: press the heel of the hand.
Hands stay flat. A lifting palm means the brake is gone.
Freestanding, range not clockKick up in open floor and go looking for the wobble instead of watching the clock.
Picture it. Open floor. Instead of freezing as still as you can, let yourself drift a few degrees off vertical and bring it back with your fingers. Then drift the other way. You are hunting the edge of what you can recover from.
Gaze between the hands, neck long. Neck flexion measurably impairs handstand performance, and closing the eyes reduces stability. (MacDonald 2025, PMID 41473027.)
Remember
Come down on the save, not on the clock.
Look at the floor between your hands.
Drift both ways. Most people can only save one.
Phase 5 · Make it robust13 min
Gate outNone. This is where it stops being a progression and starts being a practice.
Quarter pirouette outEvery sessionFrom a good handstand, turn a quarter on one hand and step down facing back the way you came.
Picture it. Up and steady. Shift onto one hand, turn a quarter around that palm, and step both feet down facing the way you came from.
Practise it from handstands that were going fine, so it is there when one is not.
Remember
Turn on one hand. Do not throw the legs.
Practise from good handstands, not only failing ones.
Same direction every time.
Parallettes holdsPick 2 of 5The same handstand with the hands on the wooden bars, wrists straight instead of bent back.
Picture it. Bars shoulder-width apart, gripped like a hammer. Kick up. Wrists straight instead of bent back, and the whole thing ten centimetres higher — narrower balance, longer fall.
PowerBars The neutral grip takes the wrist out of extension. Easy on the wrists, harder everywhere else.
Do not migrate to them because they are comfortable. The floor handstand is the one that transfers.
Remember
Grip the bars. The fingers still balance.
Bars parallel and shoulder-width. Check before kicking up.
Bail off the side, not forward over the bars.
Wall-facing shoulder tapsPick 2 of 5Chest to wall, shift all the weight onto one arm, touch the other shoulder, put the hand back.
Picture it. Chest to wall. Move your hips and ribs sideways until one arm carries nearly everything, then lift the other hand and touch that shoulder. The shift is the exercise; the tap only proves it happened.
Remember
Hips square to the wall. No twisting to buy the tap.
Shift first, lift second.
Feet stay quiet on the wall.
Straight-arm tapsPick 2 of 5Same shift, but the free arm goes out sideways with the elbow locked instead of bending to the shoulder.
Picture it. The same shift, but the free arm stays locked and swings out to the side like a wing. Longer lever, much more load on the arm still down.
Only after the shoulder taps are clean.
Remember
Free arm dead straight. A bent elbow is the easier drill.
Slowly out, slowly back. No swinging.
Stop when the supporting elbow softens.
Shape changesPick 2 of 5Go up straight, pull to tuck, open to straddle, come back to straight without coming down.
Picture it. Kick up straight and settle. Pull the knees to your chest. Open wide into a straddle. Back to straight. One leg halfway down and back. All without coming down.
Every change moves the centre of mass and the wrist has to answer it. These are also the entry to the press work in Phase 6.
Remember
Change shape slowly. Fast is just falling.
Correct at the wrist, not by piking.
Return to straight between shapes.
Eyes offPick 2 of 5Same handstand, eyes shut, back near the wall.
Picture it. Back to the wall. Kick up, settle, close your eyes. Everything will feel like it is moving, because half your balance information just left.
Read this before doing it. Closing the eyes measurably reduces handstand stability. That training blind therefore builds proprioception is a plausible but untested idea, not a proven one.
Remember
Wall behind you, always.
Open your eyes when it goes. Do not ride it down blind.
Play, not dose.
Phase 6 · Press to handstand13 min
Gate outThree clean tuck presses on the bars, then one straddle press with straight legs throughout.
L-sit on the barsEvery sessionSit between the bars, hands down, arms locked, legs straight out in front and clear of the floor.
Picture it. Sit between the two bars, a hand on each, arms straight beside your hips. Press down until your backside lifts, then lift your legs straight out in front. Your body makes an L hanging in the air.
PowerBars The bars exist for this. On the floor your heels catch; here there is room underneath, which is also what lets the legs pass through later.
Not there yet? Do it tucked and extend one leg at a time.
Remember
Shoulders pressed down, away from the ears.
Arms dead straight.
Legs together, toes pointed. Tucked is a fine regression.
Compression liftsEvery sessionSit on the floor, legs straddled wide, hands on the blocks, lift both legs clear of the ground.
Picture it. Sit with your legs straight and spread wide, a wooden block under each hand beside your hips. Press down and lift both legs off the floor. A few centimetres is a real rep.
Blocks Hands flat on the floor gives you nothing to lift into.
This is the limiter for most people, and it is not shoulder strength. A press is a compression problem first. If the legs will not come up sitting down, they will not come up in a handstand.
Remember
Sit tall. Rounding the back trains the wrong thing.
Slow up, slow down.
Legs straight. Height is irrelevant.
