Transfer Of Training
Summary
Your body adapts to exactly what you ask of it — the specific movement, speed, joint angle, and contraction type you train — so strength and skill are partly task-locked and only partially carry over to anything you did not train; the practical upshot is not "train only the exact thing you care about" (that under-builds the base) nor "any training transfers to everything" (that wastes effort), but "build a broad general base, then bias the last layer toward the specific task that matters."
Why Strong
**The specificity principle — Tier 1.** That adaptation is specific to the trained movement, velocity, joint angle, and contraction type, with partial-but-real carryover, is supported by converging independent lines: a 43-study/1,660-participant meta-analysis with no funding and no conflicts (PMC12296774), classic velocity-specificity reviews (Behm & Sale), and robust angle-specificity work (Lanza 2019). This is settled physiology/neuroscience.
• NOT Tier 0.5 because: it is a movement-domain principle, not a universal cross-pillar baseline-recovery foundation.
**The transfer magnitude and practical sequencing — Tier 2.** The ~2:1 trained-vs-transfer ratio is well-estimated, but mostly against isometric lab proxies; the leap to "general base then specific bias on real-world tasks" rests on that plus narrative review and the older-adult functional literature, without a head-to-head RCT on the sequence itself in general-population users.
• NOT Tier 1 because: the direct test of the prescription (base-first beats specific-only / general-only on real target tasks per unit time) hasn't been run; transfer to dynamic untrained tasks is less quantified than to isometric ones.
• NOT Tier 3 because: it is far more than plausibility — multiple independent meta-analyses converge on the magnitude and the moderators.
The two extremes — Tier 3 / weak. "Only exact-mimicry transfers" and "generic drills far-transfer to sport" are each weakly supported to unsupported; the far-transfer claim in particular has explicit negative reviews. They are tiered low and located honestly as commercial over-claims, not endorsed.
(Underlying exercise physiology — Bohr-effect-grade textbook material like motor-unit recruitment and fibre-type plasticity — is itself Tier-1; the tiering above concerns the transfer/outcome and prescription claims, not the basic physiology.)
Practical takeaway
The whole entry collapses into one sequencing rule and a short checklist for biasing transfer when it matters.
The rule: General base first, specific bias last (and smallest).
1. Build the general base. For nearly everyone — and especially for beginners, older adults, and anyone returning to training — the highest-transfer, lowest-cost move is to get generally strong through full-range, progressively-loaded, mostly multi-joint movements (movement_heavy_compound_lifts, physical_progressive_overload, beginner_exercise_programming_first_4_weeks). This base transfers broadly to daily life, sport, and injury resilience, and it is more valuable the weaker you currently are. Do not skip this for "functional" novelty work — a wobble-board squat you can barely load does not build a base.
2. Train through full range of motion to get strength across the whole movement, rather than relying on isometrics or partials (which strengthen mainly the trained angle, ~±15–30°). Use position-specific isometrics/partials only as a targeted add-on for a known sticking point or a specific angle that matters (e.g. a joint position you keep failing in, or a rehab angle), and train several angles if you need range coverage.
3. Add the specific layer only after the base exists, and keep it the smallest layer. If you have a specific target — a sport movement, a job demand, a particular real-life task — bias your training toward it on the dimensions that actually govern transfer (the dynamic-correspondence checklist below). This last layer is where the over-specificity industry wants you to spend all your time; spend the least of it here, because it has the narrowest payoff.
The transfer checklist (when you do want to bias toward a specific task) — match these, in rough order of leverage:
• Force-vector direction — does the target push/pull horizontally, vertically, rotationally? Train that vector (e.g. horizontal sprinting → horizontal hip-extension force, not only vertical squats).
• Movement velocity & intent — slow target → heavy slow work; fast/explosive target → train fast and train the intent to be explosive even on heavier reps (explosive_power_training). Fast training carries down to slow better than slow carries up to fast.
• Range / joint angles used — train strength through the angles the task demands; cover them with full-range work plus, if needed, position-specific holds.
• Contraction type — eccentric-heavy task → include eccentric emphasis; rebound/plyometric task → include stretch-shortening work.
