Running Mechanics · Miles of Meaning

Understand the movement, practise it gradually, and apply one useful cue during real running. This guide serves new runners, experienced runners and ultrarunners. It explains movement rather than prescribing a single perfect gait.

Five-move learning loop: RETRACT → REACH → ATTACK → CONTACT → PROPEL.

Practice progression: SHAPE → CONNECTION → MARCH → SKIP → RUN.

The five-move loop and practice progression form the Miles of Meaning teaching framework for understanding and practising running movement. Their labels organise the explanations in this guide. Research supports particular mechanical observations; it does not establish that memorising this loop improves economy, race results or injury outcomes.

1. Understand the complete cycle

A stride follows one leg from initial contact to its next initial contact. Stance runs from initial contact to toe-off; swing runs from toe-off to the next contact. The other leg is doing something different at the same time. During typical running, periods with neither foot on the ground separate the support periods. Very slow grounded running can occur without a flight phase. Walking generally includes double support and uses different mechanics. [1, 2, 16]

The five labels describe a continuous cycle in a different starting order: recovery after toe-off, forward swing, preparation for touchdown, loading, then support and propulsion. They overlap. CONTACT and PROPEL divide aspects of stance for teaching; neither is a sharply bounded conventional gait-analysis phase. A runner does not need to think through five commands every step.

The orange right leg recovers in the foreground while the blue left leg supports farther away. — **One leg through the running cycle.** Follow the same orange right leg, nearest the viewer, through five selected moments and back to recovery. The blue left leg stays farther away and has a different simultaneous role, labelled below each picture. The cycle is continuous. Panels are not equal time intervals or universal joint angles.

The orange right thigh advances with a bent knee in the foreground; the blue left forefoot supports behind the body. — **One leg through the running cycle.** Follow the same orange right leg, nearest the viewer, through five selected moments and back to recovery. The blue left leg stays farther away and has a different simultaneous role, labelled below each picture. The cycle is continuous. Panels are not equal time intervals or universal joint angles.

Teaching movement	Biomechanical context	What the cue describes
RETRACT	Early swing / recovery after toe-off	Knee flexion and heel recovery after toe-off
REACH	Forward swing, mainly hip flexion	Advancing the thigh and knee; not reaching the foot far ahead
ATTACK	Late swing / terminal swing	Limb positioning and preparation for loading; may include backward leg movement before contact
CONTACT	Initial contact and early stance / loading response	Meeting the ground, accepting load and usually early braking on level steady running
PROPEL	Stance support through later stance and toe-off	Compression, elastic return and active muscle work are connected; forward propulsion predominates later in level stance

The illustrations show selected qualitative running moments. Orange follows one leg; blue shows what the other leg is doing. They are not motion capture, an angle prescription or a diagnostic template. Speeds, proportions and phase durations vary.

Forces, muscles and elastic energy

The ground reaction force is the ground's force on the contacting foot. Its vertical component helps support and redirect the body; its fore–aft component can brake or propel. At a steady speed on level ground, early stance usually has a braking component and later stance a forward propulsive component. The runner still moves forward while braking occurs. Initial contact is not automatically propulsive. Slopes, acceleration and foot–surface interactions change this pattern. [1, 3, 16]

The spring-mass model represents the body as a mass supported by a compressible spring-like leg. It helps explain loading and rebound, but the leg is not a passive metal spring. Muscles do work and control joints; tendons store and return energy; soft tissues and shoes also contribute. Energy is not all recovered, and the model alone cannot choose an individual's ideal stiffness or technique. [1, 3, 4]

Hip muscles contribute to swing, support and propulsion. The knee manages flexion and extension and contributes to support. Ankle plantarflexors, including soleus and gastrocnemius, play substantial roles in support and propulsion. Contributions change with speed and phase; “push harder from the toes” is an inadequate summary. Tendon recoil and muscular work are related but distinct. [3, 4]

Metrics without targets

Cadence is steps per minute; a stride contains two steps. At a given speed, a higher cadence means a shorter average step length. Contact time describes how long a foot is on the ground. Flight time, vertical oscillation and joint motion vary with speed, surface, anatomy and fatigue. None is a standalone grade of running quality. Smaller vertical oscillation or shorter contact time does not necessarily cause better economy. Running economy is the energy cost at a defined submaximal speed, not how elegant running looks. [5, 6]

