Garmin calls it "oscillation." Polar labels it "running index." Whichever device you train with, there's a number quietly appearing in your post-run summary that most athletes scroll past without a second thought: vertical oscillation.
That's a mistake. Among the biomechanical metrics now available to age-group runners, vertical oscillation may be the most direct proxy for wasted energy. Understanding what it measures, what drives it, and how to reduce it can meaningfully improve your running economy, and your race times, without adding a single kilometer to your training log.
What Vertical Oscillation Actually Measures
Vertical oscillation (VO) quantifies how much your center of mass bounces up and down with each stride. It's measured in centimeters and captured via an accelerometer in your running chest strap or, on newer devices, directly from an optical wrist sensor.
The principle is straightforward: forward propulsion is the goal. Any vertical movement is, in mechanical terms, a diversion of energy in the wrong direction. A runner with 12 cm of vertical oscillation is generating a significant upward force with every step that contributes nothing to their pace, it just launches them slightly off the ground and then costs them the impact energy of landing.
Population ranges from recreational to elite give a useful reference frame:
- Recreational runners: 9–13 cm
- Trained age-groupers: 7–10 cm
- Elite distance runners: 5–8 cm
Sub-6 cm figures appear regularly in world-class marathoners and track athletes, runners who look, from a distance, like they're gliding rather than bounding.
Why It Matters for Running Economy
Running economy, the metabolic cost of running at a given pace, is the performance variable that separates athletes of similar VO2 max. A runner with superior economy uses less oxygen per kilometer, arrives at the finish fresher, and has a lower physiological ceiling on sustainable race pace.
Vertical oscillation is one of the primary biomechanical contributors to running economy because the energetic cost of bouncing is real and compounding. Every centimeter of unnecessary vertical travel requires muscular work to generate on the way up and absorbs kinetic energy on the way down. Multiply that across 90 strides per minute, 60 minutes of racing, and the cumulative oxygen cost of excessive bounce becomes substantial.
Research examining elite marathoners consistently shows that lower vertical oscillation correlates with better running economy and faster race times, even when controlling for VO2 max. The mechanism isn't complicated: athletes who stay lower to the ground redirect more of their muscular output horizontally.
The Relationship to Other Biomechanical Metrics
Vertical oscillation doesn't exist in isolation. It's tightly coupled to several other metrics that wearables now track.
Cadence. Step rate is probably the most direct lever for oscillation. Lower cadence tends to correlate with longer strides, greater ground contact time, and more vertical displacement. When athletes increase their cadence by 5–10%, vertical oscillation typically drops, shorter, quicker steps don't launch the center of mass as high.
Ground contact time. Time spent on the ground and time spent in the air are inversely related. Runners with long ground contact time tend to apply force more gradually and over-stride, contributing to both excessive vertical travel and braking forces. Tightening up ground contact generally reduces oscillation simultaneously.
Stride length. Over-striding, landing with the foot well ahead of the center of mass, is a common cause of high vertical oscillation. When the foot strikes too far forward, the body must vault over it, producing upward displacement. A shorter, faster stride with midfoot contact underneath the hip reduces this vault.
Vertical ratio. Garmin introduces a composite metric called vertical ratio: vertical oscillation divided by stride length, expressed as a percentage. It normalizes oscillation relative to how much forward distance each stride covers. A ratio under 8% is generally considered efficient; above 10% suggests meaningful biomechanical waste. This metric is arguably more useful than raw oscillation because it accounts for taller athletes and longer natural stride lengths.
What Causes High Vertical Oscillation
Several biomechanical and physiological factors drive excessive bounce, and they're worth isolating because the interventions differ.
Weak hip extensors. The glutes and hamstrings are responsible for driving the leg backward and propelling the body forward. When these muscles are underpowered, athletes compensate by using calf-dominant plantarflexion to push off, a mechanism that generates more upward force than forward force.
Over-striding. Landing with the foot in front of the body's center of mass creates a braking force and forces the trunk to rise over the planted foot. This is the single most common cause of high oscillation in recreational runners transitioning from walking gait patterns.
Low cadence. Cadence under 165 steps per minute at aerobic paces is often associated with bouncy mechanics. The physics are simple: more time in the air means more vertical displacement before returning to the ground.