• Coordination pattern / degrees of freedom — the closer the movement shape to the target, the more the coordination carries. But beware: chasing exact mimicry at light load (the classic "functional" trap) sacrifices the load that built the base. Match the qualities, don't cosplay the movement.
For motor skills (the sport-skill / Mental half): there is no shortcut around practising the actual skill — far transfer barely exists, so a balance app won't fix your tennis and an agility ladder won't make you a faster footballer. To build a skill that holds up in varied game situations rather than only the drill, use variable practice (train a family of related shots/angles/conditions, not one fixed rep) — you'll perform slightly worse on the single drilled version but generalise better to the real, messy task.
Response windows & what "working" looks like:
• Trained-task gains show up first and largest (weeks). If your trained lift is going up but your target task isn't moving, that's the specificity gap talking — not a sign training failed. It's a sign to add (or sharpen) the specific layer.
• General carryover to untrained tasks and daily function accrues over weeks-to-months and is real but roughly half the size of the trained-task gain — expect noticeable, not dramatic, transfer.
• Track the target, not just the proxy. If the goal is "carry shopping without back pain," measure that (and a close lift like a loaded carry / deadlift), not only your leg-press number. A proxy that's improving while the target is flat is the textbook specificity mismatch.
Evidence detail
Why This Entry Exists
A Realised user has a concrete goal underneath their training — they want to climb stairs without their knees complaining, carry shopping without their back giving out, keep up on a hike, play five-a-side without pulling a hamstring, or simply "be strong for real life." They then ask a reasonable question that the fitness world answers badly: does what I do in the gym actually carry over to the thing I care about?
The bad advice comes from two directions at once, which is why an honest version of this is hard to find:
• The over-specificity error. A loud "functional training" marketing layer tells them that unless an exercise looks like their target activity — wobble-board squats, cable rotations that mimic a golf swing, single-leg everything — it is useless or even injurious, and that plain barbell or machine work "doesn't transfer." This is mostly false and it sells equipment, classes, and certifications. It pushes people toward fiddly, low-load, hard-to-progress movements and away from the heavy, simple, progressible lifts that build the base everything else stands on.
• The over-transfer error. A different layer — agility ladders for field-sport speed, "brain-training" apps for reaction time, generic balance drills for "athleticism" — promises that training a generic capacity far transfers to a specific real-world skill. This is also mostly false, and it wastes the user's limited time and attention on activities that improve mainly the drill itself.
This entry exists to give the user the actual rule, which is neither extreme. Adaptation is specific — that part is Tier-1 solid and not in dispute. But specificity is partial, not total: a base of general strength genuinely does carry into untrained tasks, just at roughly half the size of the gains in the trained task. So the honest, useful instruction is a sequence, not a slogan: get generally strong first (it transfers broadly and cheaply), then spend your last and smallest layer of training making it look like the specific thing you care about.
What bad advice does this protect against? "Only sport-specific / 'functional' movements count" (over-specificity — under-builds the base, chases novelty over load). "Get strong and it'll automatically carry over to everything" (over-transfer — true only partially; the last-mile specificity still has to be trained). "This balance app / agility ladder / brain-trainer will make me better at my sport" (far-transfer marketing — almost never supported). And the inverse despair: "I train hard but I'm not better at the thing I actually want, so training doesn't work" — usually a specificity mismatch, not a training failure.
Evidence
Read this as a gradient of certainty: the principle of specificity is among the best-replicated findings in exercise science; the size of partial transfer is well-estimated in lab tasks; the practical sequencing for sport and daily life is reasoned from that plus narrative review; and the two extreme claims are weakly supported at best.
A. The strongest anchor — how much dynamic strength training transfers to an untrained strength task (Tier 1 principle, Tier 2 magnitude)
Task Specificity of Dynamic Resistance Training and Its Transferability to Non-trained Isometric Muscle Strength: A Systematic Review with Meta-analysis (Sports Medicine, 2025; PMC12296774). 43 studies, 1,660 participants, 72 resistance-training interventions, median methodological quality 17/20. Funding: "No funding was received"; authors declared no conflicts — fully independent.
This is the cleanest quantification of the whole topic:
• Trained (task-specific) gains were large: standardised mean difference SMD = 0.98 (95% CI 0.91–1.06) — i.e. you get strongly better at the dynamic exercise you actually trained.
