‹ Workout Coach

How are workouts decided?

Short version: the research and the rules decide; the AI only writes the words.Here's where the plan actually comes from — including what the science is not sure about.

Where this comes from

The coach runs on an evidence review built from the major meta-analyses of resistance training — studies that pool dozens of individual trials and thousands of participants, instead of cherry-picking one flashy result. Each claim was checked against the original paper, then graded: S strong · M moderate · W weak. The graded rules are compiled into the planner's code— they aren't suggestions the AI may or may not remember. And the coach is required to hedge anything below S when it talks to you, because a plan you can trust includes the uncertainty.

The studies doing the heavy lifting

Watch a paper become your workout

The whole pipeline for one example — a muscle-building goal, bench-press day:

  1. 1

    The papers

    Pelland 2026 ↗ — weekly sets drive growth, with diminishing returns. Robinson 2024 ↗ — closer to failure helps growth; strength doesn't care.

  2. 2

    The graded rules they became

    M#3Use ~10 sets/muscle/week as a productive hypertrophy target, but as a soft anchor on a continuous curve, not a hard cutoff.

    S#7Do not require training to momentary failure; program hypertrophy at ~0–3 RIR and strength at ~2–5 RIR (load-first).

    Below-S grades force the coach to hedge when it talks to you — that's enforced, not stylistic.

  3. 3

    The engine's numbers — live, not typed

    Weekly sets (bench pattern)

    6–16, aim 10

    rules 1 [S], 3 [M] — hypertrophy is volume-driven

    Reps

    6–12

    rule 5 [S] — 6-15 default for time efficiency

    Effort

    1–3 in reserve

    rule 7 [S] — hypertrophy 0-3 RIR

    These values are computed by the same functions the week generator runs, imported into this page — they cannot drift from behavior.

  4. 4

    What lands on your Today screen

    lift · anchor

    Barbell Bench Press

    5 sets · 6–12 reps · leave 1–3 in the tank

    The weekly 10 splits across two sessions; no session carries more than 5 sets of one movement — overflow becomes a complementary second exercise instead of junk volume.

Change one input and the machine answers differently. Switch the goal to strength and the same pipeline emits: 4–8 reps, heavier, 2–5 in reserve, and weekly volume capped at 8 sets (aim 5) — because rule #2 S says strength gains saturate early. Same evidence, different dial settings — no vibes anywhere in the chain.

The workouts you'll actually get, and why

When you tell it what you do for fun

Pick a hobby in Settings — surfing, snowboarding, skateboarding, aerial — and the plan leans its accessory choices toward what that sport asks of a body. The doses never move: the evidence rules above own every set, rep and effort number. And because sports science rarely studies surfers, these claims carry one extra grade: E expert practice — the research is silent (unlike W, where it disagrees), and the claim is a coach's call. The coach is required to say so when it leans on one.

  1. SFrame every hobby-support block as dose-dependent injury insurance first (Lauersen 2018).
  2. MSurf weeks: pulling volume + scapular-control accessories — the paddling shoulder is the sport's overuse site, and upper-body strength transfers to paddling (Remnant 2020; Coyne 2017).
  3. M/ESurf mobility finisher: thoracic rotation (named requirement, Furness 2016) + hips (practice).
  4. S/EBoard-stance sports: single-leg emphasis is stance specificity, not proven superiority — unilateral and bilateral train equally well (Moran 2021), so say "specific", never "better".
  5. MLanding-heavy variants (aerial surf maneuvers, snow, skate): keep real lower-body strength plus a small balance/ankle block (Secomb 2016; McKeon & Hertel 2008).
  6. S/MSnow/skate: wrist guards genuinely protect (RR ≈ 0.23, Russell 2007) — sayable as gear advice; beginners and first-weeks carry outsized risk, which the comeback-easing rules already encode.
  7. M/EAnti-rotation and rotational trunk work for board sports: a named requirement in surf literature, expert practice elsewhere.
  8. M/EAerial weeks lead with a shoulder built for hanging: pulling volume plus external-rotation and scapular-control accessories — shoulder is circus artists' most-injured site (Stubbe 2018), ER weakness and scapular dyskinesis predict overhead shoulder injury (Clarsen 2014), and a targeted ER/scapular programme cut shoulder problems 28% in an RCT (Andersson 2017); the handball-to-aerial transfer is a coach's call.
  9. M/EAerial wrists and grip get deliberate preparation — the wrist is a top-three circus-student injury site, and apparatus work is grip-limited; loaded carries and hangs are the practice.
  10. M/EAerial trunk work is tension before flexion — hollow-body holds and anti-extension work echo gymnastics practice, and the lower back is the number-two circus-student injury site.
  11. MFrame aerial as trainable, not dangerous: most circus injuries are minor and serious-injury rates run below NCAA women's gymnastics (Shrier 2009).
  12. MRespect freshness around aerial-heavy days: fatigue, exhaustion and prior injury raised circus-artist injury risk 1.8–2.8× (Shrier & Hallé 2011) — the freshness guardrails apply with extra teeth.
  13. SStretch training earns range: PNF and static holds beat ballistic for long-term ROM, and consistency matters more than dose details — volume, intensity and frequency weren't significant moderators (Konrad 2024).
  14. MThe lifting is itself mobility training: full-ROM resistance training improves flexibility, with higher intensities showing larger effects (Favro 2025) — a deep squat is a mobility exercise.
  15. SLong static holds go after training, never as the sole warm-up: pre-lift static stretching acutely blunts strength about 5%, worst past ~45 s per hold (Simic 2013).
  16. SMobility is for positions, not injury insurance: stretching showed no injury-prevention effect while strength training cut injuries to less than a third — say so honestly when an athlete stretches "to stay safe" (Lauersen 2014).
  17. MFoam rolling is an optional range tool — real but modest, acute and chronic (Wilke 2020; Konrad 2022); preference, not prescription.