Lean and liftPick 1 of 3From the L-sit, lean the shoulders forward past the hands until the hips float up and the knees tuck in.
Picture it. From the L-sit, lean your shoulders forward past your hands as if you were about to fall face-first. As they travel, your hips float up behind them and your knees come toward your chest. The two trade off over your hands.
That exchange is the whole press. Everything else is a variation on it.
Remember
Shoulders lead. The hips follow — you do not lift them.
Arms locked throughout.
Only as far as you can come back from.
Negative pressPick 1 of 3From a handstand on the bars, lower slowly into a straddle without letting the legs drop.
Picture it. Handstand on the bars. Now run the press backwards: open the legs wide and let them descend slowly while your shoulders travel back over your hands to counterbalance.
The press learned backwards, which is how most people find the balance point in it.
Remember
Legs straight and wide the whole descent.
Shoulders travel back as the legs come down.
Stop when it becomes a fall rather than a lower.
Tuck press to handstandPick 1 of 3From the L-sit, the legs pass back between the bars, knees to chest, then open up to handstand.
Picture it. From the L-sit, lean forward, let the hips rise, and pass your legs backward down through the gap between the bars with your knees tucked tight. Hips end up above your shoulders in a tight ball. Then extend to handstand.
This is the move you described. The legs go back through the gap, which is what the bars make possible and the floor does not.
Check the clearance on the first attempt. No published height exists for the PowerBars XL, so I do not know whether your feet clear. If they do not, run the press on the floor as a straddle and keep the bars for the L-sit.
Remember
Knees tight through the gap. A loose tuck catches the bars.
Extend only once the hips are over the shoulders.
Straddle press is the next rung.
Phase 7 · Handstand push-up12 min
Gate outFive strict full-range wall-facing reps, head to floor, no bounce.
Pike push-up, feet elevatedEvery sessionFeet up on a chair, hips high, head travels down to the floor between the hands.
Picture it. Feet on a chair, hands on the floor, hips high so your body makes an upside-down V. Bend your arms and let the top of your head travel down between your hands, then press up.
The handstand push-up with a third of the load taken out.
Remember
Elbows forward and slightly out, not flared.
Head goes between the hands, not in front.
Progress by raising the feet, not adding reps.
Wall-facing HSPU, partialPick 1 of 3Chest to wall, bend the arms until the head touches a block, press back up.
Picture it. Chest-to-wall handstand with a wooden block on the floor between your hands. Lower until the top of your head touches it — touches, not rests — then press back to straight arms.
Blocks The block makes the depth identical every rep, which is the only way to know the range is increasing.
Wall-facing, not back-to-wall. Back-to-wall lets the hips arch off and hides the range.
Remember
Head touches, never rests.
Chest stays on the wall through the rep.
Lower the block over weeks. That is the progression.
Full-range wall-facing HSPUPick 1 of 3Same thing, block removed, head to the floor and back up.
Picture it. Identical, block gone. The top of your head goes all the way to the floor between your hands and comes back up.
Five strict reps is the gate out of this phase. Not five with a bounce.
Remember
Pause at the bottom. No rebound.
Last rep slower than the first? The set was one too long.
This replaces the partial.
Deficit HSPU on the barsPick 1 of 3Hands on the wooden bars so the head can drop below them — more range than the floor allows.
Picture it. Hands on the bars, chest to wall. Now your head can pass below the level of your hands — range the floor will not give you.
PowerBars The second job the bars do that nothing else can.
Only once full-range on the floor is comfortable.
Remember
Only after five strict full-range floor reps.
A little below the bars at first, not as deep as possible.
Stop if the shoulder feels pinched rather than worked.
Phase 8 · One arm12 min
Gate outTen seconds on one arm in a straddle. Realistically a multi-year gate that may never close — which is not a reason to skip the shift work, because it improves the two-arm handstand on its own.
Two-hand weight shiftEvery sessionIn a handstand, walk the hips sideways until nearly all the weight sits over one hand, then back.
Picture it. Up in a handstand, both hands staying down. Slide your hips and ribs sideways over one hand until that arm carries nearly everything and the other is barely touching. Shoulders stay level. Hold, then travel back.
The whole skill is here. A one-arm handstand is a two-arm handstand with the weight fully stacked; taking the hand off is the last and smallest part.
Remember
Shoulders level. Tilting means you leaned instead of stacking.
Loaded shoulder ends up over the loaded hand.
Both hands stay down. This is the drill, not a warm-up for the next one.
Free hand to a blockEvery sessionSame shift, then the light hand comes up onto a block so it can only steady, not carry.
Picture it. Same stack, but the light hand rests on a wooden block. Raised, it cannot take real load — only steady you. Aim to lift it off for a second without anything else moving.
Remember
If anything moves when the block hand lifts, the weight was not across.
One second off the block is a success.
Loaded arm locked straight.
Fingertips onlyPick 1 of 2Light hand comes down to fingertips, then to one or two fingers.