• Transfer to the untrained (isometric) strength task was small: SMD = 0.42 (95% CI 0.35–0.49).
• So the task-specific effect was roughly twice the transfer effect. Both are real and statistically significant — transfer is not zero, but it is not full either. This single contrast is the entire entry in two numbers: adaptation is specific, and transfer is partial.
• Moderator that matters most: single-joint exercises transferred more (SMD = 0.70) than multi-joint exercises (SMD = 0.33). (Plausibly because a multi-joint lift's gains are spread across a more complex coordination pattern that an isometric test doesn't reproduce.)
• Training status barely changed it (transfer SMD 0.29–0.58 across sedentary/active/trained), and duration didn't (<10 weeks ≈ ≥10 weeks).
• Mechanism finding (important and honest): neither muscle hypertrophy nor measured muscle-activation change significantly predicted the strength gains (p ≥ 0.222; R² = 0–13.4%). In other words, the transfer that does happen is not mainly explained by "you grew muscle that's useful everywhere" — it points to neural / coordination adaptations that are themselves partly task-bound.
B. Velocity specificity — you get fast at the speed you train, with asymmetric carryover (Tier 1–2)
• Behm & Sale (1993), "Velocity Specificity of Resistance Training" (Sports Medicine review; independent/academic). The foundational synthesis: strength and power gains are biased toward the velocity (and the intended velocity) at which you train. A nuance that became central later — the intention to move fast is itself a training stimulus, even if the actual bar speed is slow under heavy load.
• Isokinetic velocity-specificity trials (e.g. PubMed 17685688; 7319877; independent/academic). Training fast (e.g. 300°/s) improves torque most at the trained fast velocity but also carries to slower speeds (+18% at trained, +17% at a slower 180°/s), and is accompanied by type-II fibre hypertrophy (+11%) — a structural reason fast training spreads. Training slow tends to be more velocity-specific (gains concentrated near the trained speed, driven by neural adaptation with little morphology change). Practical asymmetry: fast/ballistic training tends to carry downward to slower speeds better than slow training carries upward to fast.
• Meta-analysis of velocity-based vs traditional resistance training (J Sports Sci, 2022; independent). Lifting with deliberate movement intent / set bar speeds transfers preferentially to the matched strength quality on the force–velocity curve.
C. Joint-angle specificity of isometric training — real, with ~30° of useful carryover (Tier 1–2)
Lanza, Balshaw & Folland (2019), "Is the joint-angle specificity of isometric resistance training real? And if so, does it have a neural basis?" (European Journal of Applied Physiology; independent/academic). Trained at one knee angle (65°), strength rose most at the trained angle (+12%) and at nearby angles (+11% at 50°, +7% at 80°, +5% at 35°) — meaningful carryover across roughly a 30° window centred on the trained angle, tailing off outside it. The authors found only weak evidence for a neural basis for the angle-specificity (EMG didn't cleanly track it), leaving the mechanism partly open. Older work (PubMed 2737195) put the effective carryover narrower (~15–20°). Practical translation: isometrics build strength mainly where you hold them, so to cover a range you train multiple angles — whereas full-range dynamic lifting strengthens across the whole movement.
D. Transfer to athletic performance — partial, predictable-ish via "dynamic correspondence" (Tier 2)
• Suchomel et al. / "Training Specificity for Athletes: Emphasis on Strength-Power Training" (narrative review, 2022; PMC9680266; "no external funding," no conflicts). Frames transfer through Verkhoshansky's five dynamic-correspondence criteria: movement amplitude/direction, accentuated region of force, force–velocity dynamics, rate/timing of force, and contraction regime. Core stance, well-aligned with the lab data: build general strength capacity first, then progress toward sport-specific work — and crucially, "exercises with greater specificity tend to transfer to a greater degree in stronger athletes," i.e. weaker people get more from general heavy loading and only need fine specificity once a base exists.
• Functional-training systematic reviews (Frontiers Physiol 2021; HIFT meta-analysis PMC10707569). "Functional"/multi-joint dynamic training does improve fitness and some sport-specific measures (small-to-large effects), but reviews repeatedly note the benefits of generic resistance training are "rarely transferred to sports performance" without the specific bridge — consistent with partial, not automatic, transfer.
