What's inherited
The five-day split — Legs, Chest, Back, Shoulders, Arms — is the same one from the original program, in the same order. So is the shape of each session: a heavy main lift followed by hypertrophy accessories in the same sitting, not spread across separate days. The paused and tempo-controlled reps that program used constantly — two- and three-count pauses on squat, bench, and deadlift — are still here, not replaced by anything fancier. Saturday's superset-format arm day (a curl variant paired with a triceps variant, three pairs) matches how the later weeks of that program were already run. Even the standard warm-up — five minutes on the rower, five minutes on a backward-incline treadmill walk — carries over unchanged.
Small textures that clearly worked stayed too — Svend press, 21-rep curls, Cossack squats, dumbbell pullovers.
None of this is guesswork. Every session type was rebuilt from a union of every time that kind of day was actually logged — not from any single week, which can understate what the real program was reaching for. Week 1's leg day had six exercises; Week 6's leg day, the fullest one on record, had thirteen. Rauta's Legs day is built from that fuller, ceiling version — the maximum across every instance, never a single snapshot.
Where this comes from
- The governing rule: Anatoli is the skeleton, not the ceiling — keep what worked, and rebuild the rest from how the program was actually run, not from a blank page.
- Every exercise choice comes from a full count of how often that movement actually showed up, across every logged instance of that session type — never from a single week's list.
What changed, and why
Nothing below is invented to sound scientific. Each of these fixes something specific that happened during the original program, and each is something that program never had at all.
Shoulder corrective work
Before: none — no direct work for the shoulder joint carrying the heaviest pressing load.
Now: face pulls on both pressing days, band pull-aparts, external-rotation pre-activation before you press, wall slides — direct work for the exact joint that was never addressed.
Real deloads
Before: no planned deload — recovery happened only when something forced it.
Now: two real deload weeks (4 and 8) that cut volume and effort together, on purpose, before a crash forces one on you.
Conditioning
Before: none, across either full run — despite better cardiovascular fitness being a stated goal all along.
Now: an optional Tuesday session (easy-effort work, then short hard intervals from week 3), sized down but never fully dropped through the two hardest weeks, plus a rowing-time benchmark so that goal is actually measured, not assumed.
Loaded holds instead of passive stretching
Before: passive stretching only — and often held long enough, right before lifting, to cost real strength on the working set that followed.
Now: mobility work before you lift stays short and either dynamic or loaded; anything needing a long hold moves to the end of the session instead. For hip and shoulder range specifically, loaded end-range holds replace plain stretching, which reliably gave range that never stuck.
Five more changes round out the list: training effort is prescribed as RPE, never a number of kilos; weekly volume is checked against published landmarks rather than just "more sets," including in a shortened four- or three-day week; hip and shoulder mobility — and now aerobic capacity too — are measured on a fixed, repeated test and tracked week to week, not left to feel; hip and shoulder flare rules are written down as an explicit stop-or-continue test, not a vague "if it hurts"; and session timing and pre-training fuelling are treated as programme content, not an afterthought — skipping a meal beforehand measurably costs a working set.
Full list, with sources
- 1. Shoulder corrective work — face pulls, band pull-aparts, external-rotation pre-activation, wall slides, band dislocates — direct work for the one joint carrying the heaviest pressing load, previously given none.
- 2. Hip corrective work, including loaded end-range holds (PAILs/RAILs) — 90/90, Cossack squat, banded hip distraction, couch and pigeon stretch. A 2024 meta-analysis of 83 studies found static holds of 60 seconds or more measurably reduce maximal strength when held right before lifting; practitioners report passive-only stretching gives range of motion that doesn't hold.
- 3. The primer/finisher split — mobility work before lifting stays under eight minutes total and under thirty seconds per bout; anything held 60 seconds or longer moves to the end of the session. Same evidence as above.
- 4. RPE-based effort, zero prescribed loads. A 2025 network meta-analysis found RPE-based autoregulation produces strength gains equivalent to velocity-based training, and that this method of judging effort is reliable specifically in experienced lifters.
