Your Garmin watch spits out six running dynamics metrics after every run. Cadence. Vertical oscillation. Ground contact time. Vertical ratio. Stride length. Ground contact time balance.
If you're anything like I was two years ago, you glance at them, nod wisely, and then forget them until the next run.
Here's the thing: some of these numbers are genuinely useful. Some are interesting but not actionable. A few are actively misleading if you don't understand what they're measuring. This article is my attempt to separate the signal from the noise - what each metric actually means, what a good number looks like, and when you should (and shouldn't) try to change it.
Cadence: the 180 myth and when it matters
Cadence (or stride rate) is the number of steps you take per minute. It's the running dynamics metric with the most attention and the most bad advice attached to it.
What the number actually means
Cadence is a proxy for how your legs are moving. A higher cadence means shorter, quicker strides. A lower cadence means longer, slower strides. That's it. Everything else people claim about cadence - that it prevents injury, that it makes you faster, that there's a magic number - is downstream of this fundamental tradeoff.
The famous "180 steps per minute" rule comes from Jack Daniels' 1984 study, where he observed elite distance runners at the Olympics and found that almost all of them took at least 180 steps per minute during racing. This got simplified into "everyone should run at 180 cadence" somewhere along the way.
The reality: that study observed elites running at 5:00–5:30/mile pace. At that pace, 180 cadence is basically required because the alternative is overstriding - reaching your foot out so far in front of your body that you're braking with every step. At 8:00 or 9:00/mile pace, 180 might be aggressively high. At 11:00/mile recovery pace, it could be counterproductive.
What good actually looks like
There is no universal optimal cadence, but there are ranges that tend to work well:
- Easy runs (recovery, Z2): 155–170 is normal. Your legs are tired, your turnover is naturally slower.
- Tempo / threshold runs: 165–175. As pace increases, cadence naturally rises.
- Race pace (5K / 10K / HM): 175–185 for most competitive runners.
- Sprints / strides: 190+.
The number that matters isn't your raw cadence against some universal target. It's your cadence relative to your height, your pace, and your comfort. A 6'4" runner will naturally have a lower cadence than a 5'4" runner at the same pace.
When cadence changes actually help
There are three scenarios where consciously changing cadence makes sense:
-
Overstriding. If your foot lands significantly ahead of your center of mass (visible on video, or felt as a braking sensation with each footstrike), increasing cadence by 5–10 steps per minute will naturally shorten your stride and bring your footfall closer to your body. This reduces braking forces and often fixes the problem without you thinking about anything else.
-
Chronic injury patterns. Some runners with recurring shin splints, IT band issues, or plantar fasciitis find that a slightly higher cadence reduces impact forces enough to stay healthy. This is individual - but it's worth a 4–6 week experiment at +5% cadence if you're stuck in an injury cycle.
-
Slow turnover as a limiter. If you're a decent runner who can't seem to get faster despite adequate fitness, and your cadence at 5K effort is below 170, you might have a neuromuscular turnover problem. Strides, hill sprints, and cadence drills can help.
When to ignore cadence entirely
If your cadence at easy pace is 155–165 and you have no injury issues and you're running fine, do not force it to 180. You'll just make your runs harder, stress your calves, and probably end up with a new injury from the change itself. A forced 10% cadence increase changes your running mechanics in ways that can cause problems if done too fast. The rule of thumb: no more than a 5% increase per week, with at least 4 weeks at the new pattern before evaluating.
Vertical oscillation: the efficiency proxy everyone misreads
Vertical oscillation (VO) measures how much your torso bounces up and down with each stride. The number is usually 6–10 centimeters (around 2.5–4 inches) for most runners.
What it tells you
A lower vertical oscillation generally means better running economy - you're converting more horizontal force into forward motion, less into wasted upward bounce. Think of it as the running equivalent of pedalling in smooth circles on a bike instead of mashing up and down.
Elite distance runners often have vertical oscillation in the 4–6 cm range. Recreational runners are typically 7–10 cm. There's a clear correlation: lower VO ≤ higher efficiency at a given pace.
But this is one of those metrics where the obvious interpretation is dangerously incomplete.
Why "lower is better" can backfire
The trap is that you can reduce vertical oscillation by compromising your running mechanics in worse ways. Specifically:
-
Shortening your stride. If you take shorter, quicker steps, your vertical bounce naturally decreases. But if the stride is too short, you're not generating enough forward propulsion, so your pace drops or your effort rises. You've gained efficiency in one dimension and lost it in another.
-
Stiffening your ankles and knees. Some runners try to "run smoother" by reducing vertical motion through muscular tension. This increases energy cost, reduces elastic return from tendons (especially the Achilles), and can lead to injury. The efficient runner isn't rigid - they're compliant, using elastic recoil.