E. Motor-skill transfer (the Mental-pillar half) — narrow by default, widened by varied practice (Tier 2)
• Specificity-of-practice hypothesis (classic motor-learning literature; e.g. Schmidt; DTIC ADA156794; PLOS One 2017214). Training effects are highly task- and effector-specific: practising basketball free throws does little for jump shots; a tennis serve doesn't build a badminton smash. Near transfer (to similar tasks) is modest; far transfer (to dissimilar tasks) is essentially absent.
• Variable/varied practice (training a family of related movements rather than one fixed rep) builds a more generalisable motor schema and improves transfer to novel variations — at some cost to peak performance on the single practised version. This is the motor-learning mirror of the strength rule: specificity maximises the trained thing; variability buys breadth.
F. The bridge to everyday function (why this isn't only for athletes) — Tier 2
Machine-based RT functional-capacity meta-analysis (J Funct Morphol Kinesiol 2024); task-specific RT in older adults (PMC11091347); NSCA older-adult position statement (JSCR 2019). General resistance training does transfer to activities of daily living (sit-to-stand, stair climb, gait, carrying) — but adding some task-specific functional work yields additional ADL benefit on top of the general base. Same shape again: base transfers; the last mile is specific.
Mechanism
Why is adaptation specific at all, and why is the transfer only partial rather than zero or total? Two layers — one structural, one neural — and the neural layer is where most of the specificity lives.
1. The structural layer transfers broadly (and is the cheap, general part). When you train against meaningful load, you build muscle cross-sectional area, denser connective tissue and tendon stiffness, and — with fast training — preferentially larger type-II fibres. A bigger, tougher muscle is a general asset: it raises the ceiling of force you can produce, and that ceiling is available to any task that uses that muscle. This is the part that genuinely carries over — to untrained exercises, to daily life, to injury resilience. Notably, though, the 2025 meta-analysis found hypertrophy did not statistically predict the transfer that occurred, which tells us structure is necessary scaffolding but is not the main driver of how much carries over.
2. The neural / coordination layer is where specificity is built — and it is largely task-locked. Strength is not just muscle size; it is the nervous system's learned ability to recruit the right motor units, in the right sequence, at the right rate, at the right joint angles and speeds, for that specific movement. This is a learned motor program. Train a squat and you get superbly good at organising force in the squat pattern — that skill does not fully hand itself to a step-up, a split squat, or a vertical jump, because the coordination demand, the force vector, the joint angles, and the degrees of freedom differ. The same logic explains:
• Velocity specificity — the recruitment/rate-coding pattern that produces force fast is a different motor program from the one that grinds out a slow heavy rep; the intent to be explosive trains the fast program even under a slow bar.
• Angle specificity of isometrics — holding force at one joint angle trains the neural drive for that position; ~30° away the demand has changed enough that less of it applies.
• Motor-skill specificity — a free throw and a jump shot share muscles but are different motor programs; the body learns the program, not an abstract "shooting ability."
Why transfer is partial, not zero. Two non-trained tasks usually share some of the structural asset (the same bigger muscle) and some of the coordination (overlapping force vectors, similar joints). The more they overlap on force-vector direction, force–velocity profile, range/angle, and degrees of freedom, the more carries over. The less they overlap, the more the gain stays locked in the trained task. Specificity is therefore a similarity-weighted gradient, not an on/off switch — which is exactly why both "it's all specific" and "it all transfers" are wrong.
The recovery-register framing. For most Realised users the relevant truth is not the athlete's last 5% of sport-specific transfer; it is that general strength is a broad, durable asset that protects function — it makes the body harder to break, easier to move, slower to decline. You are not chasing a sport PR; you are restoring and protecting capacity that modern sedentary life erodes (avoid_chronic_sitting, physical_foundations_for_baseline). The base is the point; specificity is the optional bias on top.
Risks And Contraindications
The concept of training transfer carries no direct physical risk. The risks are all decision risks created by getting specificity wrong — and one real injury risk from a popular mis-application.
