What actually separates flat-out speed from change-of-direction ability from true agility, how rate of force development and impulse show up in a sprint stride, why the spring-mass model matters, what changes phase-by-phase in a sprint, how to load resisted and assisted sprint work, and how to build athletes from closed cone drills toward reactive game speed.
What actually shows up on the exam
These three terms get used almost interchangeably in casual conversation, and that's exactly why the exam draws such a hard line between them. All three describe how quickly an athlete can move, but they differ in what triggers the movement and how much of it is planned in advance.
| Term | What it actually is | Planned or reactive? |
|---|---|---|
| Speed | Moving the whole body across ground in one direction as fast as possible | Planned, linear |
| Change-of-direction (COD) ability | Decelerating, redirecting, and reaccelerating the body along a new path | Planned, pre-determined route |
| Agility | COD ability plus a perceptual-cognitive component — reading a stimulus and responding with the correct movement | Reactive |
Underneath both qualities sits the same small set of mechanical variables. Get comfortable moving between them, because test questions frequently describe a scenario (longer ground contact, shorter stride) and ask what mechanical variable is implicated.
| Term | Definition |
|---|---|
| Stride length | Distance covered from one foot's ground contact to that same foot's next ground contact |
| Stride frequency | Number of strides completed per unit time — the "turnover" rate |
| Ground contact time | How long the foot stays in contact with the ground on a single step — shorter generally means faster, more elastic running |
| Impulse | Force applied over time (the area under a force–time curve); net impulse drives the change in the body's momentum |
| Rate of force development (RFD) | How quickly force climbs per unit of time — distinct from peak force, and the variable that matters most when contact time is short |
Why RFD outranks maximal strength in a sprint stride
Producing truly maximal voluntary force takes roughly 300 milliseconds or more to develop. Sport actions — a sprint step, a plant-and-cut, a jump takeoff — are typically over in 0–200 milliseconds. That mismatch is exactly why an athlete's peak strength on a 1RM test doesn't always predict how fast they are: what matters on the field is how much force gets produced in the time actually available, which is what RFD measures and maximal strength testing doesn't.
Three overlapping systems make an athlete fast: a nervous system that can drive muscle hard and quickly, muscle-tendon tissue that stores and returns elastic energy efficiently, and technique that keeps force pointed in a useful direction. Training speed means improving all three, not just one.
| System | What improves |
|---|---|
| Neural | Faster motor unit recruitment and rate coding, better intermuscular coordination, less wasted co-contraction of opposing muscle groups |
| Musculotendinous stiffness | A stiffer muscle-tendon unit stores and releases elastic energy faster, which shortens ground contact time and improves the stretch-shortening cycle |
| Technique | Force gets applied in a more horizontal, more useful direction, with less time wasted braking against the ground on landing |
The simplest useful model of running treats the whole support leg as a single linear spring holding up a point mass (the body's center of mass). On ground contact, the "spring" compresses and stores elastic energy; it then releases that energy to launch the body back into the air for the next stride. Leg stiffness — how much the spring compresses for a given load — is the model's key variable, and athletes who can hold a stiffer leg spring through ground contact tend to post shorter contact times and better running economy.
Where the model runs out of road
The spring-mass model treats the leg as one uniform spring with constant stiffness through the whole contact phase — a simplification that holds up reasonably well at moderate running speeds. At the very high forces and very short contact times seen in elite maximal-velocity sprinting, the leg doesn't behave like one simple linear spring: joint-level stiffness actually varies across the contact phase, and the model increasingly understates what's really happening at the hip, knee, and ankle individually. Treat it as a useful conceptual shortcut, not a literal description of elite sprint mechanics.
Contact time is one of the most test-friendly numbers in this chapter because the ranges for different movement types don't overlap much — which also makes them easy to mix up under exam pressure.
| Movement | Approximate ground contact time |
|---|---|
| Walking | ~600–700 ms per step |
| Jogging | ~250–300 ms |
| Sprinting (maximal velocity) | ~80–100 ms |
| Jumping (fast, reactive ground contact) | ~150–250 ms |
A way to keep the order straight
Contact time shrinks every time the movement gets faster and more elastic: walking always has a foot on the ground (that's what makes it walking), jogging cuts contact roughly in half again, and maximal-velocity sprinting is fast enough that ground contact drops below a tenth of a second. Reactive jumping sits above sprinting because absorbing and redirecting a landing force, even quickly, still takes longer than a sprint step's brief touch-and-go.