Arm swing helps coordinate the legs and manage angular momentum. Posture and pelvic control are dynamic rather than rigid: allow normal rotation and a context-appropriate lean, without deliberately freezing the pelvis or maintaining one trunk angle. Observed associations between motion and economy cannot establish a universal pose. [5, 7]

2. Learn the five movements

RETRACT — recover and fold the leg

After toe-off, the knee flexes and the heel rises toward the body as the limb recovers. Folding shortens the effective leg lever; faster running often involves greater heel recovery. Let the movement fit the speed. A gentle easy jog does not require a sprint-style heel-to-butt position. Relax the lower leg and let recovery continue into forward swing. Do not deliberately snap the heel upward or squeeze the hamstrings on every stride. [1, 3, 16]

Here RETRACT names heel recovery after toe-off. Its meaning in this framework is defined by the movement explanation above.

REACH — advance the thigh and knee

Hip flexion moves the thigh forward and advances the knee while the trunk and pelvis support the motion. The leg is still folded during part of this swing. REACH means moving the knee and thigh forward; it does not mean casting the foot far in front of the body. Knee height and swing range increase or decrease with context. Forced high knees can add effort without improving easy-running coordination. [1, 3, 16]

Consider “let the knee travel” during a brief drill; then return to natural running. Forward swing contributes to where the next step is prepared, but step length also depends on speed, stance motion and flight. It cannot be set by pulling the knee higher alone.

ATTACK — prepare for contact

During late swing, the knee extends and the foot approaches the ground as the limb prepares to accept load. The leg may begin moving backward relative to the body before touchdown. This is not a demand that the foot match ground speed exactly or that a runner actively scrape or paw the ground. [1, 8]

ATTACK means controlled, purposeful preparation. It does not mean stomping, forceful forward kicking or trying to strike harder. Allow the lower leg to unfold without reaching for the ground. Avoid turning this label into an aggressive action that changes foot strike or creates tension.

**ATTACK: late swing preparation.** The orange right foot approaches the ground in the foreground; the blue left leg recovers farther away. Both feet are airborne. This is one selected qualitative moment, not a required position or foot-strike pattern.

CONTACT — meet the ground and accept load

Initial contact begins stance. The supporting limb manages ground reaction force through joint motion, muscle activity and tissue loading. Hip, knee and ankle behaviour varies across runners and conditions. Foot contact is often somewhat ahead of the body's centre of mass; “exactly under the hips” is not a universal touchdown requirement. Excessive forward placement may increase braking in some circumstances, but a video angle alone cannot quantify internal tissue load. [1, 6, 9]

Heel, midfoot and forefoot contacts can all occur in normal running. Changing strike redistributes loading; it does not remove loading or prove injury prevention. Loading rate is how rapidly force rises, not a direct measure of every tissue's stress. Quiet contact can be a practical observation, but a quieter sound does not establish safer loading. [9, 10]

The body and supporting leg compress under load. This compression is loading, not forward propulsion by itself. Do not deliberately crouch deeper to create a larger rebound.

**CONTACT: first contact.** The orange right foot touches the ground in the foreground with a flexed knee; the blue left foot remains airborne farther away. This is one selected qualitative moment, not a required position or foot-strike pattern.

PROPEL — support, extend and rebound

Stance continues through controlled compression, support, extension and toe-off. Muscle–tendon systems store and return elastic energy while muscles also do active work. In typical steady level running, the later fore–aft ground force is propulsive. Hip, knee and ankle contributions vary; toe-off ends contact and starts the next recovery. [1, 3, 4]

Loading and propulsion are connected but mechanically distinct. The teaching label PROPEL spans that connection; it does not imply that every moment of stance drives the runner forward. Avoid forcing an exaggerated push-off, locking the knee or staying on the toes. Think of moving smoothly through support rather than manufacturing a bounce.

**PROPEL: later stance.** The orange right forefoot supports behind the body in the foreground; the blue left thigh advances farther away. This is one selected qualitative moment, not a required position or foot-strike pattern.