• Over-specialisation / premature mimicry (the main practical risk). Skipping the general base to jump straight into low-load, sport-mimicking "functional" drills under-builds strength and is a slower, more fragile route to almost any goal. For youth athletes specifically, early single-sport specialisation is associated with higher overuse-injury and burnout risk — breadth first is the safer developmental path. (The narrative review flags early specialisation as a documented limitation of mis-applied dynamic correspondence.)
• Loading a sport-mimicking movement heavily for the sake of "specificity." Adding heavy external load to a movement designed for speed/coordination (e.g. weighted bat swings, heavily-loaded rotational mimics) can distort the very motor pattern you're trying to transfer to, and loads joints/tendons in positions they aren't built to be loaded in. Train the quality (vector, velocity, range) with appropriate tools; don't strap weight onto the skill itself.
• Don't drop the base to chase the bias. Once a general base exists it still needs maintenance; reallocating all training to a narrow specific layer lets the general capacity (and its broad protective transfer) erode.
• Injury-resilience claims cut both ways — don't over-trust transfer. General strength genuinely reduces soft-tissue injury risk broadly, but it does not fully protect a specific tissue in a specific high-demand position you haven't trained. (Classic example: strong general legs don't immunise a hamstring against high-speed sprinting strain unless you've trained that specific high-velocity, lengthened-position demand.) The partial-transfer rule is also a safety rule: assume an untrained position is under-prepared.
• Standard resistance-training contraindications still apply — the loading itself follows the usual rails (physical_progressive_overload, beginner_exercise_programming_first_4_weeks): build load gradually, respect form under fatigue, get medical clearance for relevant cardiovascular/musculoskeletal conditions. Nothing about "transfer" overrides those.
Minimal-risk bottom line: the safest and highest-transfer default for almost everyone is unglamorous — get generally strong through full-range loaded movement, then add a thin specific layer if a specific goal demands it.
Controversy
Nature: Not a scientific dispute about whether specificity exists (it does — Tier 1). It is a bidirectional commercial over-claim: two industries pull the practical recommendation toward opposite extremes, and the honest finding sits in the middle.
Position A — "Specificity is near-total" (the functional-training / sport-specific coaching layer).
• Claim: only movements that closely mimic the target activity ("functional," unstable-surface, sport-pattern) transfer; conventional lifts (leg press, machines, even barbell work) "don't carry over" and may be useless or injurious.
• Best evidence for it: specificity is real and large; dynamic correspondence does improve transfer; exact-task training does produce the biggest gains in that task.
• Where it over-reaches: it ignores the partial-transfer half (general strength carries over substantially, ~half the trained-task effect), and reviews have found no evidence for the specific claims that non-mimicking lifts (leg press, Nordic curls) harm performance or cause injury. Strong financial incentive: novelty equipment, classes, certifications.
Position B — "Transfer is broad / far transfer is real" (the brain-training, agility-ladder, generic-athleticism layer).
• Claim: training a generic capacity (reaction time, agility, balance, "athleticism," perceptual-cognitive drills) far transfers to your specific sport or task.
• Best evidence for it: some general fitness/strength genuinely transfers to general capacity, and within-task gains are easy to show.
• Where it over-reaches: the transfer it sells is to the drill, not the sport. Reviews are blunt — "there is no supporting evidence for a far transfer of general perceptual or cognitive training to sports performance," and motor-learning's specificity-of-practice principle says skill transfer is narrow by default. Financial incentive: apps, gadgets, subscription "brain/agility" products.
**The funding/bias dimension (why this entry's caution is inverted relative to a typical Realised entry):** the usual Realised pattern is industry suppressing a cheap, unpatentable intervention. Here there is no suppression and no pharma interest — instead two commercial layers over-sell in opposite directions. Tellingly, the cleanest evidence is the non-conflicted evidence: the pivotal quantification (2025 transfer meta-analysis), the angle-specificity study, and the strength-power narrative review all declare no funding and no conflicts, and they land squarely in the honest middle — specificity is real and large, transfer is real and partial, far transfer is largely absent.