At the most basic mechanical level, running speed is the product of exactly two variables, and every technical or programming decision in this chapter ultimately tries to move one or both of them in a favorable direction.
| Mechanical variable | Elite sprinters | Novice sprinters |
|---|---|---|
| Ground contact time | Shorter (~80–100 ms) | Longer (~120–160 ms) |
| Stride frequency | Higher, without a length penalty | Lower |
| Stride length | Longer, driven by force production, not overreaching | Shorter |
| Peak vertical ground reaction force | Higher, applied in less time (higher RFD) | Lower, applied more slowly |
| Braking (horizontal decelerative) impulse | Small relative to the propulsive impulse | Larger — more force wasted braking on landing |
Those are illustrative ranges, not fixed cutoffs — but the pattern is consistent: what separates each tier is mostly shorter ground contact time and a bigger propulsive-to-braking impulse ratio, not some dramatically different stride length.
A sprint isn't mechanically uniform from the gun to the finish line — the body's job changes phase by phase, and so does what "good technique" looks like at that instant.
| Phase | What should be happening |
|---|---|
| Start / acceleration | Strong forward whole-body lean; powerful, extended drive through hip, knee, and ankle; short, forceful ground contacts; aggressive arm drive; low heel recovery |
| Transition | Torso progressively rises out of the acceleration lean; stride length keeps increasing while frequency stabilizes; ground contact time keeps shortening |
| Maximum velocity | Tall, near-upright posture; high front-side knee lift; active/"clawing" foot strike landing under the center of mass; full hip extension; relaxed shoulders and face |
| Deceleration | Progressively shortening stride; increased knee flexion to absorb force; controlled, slight backward trunk lean; widened base of support |
| Characteristic | What it looks like |
|---|---|
| Posture | Tall, with a slight whole-body lean from the ankle (not a bent-at-the-hips lean); hips stacked over the base of support; minimal lateral trunk rotation |
| Arm action | Relaxed shoulders, roughly 90° elbow angle held throughout, hands driving front-to-back rather than swinging across the midline |
| Foot strike | Lands under or just ahead of the center of mass with an active, "clawing" motion on the ball of the foot; minimal heel contact and minimal time spent absorbing the landing |
| Error | Likely cause | Coaching correction |
|---|---|---|
| Overstriding (landing well ahead of the center of mass) | Consciously reaching for stride length; weak hip flexors or limited hip mobility | Cue "quick feet down," de-emphasize reaching, use wickets/hurdle-mobility drills to reinforce landing under the hips |
| Excess forward trunk lean at max velocity | Carrying the acceleration-phase posture too long; weak core/hip extensors | Cue "run tall," posture-specific drills, targeted core and glute strength work |
| Poor backside mechanics (heel kicking up too high behind the body) | Delayed hip extension; over-cycling the recovery leg | Wall drills emphasizing front-side mechanics, A-skips, cueing a quicker ground-to-recovery transition |
| Excess vertical oscillation ("bouncing") | Pushing up into the ground instead of back against it; poor pelvic control | Cue low, horizontal force application; light resisted sprints to reinforce a forward-driving push |
| Arm swing crossing the midline | Poor shoulder mechanics or compensatory trunk rotation | Arm-drive isolation drills, cue "pocket to cheek" |
| Late or weak knee drive in acceleration | Insufficient hip flexor power | Resisted starts, wall drills, targeted hip flexor strength work |
Speed training splits into three broad approaches, and a well-built program usually draws from all three across a training year rather than committing to just one.