Picture it. Same stack, light hand down to its fingertips. Then three fingers. Then one. Support removed in steps rather than all at once.
Remember
Five fingers, then three, then one. Over weeks.
Wall behind you until one finger is boring.
If the loaded shoulder moves as fingers come off, go back a step.
One arm, tuck or straddlePick 1 of 2Free hand off, legs in a tuck or a wide straddle to slow the whole thing down.
Picture it. Fully stacked over one hand, the other lifts off. Legs tucked or wide in a straddle, which makes the body slower to rotate and buys time to correct.
Prerequisites, honestly: a 60 s freestanding hold, a reliable pirouette exit, and the block work above holding for several seconds. Without those this is falling practice.
Remember
Straddle wide. Straight legs are the last rung.
Bail with the pirouette from Phase 5.
Prerequisites not met? Go back.
Strength, not preparation
Loaded wrist workStanding, palms on a bench, roll the bodyweight out over the hands and back.
Picture it. Stand in front of a bench with your palms flat on it, fingers pointing forward, arms straight. Now roll your bodyweight forward over your hands until your shoulders are past your wrists and you feel the load through the heels of your palms, then roll back. Ten slow ones. To make it harder, use a lower bench so more of you is over your hands.
New, and it is the highest-value addition to the handstand. The wrist is the balance joint: across the biomechanics literature gymnasts most often balance a handstand with a wrist strategy, the way you hold standing balance at the ankle. When it fails, the shoulders, hips and elbows take over as a mixed strategy. (MacDonald 2025 systematic review, 21 studies, PMID 41473027. No percentage attaches to "most often" — the earlier "more than 75% of the time" here was not sourced and has been removed.)
The wrist flexors carry the highest muscular demand of any group measured in a handstand: about 60% of the reference contraction on average and 200% at peak. They are doing the balancing, and they need to be strong, not just warm. (Kochanowicz 2018, Hum Mov Sci 58:195–204, PMID 29471194 — 10 adult and 15 young gymnasts.)
The wrist flexors work hardest of any muscle group measured in a handstand — that ranking is the point, not the exact number.
Standing, hands on a bench. Rock the bodyweight forward over the palms and back, slowly. Add load by lowering the bench, not by speeding up.
Do it away from the handstand session too. This is strength work, not preparation.
Remember
Slow. Two seconds out, two seconds back.
Add load by lowering the bench, never by speeding up.
Fingers stay flat and spread. No clawing.
On days you are not doing handstands — it competes with the skill work.
Phases 1 to 5 are skill work and can run any day, including every day. Phases 6, 7 and 8 carry real strength work and get three named days a week. Never Monday, never Friday. Two mornings running with stiff or sore wrists means three days off the hands, whatever the phase.
DayHandstand block may do
MBench, dips, holdsSkill only
TGround flow, no upper load6 · 7 · 8
WVO₂max intervalsSkill only
TPull-ups, row, holds6 · 7 · 8
FRestSkill, or nothing
SLower body7 · 8
SDeep flexibility, splits6
Elbow rehab, temporary: phases 6 and 7 are paused until the dips rung passes — the Elbow Governor holds the ladder. Phases 1–5 skill work continues; the handstand session entries carry their own elbow tag.
Phase 6 is Tuesday, Thursday and Sunday; phases 7 and 8 are Tuesday, Thursday and Saturday. One of them at a time, not two. Handstand push-ups compete with Monday’s bench, so they sit as far from it as the week allows; the press competes with hip and hamstring range instead, so it takes Sunday alongside the split work. Thursday’s goes in the evening, not bolted onto the gym session.
On the wrist rule: in 321 adult handstand practitioners, weekly training hours showed no association with chronic wrist pain (p = 0.455), and neither did warm-ups or braces — but 57% had chronic wrist pain. A null in a self-selected survey is not a green light, and the authors read the risk as sitting in intensity rather than volume. That is why frequency is generous here and intensity is not. (Martonovich 2025, J Funct Morphol Kinesiol 10(4):372, PMID 41133562.)
PhaseSessionStay
·Warm-up, every session3 min—
1Look12 min1–2 sessions
2Line13 min4–6 weeks
3Entry11 min3–5 weeks
4Balance12 min8–12 weeks
Phases 1–411–13 min4–6 months
Then, in no fixed order
5Robust13 minongoing
6Press13 min6–12 months
7Push-up12 min3–6 months
8One arm12 minyears
Session times include the three-minute warm-up and are computed the same way the rest of this file computes a day: work, plus rest between sets, plus a set-up allowance per exercise. They match what the timers actually count. Anything marked optional sits outside them. The calendar figures are arithmetic on the gate criteria at four to six sessions a week, not measured — the gates decide, not the clock. Phases 5, 6 and 7 can run in any order once Phase 4 closes; 8 needs 5 first.