Realised Position: Adaptation is specific — that is not negotiable and is Tier-1 settled. But specificity is partial: a general strength base transfers broadly (to untrained tasks, daily function, and injury resilience) at roughly half the magnitude of the trained-task gain, and is more valuable the less trained you are. So the prescription is a sequence, not a slogan: build a broad general base first (high transfer, low cost, the priority for almost everyone), then add the smallest necessary specific layer biased on the dimensions that govern carryover (force vector, velocity/intent, range, contraction type, coordination). We reject the "only mimicry counts" over-specificity claim and the "generic drills far-transfer to your sport" over-transfer claim equally. See What Would Change Our Mind.
Cross-Pillar Connections
• Physical — physical_progressive_overload, movement_heavy_compound_lifts, beginner_exercise_programming_first_4_weeks: the general base this entry tells you to build first. Specificity is the bias you add on top of progressive overload, not a replacement for it.
• Physical — movement_modality_selection, explosive_power_training, balance_and_proprioception_training: the specific-layer tools — velocity/intent, force-vector, and coordination biasing. Choose the modality that matches the target task's qualities.
• Physical — resistance_training_and_body_composition, physical_foundations_for_baseline, avoid_chronic_sitting: the recovery-register payoff — general strength as broad, durable, protective capacity that transfers to daily life and resists decline.
• Physical — physical_zone2_cardio: the same specificity logic in the energy-system domain (aerobic adaptations are also partly substrate/intensity-specific); a useful parallel for users who assume "fitness is fitness."
• Mental — habit_formation_fundamentals: the motor-skill half is a learning problem; varied practice and the narrowness of far transfer mirror how skills (and habits) generalise — or fail to — beyond the exact context they were built in.
What would change our mind
We would UPGRADE the practical magnitude/sequencing claim toward firmer Tier 1 if:
• Adequately-powered RCTs in healthy general-population users directly compared "general base then specific bias" against "specific-only" and "general-only" on real-world target tasks (not lab proxies) and confirmed the base-first sequence wins on transfer per unit time.
• Independent replications pinned the transfer magnitude (the ~0.42 vs 0.98 SMD ratio) across more movements and to dynamic (not just isometric) untrained tasks.
We would REVISE the mechanism emphasis if:
• Better-instrumented studies showed hypertrophy does drive transfer after all (reversing the 2025 meta-analytic null), shifting weight from the neural/coordination account back toward the structural one.
• A clean neural basis for joint-angle specificity were established (the Lanza 2019 EMG evidence was weak), which would sharpen the "specificity is neural" claim.
We would DOWNGRADE / become more cautious if:
• Far transfer from generic perceptual/cognitive or agility training to sport were robustly demonstrated in pre-registered trials (it currently is not) — that would partially rehabilitate Position B.
• Evidence emerged that general strength training meaningfully protects specific tissues in untrained positions (e.g. fully prevents hamstring strain without high-velocity lengthened-position work) — that would weaken the "assume untrained positions are under-prepared" safety rule.
What would NOT change our mind: more marketing testimonials in either direction, or more within-task improvement studies (which only ever show the trained thing got better — they say nothing about transfer).
Industry bias note
Bias risk on this topic is MODERATE and bidirectional, and it runs opposite to the typical Realised "industry suppresses the cheap thing" pattern.
• No suppression, no pharma stake. Strength training, full-range movement, and motor practice are unpatentable and free-to-cheap. There is no revenue model with an incentive to bury the honest finding.
• The incentive is to over-sell — in two directions. (1) The "functional / sport-specific" industry profits from selling specificity (unstable-surface gear, mimicry drills, branded systems, certifications) and therefore over-states that only mimicking movements transfer and that conventional lifts don't — a claim reviews have specifically failed to support. (2) The "brain-training / agility / perceptual" industry profits from selling far transfer (apps, gadgets, subscriptions) and over-states that generic drills carry over to real sport — a claim reviews explicitly reject.
• The cleanest evidence is the non-conflicted evidence. The pivotal sources here — the 2025 transfer meta-analysis, the angle-specificity study, the strength-power narrative review — all declare no funding and no conflicts, and none of them sells a product. They converge on the honest middle, which is the position Realised adopts.