-
Chasing the number. Vertical oscillation is correlated with pace. At 7:00/mile, you'll naturally bounce less than at 9:00/mile. If you compare your VO at different paces and worry about the difference, you're wasting energy on a meaningless comparison.
What to actually look at
The more useful metric is vertical ratio - vertical oscillation divided by stride length. This normalises for pace and body size. A vertical ratio below 6–7% is generally good. Above 9% suggests you're bouncing more than you should relative to how far you're covering.
The vertical ratio tells you a better story: are you getting efficient forward motion per unit of bounce? A tall runner with 8 cm VO and long strides can have a better vertical ratio than a short runner with 6 cm VO and short strides.
Here's what I watch in my own data on Baseline:
- Is VO stable across similar-effort runs? If my VO at the same pace jumps from 8 cm to 10 cm over a few weeks, it's a sign something changed - fatigue, form degradation, or a new shoe with different stack height.
- Does VO correlate with perceived effort? On days where my legs feel heavy and my VO is up, it's a fatigue signal. I back off.
- Trend over months, not days. Vertical oscillation changes slowly and should be evaluated over 4–8 week blocks, not individual runs.
Ground contact time: the fatigue detector
Ground contact time (GCT) is the duration your foot is on the ground with each step, measured in milliseconds. Typical range: 200–300 ms.
What short vs long GCT tells you
A shorter ground contact time generally indicates better running economy - you're spending less time braking against the ground and more time moving forward. Elite runners often have GCT around 160–200 ms at race pace. Recreational runners are more commonly 240–300 ms.
But here's what's more interesting: ground contact time is one of the most sensitive acute fatigue markers available from a wrist-based sensor.
When you're fresh, your nervous system fires quickly, your tendons are springy, and you're off the ground fast. When you're tired - from a hard workout, insufficient recovery, or accumulated training load - your GCT creeps up. Your foot stays on the ground 10–20 ms longer. You're spending more time in the braking-and-propulsion phase because your system can't rebound as efficiently.
This makes GCT a useful recovery check:
- Fresh legs at easy pace: ~240 ms
- Same pace, fatigued: ~260–270 ms
- Significant fatigue: 280+ ms at easy pace that's normally 240
The sensitivity is real. I've seen my GCT rise 15–20 ms after a hard block, normalize during a recovery week, and rise again the next block. It's more responsive than resting heart rate and nearly as responsive as HRV as a fatigue marker.
What affects GCT accuracy
GCT from a wrist-based optical sensor is an estimate, not a measurement. It's derived from accelerometer data, not force plates. It's directionally correct but not absolute. The trend (within the same device) is more reliable than the absolute number.
Factors that throw it off:
- Wrist motion. If your watch is loose, GCT readings degrade.
- Surface changes. Soft trail vs hard road changes GCT mechanically.
- Downhill running. GCT increases downhill regardless of fatigue.
- Shoe changes. Maximalist shoes can change GCT vs minimal shoes.
The asymmetry angle
Garmin also reports ground contact time balance - the left/right split. This is wildly underrated. A consistent asymmetry of 52% / 48% or worse suggests either:
- A leg length discrepancy (actual or functional from tightness / imbalance)
- A compensation pattern from an old injury
- A form issue that's loading one leg more than the other
Fixing GCT asymmetry can prevent overuse injuries. If you're 52%+ on one leg for months, that leg is taking more cumulative load. Address it before it becomes a problem.
How Baseline surfaces these trends
The problem with running dynamics data isn't that it's useless - it's that it's fragmented across devices, buried in workout detail screens, and almost never shown as trends.
Baseline pulls these metrics from Garmin (and other sources that record them), aligns them on a timeline, and shows you the trajectory. Your cadence at threshold pace over the last 12 weeks. Your GCT trend against your CTL (chronic training load). Your vertical ratio at easy pace week over week.
The insight isn't in any single run. It's in the line connecting them - the slow drift upward in GCT as fatigue accumulates, the stable cadence across paces after form work, the vertical ratio that improves through a build block and regresses when you're overtrained.
I check mine every Sunday morning. A 10-second scan of the trend chart tells me more about where my running form is than any individual run's data dump ever could.
The bottom line on running dynamics
Here's my ranking of these metrics from most to least actionable:
- Ground contact time - best acute fatigue signal. Watch the trend.
- Cadence - useful when it's extreme (below 155 or above 190 at normal pace), useful for fixing overstriding, largely irrelevant in the middle range.
- Vertical ratio - better than raw VO, generally correlates with efficiency, slow to change.
- Ground contact time balance - underrated asymmetry detector. Check it quarterly.
- Vertical oscillation - fine as a direction of travel, easy to overthink.
- Stride length - derived from the others. Not independently useful.
Don't obsess over any single number. The goal of running form work is not to optimise a dashboard - it's to feel smooth, stay healthy, and run faster with the same effort. The numbers are clues, not verdicts.