Confidence in the calendar column, stated plainly. Phases 1 to 4 are skill and their weeks are a reasonable guess — the gates close when they close. Phase 7 at three to six months assumes ten minutes a day is enough pressing volume to drive a strength adaptation, which is the marginal call in this table. Phase 6 at six to twelve months is the least reliable figure here: presses are limited by compression and hip-flexor range, which move slowly, and I have no data on adult learning rates for it. Nobody has published a trial on any of this. Treat the column as planning, not prediction.
Reference
▲8 blocks · tap one to open it
What this should feel like
▲Fatigue expectations · scaling up and down
A hard week you recover from is training. A hard week you do not is debt.
The full version (everything on, including Thursday's drag and leg press, about six hours a week) is the ambition, not the starting point. Scale toward it. The point of knowing what right feels like is so you can tell the difference between a hard week and too much.
Per session
Monday. Finish capable, not crushed. If you are wrung out, you have taken Wednesday's session with you.
Tuesday. Feel better after than before. If it feels like work, it is being done too hard.
Wednesday. The last interval brutal but controlled, not survived. This is the hardest session of the week and it stays that way.
Thursday. Pull-ups should feel good. If they do not, Wednesday was too much or recovery is short.
Saturday. Legs trained, not irritated. The next morning is what counts, not how the sets felt.
Sunday. Worked, never threatened. Restorative, not a seventh training session.
Across the week
By Friday you should want the rest day, not desperately need it.
Wednesday is the canary. You should be able to hit it at full effort every week. When the programme is too much, VO₂max power drops before anything else tells you.
Sleep normal. Waking unrefreshed two or three mornings running is a signal, not a phase.
Appetite and mood steady. Dreading a session you normally like is data.
The left leg fatigues differently
It does not burn. It loses position and control. When the left side starts drifting, or the lower back begins taking over earlier in a session than usual, that is its fatigue signal, and it arrives before you feel tired.
Stop the set, not the session. Pushing through here buys nothing, because the tissue that would adapt cannot be reached.
Scaling down, in this order
1 · Zone 2: Thursday elliptical, optional Monday cycling, Saturday bike
2 · Thursday's leg press, then Thursday's drag
3 · Downscale what is left before cutting it: less weight, fewer sets, fewer reps
4 · Thursday's flexibility block
5 · Monday push volume
6 · Accessories: Saturday abduction, adduction, calf, push-ups · Monday hamstring curl
7 · Sunday duration, keeping the quality
Never Monday's tracked leg press, never Saturday's squat. Never Wednesday, unless symptoms persist two weeks or more.
Scaling up
Only after two consecutive weeks in the green on the weekly traffic light. One variable. Never two, and never depth and load in the same week. Knee-work progressions are governed separately, by the Knee Governor — see What counts as a green week there.
Order of return is the reverse of the list above: Zone 2 last, not first.
Weekly Decision Rules
▲Traffic lights · decide once per week
Think in traffic lights. Decide once per week (or once per day if needed). You do not negotiate with yourself mid-session.
🟢 Green: push (or progress slightly)
Green if ALL are true:
Wednesday VO₂max: all intervals complete, last one brutal but controlled
Saturday legs: no unusual heaviness, no knee or adductor irritation
Sunday flexibility: you can sink into range without guarding
Morning resting HR and sleep feel normal
“Push” means, pick ONE:
A small load increase on the Monday leg press — the tracked set · knee
A little more depth on the back squat: pins one notch lower · knee
+2–5 W / slightly higher pace on Wednesday
+1 rep per set on one strength lift
+5–10 s longer split holds on Sunday
Never push more than one variable per week, and never depth and load in the same week. The two marked · knee are not this list's to give: knee-work progressions are governed by the Knee Governor and require its green week, not this one. A traffic-light green with no governor green means pick one of the other three.
🟡 Yellow: hold
Yellow if ANY are true:
Monday push feels harder than it should
Legs feel flat or dull on Saturday
Flexibility feels stiff but not painful
Sleep or stress is slightly off
“Hold” means: keep exact same loads, reps, times. Do not add volume. Keep structure intact.
This is where most progress actually consolidates.
🔴 Red: pull back (strategically)
Red if ANY show up:
You dread Wednesday VO₂max
Power or pace drops >5–8% mid-session
A knee red on the Knee Governor’s own criteria — do not re-judge the knee here. The governor takes over and the knee work steps back one increment; the rest of this list is unaffected
Sunday flexibility feels guarded or threatening
You want to ‘just get it done’
“Pull back” means, in this order:
1 · Remove Zone 2: Thursday elliptical, optional Monday cycling, Saturday bike
2 · Thursday's leg press, then Thursday's drag
3 · Downscale what is left before cutting it: less weight, fewer sets, fewer reps
4 · Thursday's flexibility block
5 · Monday push volume
6 · Accessories: Saturday abduction, adduction, push-ups, calf raises · Monday hamstring curl (not squat, not dumbbell RDL)
7 · Sunday duration, keeping the quality
Never Monday's tracked leg press, never Saturday's squat. Do not remove Wednesday VO₂max unless symptoms persist 2+ weeks.