• The reverse-caution. Because head-to-head RCTs of the practical sequence in general-population users are cheap-but-unfunded (no product to sell, no grant glamour), "no direct trial of base-first-then-specific" reflects no one paying to run it more than a tested negative — which is why the prescription sits at honest Tier 2 (do it, it's well-reasoned) rather than Tier 1.
Sources (23)
- *Transfer magnitude (primary anchor):**↗
- Task Specificity of Dynamic Resistance Training and Its Transferability to Non-trained Isometric Muscle Strength: A Systematic Review with Meta-analysis. Sports Medicine (2025); PMC12296774 / Springer 10.1007/s40279-025-02225-2.↗ 43 studies, 1,660 participants, 72 interventions. Task-specific SMD 0.98 (CI 0.91–1.06); transfer SMD 0.42 (CI 0.35–0.49); single-joint transfer 0.70 vs multi-joint 0.33; hypertrophy/activation did not predict gains. (No funding; no conflicts — independent.)
- *Velocity specificity:**↗
- Behm DG, Sale DG. Velocity Specificity of Resistance Training. Sports Medicine (1993). (Academic/independent — foundational review; intent-to-move-fast as stimulus.)↗
- Velocity-specific isokinetic training trials, PubMed 17685688 and 7319877. (Academic/independent — fast training +18% trained / +17% slower velocity, +11% type-II fibre hypertrophy; slow training more specific/neural.)↗
- Velocity-based vs traditional resistance training: systematic review & meta-analysis. Journal of Sports Sciences (2022); tandfonline 10.1080/02640414.2022.2059320.↗ (Independent.)
- *Joint-angle specificity:**↗
- Lanza MB, Balshaw TG, Folland JP. Is the joint-angle specificity of isometric resistance training real? And if so, does it have a neural basis? European Journal of Applied Physiology (2019); PubMed 31522276. (Academic/independent — +12% at trained 65°, carryover ~30°; weak neural-basis evidence.)↗
- Earlier angle-specificity work, PubMed 2737195 (~15–20° carryover). (Academic/independent.)↗
- *Transfer to sport / dynamic correspondence:**↗
- Training Specificity for Athletes: Emphasis on Strength-Power Training: A Narrative Review. (2022); PMC9680266. Five dynamic-correspondence criteria; "build general base first, then specific"; specificity transfers more in stronger athletes. (No external funding; no conflicts.)↗
- Effect of Functional Training on Physical Fitness Among Athletes: A Systematic Review. Frontiers in Physiology (2021). (Independent — functional training improves fitness; transfer to sport often not automatic.)↗
- High-intensity functional training meta-analysis. PMC10707569. (Independent — small-to-large effects on strength/power/sport-specific measures.)↗
- *Motor-skill / far-transfer:**↗
- There is No Supporting Evidence for a Far Transfer of General Perceptual or Cognitive Training to Sports Performance. PMC11560981. (Independent review — far transfer not supported.)↗
- Specificity-of-practice & variability-of-practice literature: DTIC ADA156794; PLOS One 10.1371/journal.pone.0174214↗ (2017); Schmidt schema theory. (Academic/independent.)
- *Transfer to everyday function:**↗
- Machine-Based Resistance Training Improves Functional Capacity in Older Adults: Systematic Review & Meta-Analysis. J Funct Morphol Kinesiol (2024). (Independent.)↗
- Task-specific resistance training adaptations in older adults. PMC11091347. (Independent — general RT transfers to ADL; task-specific work adds further benefit.)↗
- NSCA Resistance Training for Older Adults: Position Statement. JSCR (2019). (Professional society.)↗
- *Mechanism (neural vs hypertrophic transfer):**↗
- Minimal Role of Hamstring Hypertrophy in Strength Transfer Between Nordic Hamstring and Stiff-Leg Deadlift: Blinded RCT. medRxiv 2024.12.17.24319207. (Independent preprint — hypertrophy did not explain transfer; supports neural/coordination account. Preprint — weight accordingly.)↗
- Funding notation: every pivotal source above is academic/independent or explicitly declares no funding and no conflicts. The principal sources of the two over-claims (mimicry-only specificity; far-transfer brain/agility products) are commercial — branded methods, equipment, certifications, apps — and are weighted accordingly.*↗