Two legs, one continuous movement

One leg can be supporting the body while the other folds or swings. In flight, both legs move without either supplying a ground force. The legs are approximately offset in ordinary symmetrical running, but the exact timing and symmetry vary. The diagrams show phase relationships, not fixed percentages everyone must match. [1, 2, 16]

The orange right leg trails in the foreground after toe-off; the blue left foot approaches its next contact farther away. Both feet are airborne. — **Two legs: stance, swing and flight.** Follow the orange right leg, always nearest the viewer, and the blue left leg farther away. Read the two-row overview vertically to see both legs at the same selected moment. The sequence progresses from right support to flight, left support, a second flight and right contact again. Labels overlap; these are not measured phase boundaries, equal time intervals or universal joint positions. Grounded running may have different overlap.

Both legs at the same moment
Leg	Moment 1	Moment 2	Moment 3	Moment 4	Moment 5	Moment 6
Right leg · orange · near	PROPELGround support	PROPEL → RETRACTSwing / airborne	RETRACTSwing / airborne	REACHSwing / airborne	ATTACKSwing / airborne	CONTACTGround support
Left leg · blue · far	REACHSwing / airborne	ATTACKSwing / airborne	CONTACT / loadingGround support	PROPELGround support	RETRACTSwing / airborne	RETRACTSwing / airborne

The orange right leg accepts load in the foreground with a flexed knee; the blue left foot remains airborne farther away. — **Contact, loading and later propulsion.** The orange leg moves from contact and load acceptance toward later stance and toe-off. Loading and forward propulsion are connected but distinct. The pictures do not measure ground reaction forces, tissue loads or elastic energy.

3. Practise progressively

Start with a short easy warm-up. Choose one drill that helps you recognise a movement; do not complete every drill as a required circuit. The volumes below are conservative editorial starting examples, not doses established by trials of this framework. Rest enough to regain relaxed control. Repeat or regress a stage before adding speed or impacts. Stop a drill if it causes pain, instability or an altered gait; do not use technique practice to work through symptoms.

SHAPE builds recognition; CONNECTION links movements; MARCH adds alternating rhythm; SKIP adds a small elastic demand; RUN transfers one observation into normal running. Skipping is optional. An experienced runner can move directly from a controlled march to a short easy-running test.

SHAPE 1 — supported fold and return

Purpose: recognise RETRACT without a forceful heel kick.
Instructions: stand beside a wall, hold it lightly, bend one knee so the heel rises comfortably, then lower the foot. Keep the thigh relaxed rather than pulling the heel to the buttock.
Mechanics: knee flexion shortens the recovering lever; standing isolates a shape and does not reproduce running timing or loads.
Volume: 1–2 sets of 5 slow repetitions per side.
Common mistakes: arching the back, squeezing hard, chasing heel height.
Progress / regress: connect the fold to a small knee advance; regress to a smaller bend with more support.
Safety: use a stable support and pain-free range; stop if balance or hamstring discomfort worsens.
Evidence: anatomy and recovery mechanics are established observations; benefit from this isolated drill is untested.

SHAPE 2 — supported knee advance

Purpose: distinguish REACH from forward foot reaching.
Instructions: hold a wall lightly. Lift one thigh a comfortable amount with the knee bent; return gently. Keep the trunk relaxed and avoid a high-knee target.
Mechanics: hip flexion and knee advancement; balance and pelvic control.
Volume: 1–2 sets of 5 repetitions per side.
Common mistakes: kicking the foot forward, leaning back, holding the breath.
Progress / regress: alternate legs with less hand support; regress to a lower knee lift.
Safety: stand on a level surface; stop for pain or instability.
Evidence: the movement demonstrates joint action; improved running performance is not established for this drill.

CONNECTION 1 — fold, advance, unfold

Purpose: link RETRACT, REACH and ATTACK slowly.
Instructions: with support, fold the leg, move the bent knee a little forward, let the lower leg unfold toward a comfortable step position, then place the foot gently. Pause between repetitions.
Mechanics: a simplified swing sequence; it is deliberately slower than actual running and omits flight.
Volume: 3–5 controlled cycles per side.
Common mistakes: exaggerated circles, forceful forward kicking, dragging or scraping the foot backward.
Progress / regress: reduce pauses; regress to two connected components.
Safety: remain below a range that produces tension; no stomping or forced foot strike.
Evidence: educational coordination synthesis; transfer and optimal volume are uncertain.