| Method | What it does |
|---|---|
| Resisted training | Adds an external load (sled, band, incline) so the athlete has to produce more horizontal force to move — targets acceleration-phase strength |
| Assisted (overspeed) training | Has the athlete move faster than their own unassisted max — targets stride frequency and the ability to tolerate higher movement speeds |
| Technique/form-based training | Unloaded drills (wall drills, A-skips, wickets) that isolate and reinforce a specific piece of sprint mechanics |
| Method | Category | General loading guidance |
|---|---|---|
| Sled towing | Resisted | Light loads preserve sprint mechanics; keep the velocity decrement modest to stay sprint-specific, load heavier only when the goal is explicitly strength-speed |
| Sled pushing | Resisted | Similar loading logic to towing, with a more forward trunk position and a strong emphasis on horizontal force production |
| Resisted bands / harness | Resisted | Coach- or partner-controlled tension over short accelerations; easy to titrate moment-to-moment |
| Overspeed towing | Assisted | Modestly faster than the athlete's own unassisted top speed; used sparingly — too much speed too soon breaks down mechanics |
| Downhill sprinting | Assisted | A shallow grade increases stride frequency without excessive mechanical disruption; too steep a grade encourages overstriding and braking forces |
The number that matters more than the load itself
For resisted sprinting to still train sprint-specific mechanics — rather than just turning into a generic strength exercise — the load has to stay light enough that the athlete's velocity decrement stays modest. Push the sled too heavy and the movement pattern stops resembling a sprint at all: contact times lengthen, posture changes, and the transfer to actual unloaded sprinting drops off. The same logic caps assisted work from the other direction — too much added speed and the athlete simply can't hold sprint mechanics together at all.
True agility layers a decision-making process on top of the same physical qualities that drive COD ability. That decision-making process is built from several distinct perceptual-cognitive skills, and a training program that only drills the physical side never fully trains the whole quality.
| Component | What it involves |
|---|---|
| Visual scanning | An efficient search strategy for pulling relevant information out of a busy visual field |
| Anticipation | Using early, subtle cues (an opponent's hips, a ball's early flight) to predict what happens next |
| Pattern recognition | Recognizing recurring structures in how play or movement unfolds, built from repeated exposure |
| Knowledge of situations | Contextual/game awareness — understanding what's likely given the score, position, or phase of play |
| Reaction time | How quickly the athlete responds once a relevant stimulus has actually been perceived |
| Decision-making speed | How quickly the athlete converts perceived information into the correct motor response |
Because COD ability is the purely physical half of agility, it responds to the same strength-and-power training that improves any explosive quality — just applied through movement patterns that specifically load deceleration, redirection, and reacceleration.
| Strength quality | Why it matters for agility |
|---|---|
| Maximal strength | Sets the ceiling for how much braking and reaccelerating force the athlete can ever produce |
| Reactive/eccentric strength | Determines how quickly the athlete can absorb landing forces and redirect them — the shorter the ground contact, the faster the change of direction |
| Rate of force development | Governs how much of that maximal strength is actually usable in the brief window a cut allows |
| Core and rotational strength | Stabilizes the trunk and transfers force between the lower body and upper body during a cut |
| Unilateral (single-leg) strength | Most COD actions plant, decelerate, and reaccelerate off one leg at a time — bilateral strength alone doesn't fully transfer |
| Movement | Coaching points |
|---|---|
| Backpedal | Stay low in an athletic stance; quick, short steps with weight on the balls of the feet; drive the knees back and down rather than reaching; keep the chest over the base of support; glance over the shoulder periodically to reorient |
| Side-shuffle | Stay low in an athletic stance; feet never cross; push off the trailing leg to generate lateral force; keep the shoulders and hips square to the target; quick, light ground contacts |
A third term worth separating out here is maneuverability — the ability to make fine, controlled adjustments to body position (a fake, a subtle weight shift) without necessarily completing a full change of direction. Test selection has to match the quality actually being measured.
| Test | What it measures |
|---|---|
| 40-yard dash | Linear sprint speed — acceleration phase plus approach to maximal velocity |
| 10-yard (10 m) sprint | Acceleration specifically, isolated from maximal-velocity mechanics |
| Pro agility (5-10-5 shuttle) | Closed-skill, pre-planned COD ability — lateral acceleration and deceleration under a known route |
| 3-cone (L) drill | Multidirectional, pre-planned COD ability combining a forward sprint with lateral and backpedal transitions |
| T-test | Pre-planned COD ability across forward, lateral, and backward movement in one sequence |
| Reactive agility test (with a stimulus) | True agility — COD ability plus the perceptual-cognitive/reaction component that a pre-planned test can't capture |
Where people lose points
A fast pro agility or 3-cone time tells you the athlete has good COD ability — it does not by itself tell you they're agile. Without an unpredictable stimulus built into the test, there's no perceptual-cognitive component being measured at all, no matter how many direction changes the course has.