Poor trunk stability. Excessive vertical movement sometimes originates from the core rather than the legs. If the trunk can't stay stiff during the power application phase, energy leaks upward rather than translating into forward velocity.
Fatigue. Even athletes with efficient mechanics at fresh tend to see oscillation creep up in the final kilometers of long runs and races. As the glutes and core fatigue, compensatory movement patterns emerge.
How to Reduce Vertical Oscillation
Reducing vertical oscillation is a training adaptation, not a quick fix. The following interventions are evidence-supported and practically achievable for age-group athletes.
Cadence work. The fastest route to lower oscillation is usually increasing step rate. Use a metronome app targeting 170–180 steps per minute during easy runs. Don't chase 180 as a magic number, aim for 5–8% above your current natural cadence and let the adaptation happen over four to six weeks. The reduced stride length that comes with higher cadence is the primary mechanism driving oscillation down.
Glute and hip extensor strengthening. Single-leg work is particularly effective: Bulgarian split squats, single-leg Romanian deadlifts, and hip thrusts. Two strength sessions per week targeting these muscle groups will improve force application direction over a training block of eight to twelve weeks.
Running drills. A-skips, high knees, and butt kicks train the movement pattern of keeping contact time short and driving the leg down and back rather than up and forward. Incorporate 10–15 minutes of drills before speed sessions.
Hill running. Uphill running naturally encourages more forward lean, shorter ground contact, and less vertical displacement. Including hill repeats in your weekly schedule reinforces efficient mechanics that transfer to flat terrain.
Posture cues. Running tall with a slight forward lean from the ankles, not the waist, keeps the center of mass moving forward rather than bouncing. A common cue is "tall spine, fall forward", think of falling into a controlled lean rather than pushing off the ground.
How to Interpret Your Own Data
Raw oscillation numbers need context. Here's a practical framework for making sense of your own metrics.
Start by establishing your baseline over several easy runs. Oscillation is pace- and fatigue-sensitive, so comparing across different conditions creates noise. Easy aerobic runs at consistent effort give the cleanest picture.
If your vertical ratio (if your device calculates it) is above 10%, treat that as an actionable inefficiency worth addressing. If raw oscillation is above 10 cm at easy paces, cadence and hip strength should be your first interventions.
Track trends over a training block, not single sessions. A meaningful reduction in oscillation, say, 1–2 cm, over eight weeks of targeted work is a genuine biomechanical improvement. Session-to-session variation is normal and not diagnostic.
Cross-reference with running economy proxies. If cadence-adjusted heart rate at a given pace is decreasing over the same period your oscillation is dropping, the efficiency improvement is real.
One important caveat: some vertical oscillation is necessary. The spring-mass model of running depends on elastic energy storage and return through the tendons. Attempting to completely eliminate bounce would require constant muscular activation that's metabolically more expensive than a small amount of controlled vertical displacement. The goal isn't zero oscillation, it's optimal oscillation for your anthropometry and pace.
Tracking Tools and Devices
Vertical oscillation is available across most major running hardware platforms. A chest strap remains the most accurate capture method, wrist-based accelerometry introduces more noise, though newer optical sensors have closed the gap considerably.
Garmin's HRM-Run and HRM-Pro straps provide oscillation, vertical ratio, and ground contact time balance. Polar's H10 paired with Polar watches gives similar data under the "running power" ecosystem. Wahoo's TICKR Run captures the same metrics for Wahoo and third-party device users.
In the Triforge platform, these metrics feed directly into the AI coach's running dynamics analysis, allowing trend tracking across training blocks and flagging outliers that correlate with form breakdown or fatigue accumulation.
The Efficiency Equation
Vertical oscillation is a rare metric in the wearable data landscape: it's mechanically meaningful, actionable, and responsive to targeted training. Unlike VO2 max, which changes slowly and is largely genetically constrained, oscillation can shift meaningfully within a single training block with the right interventions.
For competitive age-groupers looking for performance gains beyond volume and intensity, this is exactly the kind of metric worth pursuing. Not because a lower number looks better on a dashboard, but because less bounce translates directly to more efficient running, lower oxygen cost, and, eventually, faster races at the same effort.
→ Lower oscillation
→ Less bounce
→ Forward force, not up