Every citation in the hobby corpus was resolved against the research registry and its abstract read before the claim was written down — including the two that cut against us: single-leg training is notproven superior (we choose it for stance specificity and say so), and balance training's performance transfer is unclear (so the balance work stays small, justified by ankle-sprain prevention instead).

The research behind each sport

Every sport

Surfing

Snowboarding

Skateboarding

Race builds

Mobility & flexibility

Aerial arts

What the AI cannot do

The guardrails, with their numbers

Live values, rendered from the same constants the engine enforces — this page cannot claim something the code doesn't do.

The full rule list, graded (21 rules)tap to expand
  1. STreat weekly set volume per muscle as the primary hypertrophy dial; more sets = more growth with diminishing returns and no clear plateau.
  2. SCap strength-focused volume early — strength gains largely saturate around ~4 hard sets/muscle/week (functional plateau); push hypertrophy volume higher.
  3. MUse ~10 sets/muscle/week as a productive hypertrophy target, but as a soft anchor on a continuous curve, not a hard cutoff.
  4. SSet training frequency to whatever lets the athlete accumulate quality weekly volume; frequency itself adds little hypertrophy once volume is equated, and modestly helps strength.
  5. SFor muscle growth, allow any load from ~30–100% 1RM (≈6–35 reps) provided sets end near failure; default to 6–15 reps for time efficiency.
  6. SFor maximal 1RM strength, program heavy loads (≥80% 1RM), full ROM, 2–3 sets, strength lifts first, ≥2×/week.
  7. SDo not require training to momentary failure; program hypertrophy at ~0–3 RIR and strength at ~2–5 RIR (load-first).
  8. M/WEmphasize the lengthened/stretched position via full ROM (or lengthened partials as an equivalent option); do not claim partials are superior.
  9. MRest ≥2 min between hard sets for trained lifters; offer supersets/drop sets/rest-pause as time-savers that favor hypertrophy.
  10. MTreat periodization model as adherence-driven for hypertrophy; prefer undulating over linear for trained strength athletes; skip complex periodization for beginners.
  11. S/WDistribute endurance intensity as polarized *or* pyramidal — both work; reserve extra high-intensity interval work for trained athletes and short blocks.
  12. MAny structured aerobic training raises VO₂max; add HIIT for a modest, athlete- and beginner-dependent edge, not as a decisive lever.
  13. STaper 8–14 days before key events: cut volume ~40–60%, hold intensity/frequency; add a pre-taper overload block.
  14. SFrame cardio as a modest weight-loss aid (~0.5 kg per +30 min/week; ≥150 min/week for meaningful fat/waist change); diet leads.
  15. SPrescribe resistance training during any calorie deficit to preserve lean mass and increase fat loss, independent of scale weight.
  16. SSet age-adjusted step goals (~6–8k for 60+, ~8–10k for under-60; ~7k captures most benefit); prioritize total volume over cadence.
  17. MBuild adherence with action planning, self-monitoring, cues, and barrier problem-solving; reinforce past 12 weeks; avoid social-reward gimmicks.
  18. SAnchor general fitness to WHO/ACSM: 150–300 min/week aerobic + strengthening ≥2 days/week, dose-response, some-is-better-than-none.
  19. MSupport a minimal effective RT dose (~4 sets/muscle/week; single sets and "exercise snacking" for novices/older adults).
  20. M/STreat equipment as interchangeable for muscle and general strength — bands, machines, free weights, and bodyweight all transfer when volume and effort match; reserve heavy free-weight specificity for max-1RM goals.
  21. SEncode prescriptions as ranges, not false-precision setpoints — the evidence base is mostly short-term and on untrained subjects.meta-principle

And it shows its work

Every plan change is recorded with the rule that caused it, and every plan version is kept — "why did my Thursday change?" always has an answer. Ask the coach in Chat for the evidence behind any session and it answers from this review, citation and grade included — or says "the research doesn't cover that" when it doesn't.

Back to your plan