When two of these disagree
Systemic fatigue overrules everything. If the whole system is tired, that is when ordinary movement injures you, and the week comes down regardless of what any single joint says — but it governs the week's volume, not the Knee Governor's increment. If the system is tired and the knee is red the same week, the week comes down and the knee work still steps back one. The override decides precedence, not whether the second rule applies.
Otherwise a knee red overrides this list: the knee work steps back one increment even though the general rule protects it.
A challenged knee with a fresh system is not a reason to lose the session. Do as much of the rest of the workout as it allows.
Non-negotiableNever “make up” a missed session.
Knee Governor
▲Judged on the next morning, not on the set
Every exercise carrying the knee tag is governed by this, with one stated exception: Wednesday’s intervals carry knee · tracked and are attributed, not capped — no ≤2/10 rule, no ceiling, just the next-morning number marked Wed. One number decides: how the knee feels the next morning, rated before you get out of bed. Mild symptoms that settle overnight are expected. They are the signal you are at the right dose, not a warning.
🟢 Green
Next morning back at baseline · no rise across the week
Progress one variable: leg-press load, or squat depth, or a rep. Never two.
🟡 Amber
Next morning mildly up but settled within the day · no week-on-week rise
Hold. Repeat the same session at the same numbers. This is where it consolidates.
🔴 Red
Next morning still up at 24 h · or rising week on week · or new morning stiffness three mornings running
Step back one increment. An increment, not the whole thing.
Stiffness counts, not just pain, and it usually arrives first — but it follows the same rule as pain: one stiff morning is noise, three in a row is signal. Do not act on one.
What counts as a green week
One definition, and it is the green box above: next morning back at baseline, no rise across the week. It is judged once a week, not after each session — it was always a weekly rule. Every green-week count in the programme uses this one — the two weeks before Thursday’s leg press opens, the two in the exit gate, the four before single-leg right. The weekly traffic light is a separate, whole-programme judgement about whether to push; it does not feed these counts.
A green week buys load, or depth, or a rep. One of them, never two, and never depth and load together. When both depth and load are eligible, depth takes precedence — depth is what the deep-range design is buying, and load can go on the following week. Set count is not a progression variable anywhere in the knee work. (Decision rule, not a finding; nothing in the literature ranks them.)
“Baseline” is not set yet. It is derived, not declared — read off the first three weeks of next-morning numbers and written down at the first monthly review (Debrief Protocol §5). Until it is set, the first line of each band above is read as no rise across the week, which needs no baseline, and exit-gate criterion 3 cannot be assessed. A session that hurts materially more than usual is said at the time, in words, in Flag — it does not wait for the next morning and it does not get a number. The escalation list below is separate and always live.
Which rule wins
Systemic fatigue overrules everything, including this — but it governs the week's volume, not this rule's increment. If the system is tired and the knee is red in the same week, the week comes down and the knee work still steps back one increment. The two point the same way; the override decides precedence, not whether the second rule applies. Protecting Monday's tracked leg press and Saturday's squat from the scale-down list protects them from removal, not from a load or depth step-back.
Below that, a knee red overrides the weekly traffic light, which otherwise protects the knee work.
A red knee is not a lost session. Do as much of the rest of the day as the knee allows.
Tracked, but not capped
The capped knee tag covers seven elements across three days — drag and leg press on Monday, Thursday and Saturday, plus Saturday’s squat. Three exercises, seven exposures. It does not cover most of the knee bending you actually do: Wednesday’s intervals, Zone 2 on Monday, Thursday and Saturday, Sunday’s horse-stance pulses and side lunges, two walks a day and two dance blocks a day. None of that is being capped or cut. All of it is being attributed.
Sunday earns its place on this list rather than sitting outside it: a side lunge through 40–100° loads the kneecap joint more than a forward lunge, and any one-legged position loads it 2.4–2.7× a two-legged one. Shallow and slow is what keeps it cheap. If a Sunday night is followed by a raised Monday morning, write sunday next to the number.
Write the day’s aerobic content next to the next-morning number — one word is enough: elliptical, bike, long walk, dance.
Three consecutive rises that line up with an aerobic day and not with a tagged day: change the machine, the resistance or the duration first. Nothing tagged steps back on that evidence.
This is a record-keeping rule, not a restriction. There is no PFJ stress measurement for an elliptical in the literature, which is a reason to log it rather than a reason to assume it is free.
When this ends
Three conditions, all of them. This is the only exit gate that exists, and its wording matches the debrief protocol word for word, so there is one gate and not two:
The tracked Monday leg-press load has increased on three consecutive progressions, at full range, at ≤2/10, without a red in between. (“Meaningful load” had no definition until 8 August 2026 and was the one governing threshold in the system carrying no number. This one is read off what you already log rather than invented.)
On the monthly test, single-leg decline squat onset is at parallel or deeper, with pain ≤2/10.