COD and agility training follows a predictable progression, moving an athlete from fully controlled conditions toward the unpredictable demands of real competition.
| Stage | Characteristics | Example |
|---|---|---|
| 1 — Closed, planned, simple | Fixed pattern, known in advance, a single direction change | Single cone cut at a pre-set angle |
| 2 — Closed, planned, complex | Fixed pattern, known in advance, multiple direction changes | Pro agility shuttle, T-drill, 3-cone drill |
| 3 — Open, simple reactive | An external, low-complexity cue (a color, a number, a coach's hand signal) determines the route in real time | Cone drill with a called-out direction |
| 4 — Open, sport-specific reactive | A realistic opponent, ball, or game stimulus drives the response | Mirror drills, small-sided games, live 1-on-1 reaction drills |
Speed and agility work still follows the same acute variables as any other training — sets, reps, load, rest — but a few variables are specific enough to this kind of training that they deserve their own attention.
| What to track | How often |
|---|---|
| Sprint splits (10 m, 20 m, 40 m via timing gates or GPS) | Weekly to biweekly |
| Maximal velocity (GPS or radar) | Weekly to biweekly |
| Countermovement jump (as a neuromuscular readiness proxy) | Session-to-session or weekly |
| COD/agility test battery (pro agility, 3-cone, reactive test) | Every 4–6 weeks |
| Technique checklist / coach observation | Every session |
| Perceived exertion / fatigue and recovery status | Every session |
What is the one component that separates true agility from change-of-direction (COD) ability?
B. COD ability is the purely physical capacity to decelerate, redirect, and reaccelerate along a known route. Agility adds perception and decision-making — reading and responding to an unpredictable stimulus — on top of that same physical base.
Sprint speed is the direct product of which two variables?
C. Speed = stride length × stride frequency. Every other mechanical variable in this chapter ultimately acts on one or both of those two terms.
Why does rate of force development (RFD) matter more than maximal strength for sprint performance?
B. Maximal voluntary force takes roughly 300 ms or more to develop, while a sprint ground contact is typically well under 100 ms at max velocity. There simply isn't time to express peak strength — what matters is how much force can be produced in the brief window available.
Which movement type has the shortest typical ground contact time?
D. Maximal-velocity sprinting sits around 80–100 ms, shorter than jogging (~250–300 ms), reactive jumping (~150–250 ms), or walking (~600–700 ms).
Compared to a novice sprinter at maximum velocity, an elite sprinter typically shows which combination of traits?
B. Elite sprinters combine shorter ground contact time with a more favorable propulsive-to-braking impulse ratio — less force gets wasted decelerating the body on each landing.
What is the main limitation of the spring–mass model when applied to elite maximal-velocity sprinting?
C. The model's simplification — one linear spring with constant stiffness — holds up reasonably well at moderate speeds but increasingly misses what's really happening at the hip, knee, and ankle individually once forces and speeds reach elite sprinting levels.
A coach loads a sled heavily enough that the athlete's sprint velocity drops well beyond the recommended decrement threshold. What's the most likely consequence?
C. Once resisted-sprint loading pushes the velocity decrement too high, the movement pattern stops resembling an actual sprint — contact times lengthen and posture shifts — so transfer to real sprinting mechanics drops.
Which test is most likely to actually measure agility rather than pure change-of-direction ability?
C. Without an unpredictable stimulus the athlete has to perceive and react to, a test — no matter how many direction changes it includes — only measures COD ability, not agility.
Which strength quality is most directly responsible for how quickly an athlete can absorb a landing force and redirect it during a cut?
B. Reactive/eccentric strength governs how fast the athlete can absorb and redirect force during a brief ground contact — the shorter that contact, the faster the direction change can be completed.
A drill has an athlete perform a fixed 3-cone pattern that never changes from rep to rep. Where does this sit in the COD/agility drill progression?
C. A fixed, memorized route with multiple direction changes and no reactive element is closed-skill and planned — it's more complex than a single cut (Stage 1), but it still isn't reactive, so it belongs at Stage 2, not Stage 3 or 4.
This chapter is original study material covering topics also addressed in Essentials of Strength Training and Conditioning, 5th edition (NSCA/Human Kinetics). It is an independent work, not a reproduction or paraphrase of that text. This guide is not affiliated with, endorsed by, or sponsored by the National Strength and Conditioning Association (NSCA). CSCS® and Certified Strength and Conditioning Specialist® are registered trademarks of the NSCA.