- 5. A real eight-week wave with planned deloads at weeks 4 and 8. Rogerson, Bell et al., Sports Medicine – Open, 2024 — a real deload cuts volume and intensity together, not one alone.
- 6. Weekly volume checked against published landmarks (minimum, maximum adaptive, and maximum recoverable volume per muscle), including in a shortened week. Schoenfeld, Ogborn & Krieger, Journal of Sports Sciences, 2017.
- 7. Conditioning — an optional Zone 2 plus short hard-interval session, sized to the actual week rather than switched fully on or off. The interval protocol is Helgerud et al., Medicine & Science in Sports & Exercise, 2007.
- 8. A fixed benchmark test for hip and shoulder mobility, and a rowing time-trial for conditioning, both tracked week to week rather than judged by feel. The rowing benchmark comes from Ingham, Whyte, Jones & Nevill, European Journal of Applied Physiology, 2002 — a real, disclosed limitation: that correlation between rowing time and aerobic capacity was established in elite competitive rowers, extrapolated here as a personal trend marker rather than a population score.
- 9. Explicit flare rules for hip and shoulder pain — a dull ache is fine to continue through, a sharp pinch means stop.
- 10. Session timing and pre-training fuelling treated as programme content, not an afterthought.
The method: never a number of kilos
Every working set in the programme is a set count, a rep range, and a target RPE. RPE 10 means 0 reps left, RPE 9 means 1, RPE 8 means 2, RPE 7 means 3 — the higher the number, the fewer reps you had left, not more.
4 × 6 @ RPE 7 · tempo 3-1-1-0
No load written anywhere. You choose the weight that makes that rep range land at that effort, today, and log the actual number afterward — that's the only place a kilogram appears.
This isn't hedging. Training energy isn't constant — it moves with how hard the week's work has been, and drops hard around an extended fast, multi-day water fasts included. A fixed kilogram number is either too easy on a strong day or an invitation to grind through a bad one; a target effort follows you instead of the calendar. It's also close to how the original program actually ran — an RPE was already noted next to most working sets in those logs. The programme never has you lift during a fast at all — it drops to mobility and easy walking only, and waits for a real re-feed day before the bar comes back out.
Evidence and citations
- The RPE numbers themselves map to reps in reserve (RIR) exactly as stated above — Zourdos et al., 2016, Journal of Strength and Conditioning Research ("Novel Resistance Training-Specific Rating of Perceived Exertion Scale Measuring Repetitions in Reserve"), the study that formalized this scale from Tuchscherer's earlier practitioner version. RIR estimates are most reliable close to failure (RPE 8–10) and looser further out, which is why the app's separate mobility/primer effort scale (2–8) never uses reps-in-reserve language at all — a held stretch has no "reps left" to estimate.
- RPE/RIR-based autoregulation matches velocity-based training for strength gains, and is reliable in experienced lifters specifically — a 2025 network meta-analysis.
- Strength returns faster than tendon adapts after time off — the reason effort, not volume, is the lever capped early in a return to training: a 2020 study found strength recovering in about 8 weeks after a layoff, and a 2024 study found 5 weeks of retraining matched an original 10-week block.
- Fasted resistance training measurably blunts strength output — Trabelsi et al., 2023; a 2025 systematic review.
- Training under-fuelled measurably costs reps and sets once a session runs long — the published research behind that, not a single logged example, is cited on the Nutrition page's Fasting section.
What's judgment, not evidence
Most of this page points at a citation or a line in the source logs. Some of it doesn't, and it's worth saying which is which plainly, rather than dressing a coaching call up as settled science.
- Saturday is Arms and Sunday is real rest — not a research finding, a correction. Reading the source log closely, arm day usually landed on Saturday and only slipped to Sunday when the week ran late; an earlier pass copied the slip instead of the intent.
- Tuesday is where the optional conditioning work sits, and in the two hardest weeks a single hard interval session moves to Saturday instead. That placement is reasoned from how long a heavy lower-body day needs to recover before the next one — a considered call, not a specific study's result.
- If a week only allows four or three days, Arms is the first thing dropped, never Legs or Back — because those two carry the most spillover to every other muscle, not because a paper ranked them.