CONNECTION 2 — small step, load, release

Purpose: feel that CONTACT loading precedes the release toward the next step.
Instructions: take a small walking step, allow a little comfortable joint flexion, then move through support into the next step. Use support if needed and keep the movement small.
Mechanics: introduces load acceptance and progression through support; walking is an analogy, not a running-force reproduction.
Volume: 2 passes of 6–8 steps.
Common mistakes: deep squats, deliberately landing harder, pushing forcefully from the toes.
Progress / regress: connect continuous steps; regress to a supported weight shift.
Safety: choose an even surface and comfortable range; stop for joint or calf pain.
Evidence: a teaching analogy, not a validated spring-stiffness or injury-prevention exercise.

MARCH 1 — relaxed alternating march

Purpose: connect alternating legs, arm rhythm and posture.
Instructions: march slowly with modest knee lift. Let the opposite arm accompany the leg. Place each foot comfortably, without stamping or locking the supporting knee.
Mechanics: REACH and CONTACT coordination, with supported balance; marching lacks running flight.
Volume: 2 × 20 seconds with easy walking between.
Common mistakes: forced high knees, rigid pelvis, breath holding.
Progress / regress: move forward at a natural rhythm; regress to slower steps or wall support.
Safety: leave clear space and stop if balance declines.
Evidence: coordination practice is a plausible teaching aid; its independent effect on running economy is uncertain.

MARCH 2 — low step-over rhythm

Purpose: keep the foot connected to the thigh's movement instead of kicking ahead.
Instructions: imagine stepping over a very low line as you march. Advance the knee, let the lower leg unfold, and place the foot gently. There is no physical obstacle to clear.
Mechanics: slow REACH–ATTACK–CONTACT transition.
Volume: 2 passes of 10–15 metres, or 2 × 20 seconds in a clear space.
Common mistakes: oversized steps, exaggerated ankle action, intentional ground scraping.
Progress / regress: link to a few easy running steps; regress to smaller stationary steps.
Safety: stay on level ground and reduce range when uncertain.
Evidence: editorial drill; no guaranteed performance or injury benefit.

SKIP 1 — low alternating skip

Purpose: introduce rhythm and a small elastic load.
Instructions: use a gentle step-hop rhythm, alternate sides, keep the hops low and the arms easy. Let the whole foot–leg system accept load without dictating a strike pattern.
Mechanics: a simplified CONTACT–PROPEL connection plus swing; skipping differs from running.
Volume: 2 × 10 seconds, with full walking recovery.
Common mistakes: height competitions, stiff knees, forcing the toes, adding speed too soon.
Progress / regress: slightly extend duration while control stays comfortable; regress to marching or omit skipping.
Safety: use a level nonslip surface. New impact work adds load; skip this stage if hopping causes pain or you have not tolerated comparable impact.
Evidence: spring-like loading is a mechanical observation; this exact skipping progression has not been validated for ultra outcomes.

SKIP 2 — skip, then settle

Purpose: transfer elastic rhythm without carrying an exaggerated drill into running.
Instructions: make a few low alternating skips, then walk naturally. Notice the contrast between deliberate coordination and automatic movement.
Mechanics: alternation, loading and release, followed by reduced conscious control.
Volume: 3 rounds of 5 seconds skipping and 20 seconds walking.
Common mistakes: racing the drill, bouncing higher, continuing while coordination deteriorates.
Progress / regress: transition into a few easy-running steps; regress to march–walk transitions.
Safety: restore comfortable breathing and balance between rounds; stop for calf or tendon symptoms.
Evidence: editorial transfer exercise; neither the timing nor injury benefit is established.

RUN 1 — one cue, then natural running

Purpose: test a useful observation during ordinary easy running.
Instructions: run naturally, try one gentle cue such as “let the knee travel” or “relaxed recovery” briefly, then let the cue go. Compare comfort and effort at a similar pace and surface.
Mechanics: integrates a selected swing observation into the complete gait cycle.
Volume: 3 × 20 seconds of attention, separated by at least 60 seconds of natural easy running.
Common mistakes: reciting all five labels, chasing a pose, treating a quieter sound as proof of safer loading.
Progress / regress: retain only repeatedly useful cues and reduce attention; regress to a march or omit the cue.
Safety: return to natural running if effort, tension or discomfort rises. A cue must never override pain or loss of control.
Evidence: a practical self-observation experiment; immediate comfort is not proof of long-term economy or injury prevention.