Next mornings at baseline for two consecutive weeks.
Then the ≤2/10 rule comes off, the knee tags come off, and the leg press progresses like any other lift. It stays in the week permanently. There is no stage after this one, and no end date.
Raise it directly if
Swelling, locking, the knee giving way, or pain at night or at rest
No meaningful improvement after 6–12 weeks of loading done properly. Counting from 10 August 2026, that window is 21 September to 2 November 2026. Put both dates in the calendar now. At the first, ask the question; at the second, it stops being a question.
Anything anterior in the left knee. Your right quadriceps is normal and the left is not, so the two knees are not carrying the same risk — but the left is loaded three times a week and has never been symptomatic. A watch item: worth raising if it appears, not a prediction that it will.
If it hurts on a given day
A tailored medial patellar tape is the one thing with good evidence for taking pain down on the day. It does nothing lasting and is not treatment: use it to get through a session, not as part of the plan.
Isometrics before the session are optional and cheap. The best replication got about 1/10 and it did not hold. Use it if it helps you, drop it without regret if it does not.
One-Page Reference Sheet
▲The week at a glance
Monday: Push
Pair: bench ×5 / handstand ×4 → batch: lateral raise, hip flexion
Rotation: curls, hip thrust, dips → drag, leg press, hamstring curl knee · optional Zone 2 cycling (cut first)
Rule: must finish feeling capable, not crushed
Tuesday: Mobility + Balance
Ground: breathing → spinal rocking → psoas activation (left) → dead bug → wall press
Hips + spine: figure-four → seated hip CARs → thread the needle → bridge (if the low back is quiet)
Balance: single-leg stand → split-stance shifts → tandem stance
Close: free movement, non-negotiable · Rule: feel better after than before
Wednesday: Primary VO₂max
Warm-up → 16 min ON (hard, controlled) → cool-down · Rule: highest-priority session
Thursday: Pull + Zone 2
Batch: pull-ups / crossover / handstand → alternating: farmer carry / single-arm row
Batch: back extension, side plank, leg raise → drag, leg press kneeafter 2 green weeks → flexibility · Zone 2, no added weight · Rule: no HIIT
Friday: Recovery
Full rest or gentle movement only · Rule: do less than you want to
Saturday: Leg Strength
Batch: swing, calf raises → drag, squat, leg press knee → dumbbell RDL, push-ups
Rotation: abduction, adduction
Optional light Zone 2 · Rule: legs should feel trained, not irritated
Sunday: Deep Flexibility
Prep → front split (one cycle each leg) → middle split (one round) → pancake
Spine: Jefferson curl → bridge · close: savasana, legs supported · Rule: worked, never threatened
Elbow Governor
▲Temporary · judged before each rung, not on the calendar
Right triceps MTJ strain, 10 August 2026, aggravated 12 August; day zero 12 August; grade-2 working read — a read, not a diagnosis. The left arm is the reference side. Every exercise carrying an elbow · out or elbow · light tag is governed here, and tags change state here — one place to update as rungs pass, not eleven. Decision source: the 12 August integration plan, ratified the same day. The evidence behind every rule sits in Elbow Evidence Record.md; the operational mirror is in the debrief protocol.
The gate, before every rung
Pain-free resisted elbow extension at the current effort level, judged at the weekly rung decision — the same moment as the knee governor’s weekly check. The dates below are earliest-plausible only. Radiological grading predicts return time poorly — 7.6–11.9% of the variance (PMID 29460072) — which is why the criteria govern and the calendar does not.
The ladder · one rung a week, nothing skips
Now–15 Aug · inflammation: unloaded ROM only, nothing loading elbow or shoulder extension
16 Aug → early repair: isometric holds enter on the low-effort gate; pressdown follows once they are pain-free at moderate effort
1 · Bench at reduced load, 3 s lowering — 24 Aug
2 · Elevated push-ups, Saturday — 31 Aug · the Sunday wall walk-down follows at light
3 · Pulldown — 7 Sep · an ordinary lat pulldown, a gated exposure rather than a programme entry. The ski-pole machine is retired outright: a one-off tried 12 Aug as a VO₂ stimulus, the aggravating exposure, not prescribed, not returning
4 · Pull-ups, and the daily bar block with them — 14 Sep · the overhead extension enters here too
5 · Handstand, holds only, practice phases 1–5 — 21 Sep
6 · Dips — 28 Sep · phases 6–7 follow after this rung, and the pressdown retires or converts
What advances, what steps back
Advance: the gate passes and the previous rung ran a full week without flare.
A flare during or the morning after a rung → repeat the week. A flare that persists → drop one rung.
The temporary next-morning elbow number, 0–10, rides beside the knee number in the debrief and is what the weekly call reads. It is taken from what you share — logged, not chased — and it retires at the exit gate.
