Cadence Calculator Moving Average: How to Read It

A cadence calculator with a moving average smooths your raw step or stride data over a set window (for example, the last 30 seconds) so you see a stable, readable number instead of a jumpy one. You enter your cadence readings or a time series, choose the window length, and the calculator returns the averaged value at each point, plus the overall trend.

Formula

The moving average cadence is calculated as a weighted mean of the most recent raw value and the previous average. For a simple exponential moving average (EMA), the formula is:

EMA = (Raw × k) + (Previous EMA × (1 − k))

where k = 2 / (window + 1) for a standard EMA, or k = 1 / window for a simple average over the last N seconds. In the calculator above, we use a linear weighting based on the time window: if you enter a 30-second window, the new reading gets a weight of 30/60 = 0.5, and the previous average gets the remaining 0.5. This gives a smooth response that reacts quickly to real changes but ignores single-second spikes.

Variables:

• Raw cadence (c) – your latest measured steps per minute (or strides per minute for running).
• Previous average (p) – the moving average value from the prior update.
• Time window (t) – the length of the averaging period in seconds. A longer window (60 s) smooths more; a shorter window (10 s) reacts faster.

Worked Example

Using the default values from the calculator: window = 30 seconds, raw cadence = 170 steps/min, previous average = 168 steps/min.

1. Convert the window to a fraction of a minute: 30 / 60 = 0.5.
2. Weight the raw value: 170 × 0.5 = 85.
3. Weight the previous average: 168 × 0.5 = 84.
4. Add the weighted parts: 85 + 84 = 169.
5. Divide by the total weight (0.5 + 0.5 = 1.0) → 169 steps/min.

The moving average cadence is 169 steps/min, and the change from the previous average is +1 step/min.

Why Use a Moving Average for Cadence?

Raw cadence data from a watch or foot pod is noisy. A single step can be miscounted, a GPS glitch can add a false stride, or you might briefly slow down to avoid a pothole. The moving average filters out that noise so you can see your true cadence trend. Runners and walkers use this to:

• Check if they are maintaining a consistent stride rate.
• See the effect of a hill or a tailwind on their cadence.
• Compare cadence across different segments of a workout without being distracted by second-by-second jumps.

Coaches often recommend a target cadence (e.g., 170–180 steps/min for running), but a single raw reading rarely equals that target. The moving average gives you a fair comparison.

How to Choose the Window Length

The window length changes how the average behaves. There is no single “correct� value, but here are practical guidelines:

• 10–15 seconds – good for interval training where you want to see quick changes.
• 30 seconds – a balanced default that smooths most noise but still reacts within a minute.
• 60 seconds or more – best for steady-state runs or long walks where you want a very stable number.

If you are comparing your cadence to a published average (e.g., “elite runners average 180�), use a 30-second window because that matches how most sports watches compute their displayed average.

Common Mistakes and Misconceptions

One frequent error is mixing up steps per minute with strides per minute. A stride is two steps (one with each foot), so a cadence of 180 steps/min equals 90 strides/min. Always check which unit your device reports.

Another mistake is applying a simple arithmetic mean to a time series without weighting. If you just average all raw values equally, a long slow section will drag the average down even after you speed up. The moving average solves this by giving recent data more weight.

Finally, don’t confuse the moving average with the instantaneous cadence. The moving average lags behind reality by roughly half the window length. If your raw cadence jumps from 160 to 180, a 30-second EMA will show about 170 after 15 seconds, not immediately.

Limits of the Formula

The moving average is only as good as the raw data. If your device samples at 1 Hz (once per second), the average cannot be more precise than that. Also, the EMA formula assumes a constant time interval between samples. If your device records irregularly (e.g., only when GPS updates), the simple weighted formula above becomes inaccurate – you would need a time-weighted average that accounts for the actual gaps.

For walking, cadence is typically 100–120 steps/min; for running, 150–200 steps/min. The calculator works for both, but the default of 170 is aimed at runners. Adjust the raw input to your activity.

Alternatives to a Moving Average

If a moving average feels too laggy or too smooth, consider these alternatives:

• Median filter – takes the middle value of the last N samples. It removes outliers better than a mean but is slower to compute.
• Exponential weighted average with a faster decay – reacts quicker than a linear window but still smooths.
• No smoothing – use raw cadence for short sprints where every step matters.
• Stride length × cadence – instead of cadence alone, some runners track speed = cadence × stride length. The moving average can be applied to speed directly.

Frequently Asked Questions

What is a good cadence for running?

Most recreational runners have a cadence between 150 and 170 steps/min, while elite distance runners often hit 180 or higher. The “180� figure is a common benchmark, but it is not a rule – taller runners naturally have lower cadence. Focus on consistency rather than hitting a specific number.

Does a moving average change the final average?

No. If you average the entire moving average series, you get exactly the same result as averaging all raw values (assuming equal time intervals). The moving average only redistributes the weight across time; it does not invent or remove data.

Can I use this calculator for cycling cadence?

Yes. Cycling cadence is measured in pedal revolutions per minute (RPM). The same smoothing logic applies. Enter your RPM readings as the raw cadence, and the moving average will show your smooth pedal rate. Typical cycling cadence is 70–100 RPM.

Why does my watch show a different number than this calculator?

Watches often use proprietary smoothing algorithms, sometimes with a shorter or longer window, or they may display a 10-second average that updates every second. This calculator uses a transparent formula, so you can replicate the result if you know the exact raw values and window length. Differences are usually due to window length or how the watch handles gaps in data.

How do I get raw cadence data to input?

Most sports watches and phone apps export CSV or FIT files with per-second cadence. You can open the file in a spreadsheet, copy the cadence column, and paste the values into a time-series calculator. If you only have a single average, you cannot compute a moving average – you need the underlying raw data.

Related

• Binomial Distribution Equation Calculator
• Binomial Cumulative Distribution Calculator
• Binomial Distribution Calculator Keys Explained

in t | Time window (seconds) | s | 30 | 1
in c | Raw cadence (steps/min) | steps/min | 170 | 0
in p | Previous average (steps/min) | steps/min | 168 | 0
out avg | Moving average cadence | steps/min | (c * (t/60) + p * (1 - t/60)) / (t/60 + (1 - t/60)) | 1
out diff | Change from previous | steps/min | avg - p | 1