Protocols
Almost nothing in the movement library has a paper behind it. A back squat was not published by anyone — it just exists, and saying otherwise would be the first dishonest thing on this site. Five protocols are different: each is named, published, and linked here to the trial or to the body that publishes it. Fifteen movements out of 228 come from them. Everything else carries no source, and that absence is the honest answer rather than a gap.
N=1, not medical advice. These are other people’s trials, run on other people. None of it was tested on me, and a protocol that reduced injuries across a squad says nothing about what it will do for one person.
Last updated 14 September 2026.
What a source has to clear
A movement gets a source only when there is a named published protocol behind it and a link that resolves to a primary or near-primary source — the body that publishes the protocol, or the trial itself. Not a blog, not an aggregator, not a content farm.
No movement here is attributed to a person as an endorsement. A search for documented athlete routines returned aggregator pages with no primary source behind any of them, so none of it is on this site and none of it will be.
Evidence is graded strong, moderate, contested or untested. Contested is a real grade and it is used below: one of these five is genuinely disputed in the literature, and burying that would defeat the point of having grades at all.
FIFA 11+
football — also trialled in basketball · evidence: strong
A 20-minute warm-up published by FIFA’s medical research centre. Three parts: running, then strength / plyometrics / balance, then running with changes of direction.
A 2017 systematic review and meta-analysis of the FIFA 11 and 11+ programmes found a reduction in overall injury rate in football. A separate cluster-randomised trial in elite male basketball players found the same programme reduced injuries in a sport it was not designed for.
The trial · A second trial · The protocol itself
In the library — 9 movements:
- Nordic hamstring curl — called “Hamstrings, advanced” there
- Walking lunge — called “Walking Lunges” there
- Bodyweight squat — called “Squats with toe raise” there The 11+ version adds a calf raise at the top.
- Nordic curl, assisted — called “Hamstrings, beginner” there
- Plank — called “The Bench” there
- Side plank — called “Sideways Bench” there
- Side plank with hip dip — called “Sideways Bench — raise and lower hip” there
- Box jump — called “Box Jumps” there
- Lateral bound — called “Lateral Jumps” there The 11+ version is continuous for 30 seconds rather than sets of bounds.
Nordic hamstring protocol
football · evidence: contested
Partner- or strap-anchored eccentric knee flexion. Lower under control, catch yourself, push back up.
A 2019 meta-analysis across 8,459 athletes reported that injury-prevention programmes including the Nordic hamstring exercise halved hamstring injury rates. A 2021 methodological reappraisal argued the pooled effect is inconclusive because of how the original analyses were constructed. Both positions are live; this is not a settled question.
In the library — 3 movements:
- Nordic hamstring curl
- Slider hamstring curl Same eccentric emphasis, different implement. The trials used the partner-anchored version.
- Nordic curl, assisted
Adductor Strengthening Programme
football · evidence: moderate
The Copenhagen adduction exercise, built up over weeks from a short lever to a long one.
A cluster-randomised controlled trial in male football players found the programme reduced the prevalence of groin problems across a season.
In the library — 2 movements:
- Copenhagen plank, short lever — called “Copenhagen adduction, short lever” there
- Copenhagen plank — called “Copenhagen adduction, long lever” there
Norwegian 4×4
endurance science — NTNU, Trondheim · evidence: strong
Four intervals of four minutes near maximum, three minutes of active recovery between them.
The 2007 trial compared four training protocols at matched work and found high-intensity 4×4 intervals raised VO₂max substantially more than moderate continuous training. VO₂max is among the strongest modifiable predictors of all-cause mortality.
No movement in the library carries this as a source.
Heavy Slow Resistance
sports medicine — tendon rehabilitation · evidence: moderate
Heavy loading through a full range at a deliberate tempo, roughly three seconds up and three seconds down. A loading scheme rather than a movement.
A randomised controlled trial in Achilles tendinopathy found heavy slow resistance produced outcomes comparable to eccentric training at twelve weeks, with better patient satisfaction and far lower time cost.
In the library — 3 movements:
What isn’t working
- One of the five is contested, and it is the one with the most movements attached to the idea. The Nordic hamstring protocol has a 2019 meta-analysis reporting halved hamstring injury rates and a 2021 reappraisal arguing the pooled effect is inconclusive. Both are live. Three movements in the library point at it, and the grade on every one of them says contested rather than hiding behind the meta-analysis.
- Two of the five have no trial behind the specific movements drawn from them. Heavy Slow Resistance was trialled in Achilles tendinopathy; the three movements here apply its loading scheme to a leg press, a calf raise and a split squat, which is not what was tested. The Norwegian 4×4 has no movement at all — it is a cardio option in the builder, not something the library can draw, so nothing carries it as a source.
- The trials tested programmes, not exercises. The FIFA 11+ and the Adductor Strengthening Programme were trialled whole. Nine movements in the library are exercises from the 11+, but drawing one of them into a session is not running the 11+, and no result from those trials transfers to a movement pulled out on its own.
- Two of the mappings are approximate and say so. A bodyweight squat is not the 11+’s squat with a toe raise, and a set of lateral bounds is not its thirty seconds of continuous lateral jumps. Where the movement is not the same thing, the line under it says what differs rather than implying they are equivalent.
- Fifteen out of 228 is a thin layer. Most of what gets trained here has no published protocol behind it, and adding more sources would mean lowering the bar rather than finding more evidence. Functional Range Conditioning, the knees-over-toes material and the Jefferson curl were all considered and left unsourced: named systems, no located trials. The movements stay; the citations they do not have stay absent.