Which rule wins
Systemic fatigue overrules everything, including this — but it governs the week’s volume, not this rule’s rung. If the system is tired and the elbow flares in the same week, the week comes down and the rung still repeats or steps back. The override decides precedence, not whether the second rule applies.
The two governors are parallel and independent. A knee red and an elbow flare in the same week are handled separately, each by its own rule.
The rehab work
Daily ROM, the isometric holds, the pressdown and the overhead extension live in the Elbow rehab block beside the daily practice sections, each with its entry condition stated on the exercise.
When this ends
Pain-free resisted elbow extension at full effort, matching the left arm, on two consecutive weekly checks, with every rung back at working load and no flare in the preceding two weeks. Then the elbow tags come off — except elbow · watch on dips and handstand pressing (practice phases 6–7), which persists until the first monthly review after exit.
This wording matches the debrief protocol word for word, and those are the only two copies. If one is edited, edit both in the same session.
Full-effort symmetry is deliberately stricter than field practice — 67% of “clinically recovered” professionals still carried >10% strength deficits at return (PMID 24493666) — and it is an operational definition, not a validated threshold; none exists. The watch tag outlives the gate because tendon-side involvement carries moderate evidence of higher re-injury risk (PMID 28259847), and in the one athlete series 1 of 10 partial triceps tears under continued heavy load progressed to complete (PMID 14977669).
At exit: this block and the Elbow rehab block come out in the same session the protocol’s elbow rows retire. The evidence record stays, as record.
Raise it directly if
A palpable gap, inability to extend against any resistance, or significant swelling → assessment, not waiting. Complete ruptures are repaired within ~3–4 weeks of injury (PMID 23016111), so that window is not free.
Load Caps
▲Two real, one milestone
Two of these are real constraints and one is a milestone dressed up as a limit. Which is which is marked below: a cap you cannot justify is a cap you will quietly ignore, and then you will not trust the ones that matter either.
Bench press: 100 kg is a milestone, not a cap
You asked whether 100 kg is a good ceiling. For your stated goals, strong and lean rather than big, there is no evidence that an absolute bench ceiling protects anything. Shoulders are hurt by volume, speed and position, not by the number on the bar. So treat 100 kg as the number worth reaching, roughly 1.2× bodyweight and a solid lifetime strength marker, and keep going past it if the joints stay quiet.
Progress via reps up to 8, 1–2 s pauses on the chest, 3–4 s eccentrics: all of which build strength without adding absolute load.
During the elbow rehab the bench runs at the Elbow Governor’s rung-1 reduced load with slow lowering.
Stop the set for bar speed dropping, shoulders shrugging off the bench, or any need to grind, not because of the number.
Back squat: hard cap 90 kg, and this one stays
Kept, for a reason specific to you rather than a general one: you were safe at 60 kg, hurt yourself at 70, then avoided the lift for five months. 90 kg is about 1.1× bodyweight and sits well above where you will be for a long time, so it is not the binding constraint and is not meant to be. It is a backstop for the day you feel invincible.
The research supports a ceiling on one of load or depth, not both: patellofemoral stress rises with each and the two interact (Zavala 2021, trained lifters — female-only; the equivalent study in men does not exist). Depth is the one you are deliberately reclaiming, so load is the one that gets capped.
Progress via depth, 4–5 s eccentrics, up to 6 reps, cleaner knee tracking.
The real governor is the ≤2/10 rule and the next morning. No forced symmetry.
Dumbbell RDL: working load 20 kg/hand
Governed by back protection and left-side compensation, not a load ceiling: no cap to chase
Progression: slower eccentrics · paused hinges · more reps (up to 8) · better posture and breathing · load increase only if lumbar stays completely neutral AND left side shows no end-range compensation, and only in 2.5 kg increments
No max testing · no load increase if left side compensates · no grip-limited failure
Pull-ups: no cap. The weekly ceiling has been removed.
You asked why the ceiling existed. The honest answer is that it was convention, not evidence: a volume cap of the same family as the 10% rule, which was tested head to head in a 532-person randomised trial and had no preventive effect at all (Buist 2008, 20.8% vs 20.3% injured). I have removed it.
What replaces it is a response rule rather than a number: every set stays at least 2 reps in reserve · add volume by adding sets, never by pushing sets closer to failure · two consecutive tender-elbow mornings means back off for a few days.
Goal 20 strict, 30 aspirational. Separately: 2 min dead hang, from about 1 min now.
Carries · cap: bodyweight total load
Progress via longer clean posture, better breathing, less grip fatigue
Accessories (rows, raises, machines)
If adding load reduces control or increases recovery cost → stop. Accessories never justify fatigue spillover.
Where the block sits
▲Week 1 · deload at week 10
This is week 1 of a build-up, not a deload. The legs are starting light on two lifts because of the knee: that is a starting position, not a downshift, and everything else is progressing normally from here.
When to deload
Week 10. Put it in the calendar now rather than deciding later.