RUN 2 — context transfer and release

Purpose: learn that a cue's usefulness depends on terrain and speed.
Instructions: after a comfortable easy-running test, try the same observation on a familiar gentle incline or a slightly quicker relaxed segment. Return to easy running and decide whether to keep, modify or discard it.
Mechanics: adaptable coordination rather than a fixed gait; use the Form Coach's terrain-specific controls.
Volume: 2 × 15–20 seconds with at least 60 seconds easy recovery. This replaces part of an existing run.
Common mistakes: using a flat-ground cue on every slope, accelerating hard, adding drills when already exhausted.
Progress / regress: test one new context at a time; regress to familiar flat easy running.
Safety: avoid technical descents for a first test; stop cueing if control deteriorates.
Evidence: editorial application; no trial establishes this loop's efficacy across terrains or ultra fatigue.

4. Apply it during real running

Context	What may change	Useful observation / decision
Easy running	Smaller swing and heel recovery than fast running	Stay relaxed; do not enlarge movements to match an illustration
Faster running	Speed commonly changes step length, cadence, swing and contact time	Let movement scale with speed; do not force sprint mechanics into an easy jog
Uphill	More work against gravity; gradient changes joint contributions and step pattern	Choose a comfortable shorter step or hike when effort warrants it; no fixed lean or cadence target
Downhill	More negative work and braking; control and surface matter	Reduce speed when control declines; do not remove all braking or force a forefoot landing
Technical trail	Variable foot placement, step width and timing	Look along a safe line and prioritise secure placement over a symmetrical appearance
Long-run fatigue	Motion, muscle activation and contact timing can change	Notice loss of control; reduce attention to technique, adjust pace, walk or end the run when needed
Ultras	Long duration, walking, terrain, sleep loss and fueling interact	Use familiar cues sparingly; do not start a wholesale gait change on race day

Slope and fatigue research describes changes under specific conditions; it does not validate a single technique for every ultra. More economical-looking motion is not a reason to ignore symptoms. [11, 12]

Open the Form Coach for terrain selection, brief cue tests, rejection rules and a coaching log. Its existing Relax–Stack–Swing–Land–Flow scan is an application checklist, not a competing gait-cycle theory. Learn here → practise a drill → apply one cue there.

Connect technique practice to Ultra Training, the training protocols and your existing training load. Drill time replaces a small part of a session; it is not automatically extra training. Use the glossary for shared terms.

5. Evidence and limitations

Descriptive mechanics: stance/swing, joint actions, muscle contributions and force patterns are established concepts. Models and laboratory measurements simplify a three-dimensional, individual movement. They do not determine your ideal gait.

Cadence: controlled studies show that small step-rate changes at a matched speed can alter some joint mechanics. That does not create a universal 180-steps-per-minute target, nor establish that every runner should increase cadence. [6, 15]

Foot strike: reviews do not establish one universally superior strike for injury risk or running economy. Changing strike can move demand toward different tissues. An external loading-rate reduction cannot be assumed to reduce all internal loads or prevent injury. [9, 10]

Gait retraining and injury: some targeted programmes change measured mechanics, but outcomes depend on intervention, population and adherence. A novice-runner trial reported fewer injuries after feedback-based retraining; a later wearable-feedback randomised trial found no injury benefit in its primary intention-to-treat comparison. Its more favourable as-treated comparison partly breaks randomisation and cannot replace the primary result. Neither validates this five-move curriculum. [13, 14]

Running economy: a systematic review finds associations between several biomechanical characteristics and economy, with variable strength and context. Association does not establish that deliberately changing a metric will reduce your energy cost. Read the published correction alongside the review. [5]

Drills and fatigue: this progression is a transparent coaching synthesis. Optimal drill dose, transfer, lasting benefit and injury outcomes for this exact framework are unknown. Fatigue-related changes are not automatically defects to correct, and short laboratory studies cannot establish multi-day efficacy. [12]

Record one observation, the context, the cue, immediate comfort/effort and any symptoms. Keep, modify or discard the cue after repeat comparable sessions. Seek an appropriately qualified assessment for persistent pain or marked functional difficulty rather than diagnosing yourself from a video or wearable.