Or immediately if: VO₂max power drops · pull-ups stall >3 weeks · knees or psoas feel flat · sleep quality declines
First deload: week 10, counting from 10 August 2026.
Deload week rules
Strength: reduce all loads 15–20% · same exercises, reps, order · no new progressions
VO₂max: keep Wednesday session · reduce ON time ~30% (e.g. 16 → ~11 min)
Zone 2: keep as written, no pushing pace
Tuesday + Sunday: unchanged · preserves range during the strength downshift
Pauses the count: a deload week neither resets a consecutive-green-week run nor counts toward one. The count picks up where it left off the following week, provided the next mornings stay at baseline. A deload week cannot be evidence of tolerance under load, because it is not load. It does not buy a progression either. This matters because week 10 falls inside both windows and pushes single-leg right roughly a week later against a fixed 2 November date.
Success Metrics
▲Quarterly · every 12–16 weeks
Metrics reflect capability, not ego.
1 · Strength expression
Max strict pull-ups: one set, perfect form, full hang, no kipping. Direct bodyweight strength, joint-honest. Goal 20, aspiration 30.
Longest dead hang: one hang, active shoulders. About 1 min now, goal 2 min.
Longest freestanding handstand: goal 2 min.
All three are collected at the monthly review, so the quarterly read is a series rather than a single day.
Longest dead hang: one hang, active shoulders. About 1 min now, goal 2 min.
Longest freestanding handstand: goal 2 min.
All three are collected at the monthly review, so the quarterly read is a series rather than a single day.
2 · Conditioning (longevity)
Power or pace in final VO₂max interval: same machine, protocol, resistance. VO₂max is the strongest single mortality predictor.
3 · Movement freedom (disability-aware)
Joint-confidence score, 1–5 (1 = guarded, 3 = neutral, 5 = confident and fluid): knees · hips/psoas · spine · shoulders. Collected at the monthly review, not weekly. Early warning beats pain.
4 · Knee, monthly
Single-leg decline squat: how bad at its worst, and where in the descent it starts: quarter, half, parallel, deep, or not at all. Onset depth is the sensitive measure and it moves before the pain number does.
Plus the week's rough unloaded / loaded / max. Symptom baseline 7 Aug 2026: 2 / 4 / 5 — this is the decline-squat and week-pain baseline, a different quantity from the next-morning baseline the Knee Governor runs on, decline squat 3–5/10, onset partway down.
Plus the week's rough unloaded / loaded / max. Symptom baseline 7 Aug 2026: 2 / 4 / 5 — this is the decline-squat and week-pain baseline, a different quantity from the next-morning baseline the Knee Governor runs on, decline squat 3–5/10, onset partway down.
Single-leg right leg press, one set. The load that sits at ≤2/10 at the same depth, written beside the current two-legged load. The ratio between them is the read on what share of the bilateral load the right leg is carrying — the one thing the two-legged press cannot tell you. One set a month. This is the measurement and it runs every month regardless of any gate. The separate question of when single-leg right enters the session proper is the Knee Governor's, not this list's.
5 · Flexibility / range
Not measured. Range is visible without a tape measure, it moves too slowly for a number to change a decision, and if something stops working it will surface in a debrief before any measurement would catch it. The test that stands: enter the deep ranges without guarding. Quality > centimetres.
Post-Cap Expression Phase
▲Activates when a primary lift hits its cap
Begins immediately when bench = 100 kg or back squat = 90 kg. The dumbbell RDL is governed by back protection and left-side tolerance, not a cap: it does not trigger this phase. From that point: no additional load on capped lifts, ever. Progression switches from load to expression.
At 40+, the strength gains that matter are cheaper to express, more repeatable, less injurious, more transferable. This phase trains neural efficiency · tendon tolerance · coordination under load · longevity-compatible strength.
Core rules
Same exercises · same weekly structure · same loads for capped lifts
No grinding · no testing · no chasing fatigue
If it feels hard in a new way: you are doing it wrong.
Bench press: expression mode
Load fixed (≤100 kg). Rotate, never stack: paused reps (1–2 s) · slower eccentrics (3–4 s) · more reps (up to 8) · fewer warm-up sets for same working load
Back squat: control mode
Load fixed (≤90 kg): slower eccentrics (4–5 s) · cleaner depth (never forced) · longer isometric pause above bottom · more symmetrical tracking
Dumbbell RDL: back-protection mode
Starts at 20 kg/hand. Increases only with perfect lumbar neutrality and zero left-side compensation: slower eccentrics · paused hinges · more reps (up to 8) · better breathing under load
Pull-ups: primary strength driver
Uncapped, and no weekly rep ceiling either: see Load Caps for why that was removed. Quality-governed: more total clean reps · higher density · longer isometrics · every set 2 reps in reserve. Becomes the main visible strength signal once barbell loads are capped.
Unchanged in this phase
VO₂max structure · Zone 2 · Tuesday mobility · Sunday deep flexibility: exactly the same. Deloads continue on the same week-10 cycle. No end date.