6. Definitions used in this guide

Term	Meaning
Initial contact / toe-off	Start / end of foot–ground contact
Stance / swing	Contact period / limb recovery and advance between contacts
Hip flexion / knee flexion	Thigh advancing relative to pelvis / knee bending
Knee extension / plantarflexion	Knee straightening / ankle action pointing the foot away from the shin
Ground reaction force	Force of the ground on the foot
Braking / propulsion	Backward / forward fore–aft force component on the runner
Loading rate	Rate at which a measured force increases; not a universal tissue-stress measure
Centre of mass	Mass-weighted location of the body; changes with limb positions
Elastic return / muscle work	Energy recovered from stretched tissues / energy supplied or absorbed by muscle
Cadence / step / stride	Steps per minute / one foot contact to the opposite / one foot contact to the same foot again

Framework and terminology

RETRACT → REACH → ATTACK → CONTACT → PROPEL organises the continuous running movement into five teaching cues. Movements, positions and gait phases can be described using different terms. This guide defines the terms needed for its explanations; it does not attempt to catalogue every alternative name or treat these five labels as a universal biomechanical naming standard.

Miles of Meaning makes no claim of ownership or exclusivity over these movements, positions or their names. You can use your own words and cues for the same movements.

Routine research updates can flag a claim or illustration for review; they must not silently rename the five movements or alter their definitions. Changes to the framework need an explicit editorial proposal, bilingual review and the normal authorization process. Research does not automatically change drill doses or safety boundaries.

7. Sources and review status

The references below support the specific observations discussed; none tests the entire curriculum. Evidence coverage is topic-specific, not a claim of a complete literature search or clinical review. Named human expert review is not established. Independent AI assessment and publication authorization are recorded separately in the content's review packet.

Muscle contributions to propulsion and support during running (opens in a new tab) · author full text.
Mechanics, energetics and implementation of grounded running technique: a narrative review (opens in a new tab) · primary abstract.
The spring-mass model for running and hopping (opens in a new tab) · primary abstract.
Speed-specific optimal contractile conditions of the human soleus muscle from slow to maximum running speed (opens in a new tab) · primary full text.
The Relationship Between Running Biomechanics and Running Economy: A Systematic Review and Meta-Analysis of Observational Studies (opens in a new tab) · primary full text; 2025 correction inspected.
Effects of step rate manipulation on joint mechanics during running (opens in a new tab) · author full text.
The metabolic cost of human running: is swinging the arms worth it? (opens in a new tab) · primary full text.
Swing-leg retraction: a simple control model for stable running (opens in a new tab) · primary abstract.
What are the Benefits and Risks Associated with Changing Foot Strike Pattern During Running? A Systematic Review and Meta-analysis of Injury, Running Economy, and Biomechanics (opens in a new tab) · publisher primary abstract.
Ground reaction force metrics are not strongly correlated with tibial bone load when running across speeds and slopes: Implications for science, sport and wearable tech (opens in a new tab) · primary full text.
Biomechanics and Physiology of Uphill and Downhill Running (opens in a new tab) · publisher primary abstract.
Effects of running-induced fatigue on joint kinematics and kinetics during overground running: a systematic review and meta-analysis (opens in a new tab) · primary abstract.
Gait Retraining for the Reduction of Injury Occurrence in Novice Distance Runners: 1-Year Follow-up of a Randomized Controlled Trial (opens in a new tab) · publisher primary abstract.
The Effect of Wearable-Based Real-Time Feedback on Running Injuries and Running Performance: A Randomized Controlled Trial (opens in a new tab) · primary full text.
What is the Effect of Changing Running Step Rate on Injury, Performance and Biomechanics? A Systematic Review and Meta-analysis (opens in a new tab) · primary full text.

Correction to reference 5: published figure correction (opens in a new tab). Access is recorded per source; abstract-only records do not establish full-text review.

The biomechanics of running (opens in a new tab) · Primary full text retrieved via third-party mirror; article identity verified; foundational 1998 review, not current intervention evidence.