Your data
Paste a table straight from Excel or Google Sheets, or drop a CSV file. Nothing is uploaded anywhere — the whole calculation runs inside this browser tab.
CSV, TSV or plain text. Files saved in windows-1251 are detected and re-read automatically, so Cyrillic headers do not turn into garbage.
Parsing options
Columns
Result
Save and share
The drawing library is downloaded only when you press the button, and only on this page.
This page has no server side at all. The table you paste is parsed by JavaScript inside your own browser, the chart is drawn on a canvas element on your machine, and the export file is assembled locally. Nothing is uploaded, stored or written to any log. You can disconnect from the network after the page has loaded and everything will still work — which is the simplest way to verify the claim yourself.
Facts and limits of this method
| Method | Simple moving average — unweighted mean of the last N values |
|---|---|
| Window | Chosen automatically: 7, or a third of the series length if that is smaller |
| First values | Left empty until a full window is available, rather than averaged over fewer points |
| Lag | Roughly half the window — a 7-point average reacts about 3 points late |
| Chart | Original series in grey behind the smoothed line |
| Minimum rows | 3 |
When it misleads you
- A moving average always lags. It reacts to a turning point about half a window after it happens, which is not a defect to be fixed but the direct consequence of averaging past values. On a 7-day window a change on Monday becomes visible around Thursday.
- The window length determines what you see. A short window leaves noise in, a long one can smooth away the very change you were looking for, and there is no objectively correct choice — only one appropriate to the cycle in your data.
- The average is dragged by outliers exactly as any mean is. A single enormous day raises the line for the whole window it belongs to, producing a plateau that looks like a sustained shift and is not.
- Smoothing removes the variability, and the variability is sometimes the finding. A process whose daily numbers doubled in spread while holding the same average looks completely unchanged after smoothing.
How it is calculated
Each point is replaced by the unweighted mean of itself and the preceding values inside the window. Nothing is weighted and nothing is centred, which makes this the simple moving average — the version that lags by half a window rather than the centred version that cannot be computed at the end of a series.
The window is chosen as 7 or a third of the series length, whichever is smaller. Seven is a sensible default because most business data carries a weekly cycle and a seven-point window cancels it exactly; the one-third cap prevents a short series from being reduced to two or three usable points.
The first few positions have no full window behind them and are left empty rather than averaged over whatever happens to be available. Averaging over three points and calling it a seven-point average would make the start of the line jumpier than the rest and hide that fact.
The chart draws the original series in grey behind the smoothed line, because a smoothed line on its own conceals how much was removed. Seeing the two together is what makes the smoothing honest.
Questions and answers
Why does my smoothed line start late?
Because the first positions have no full window behind them. Averaging fewer points would produce values that look comparable but are noisier, so they are left blank instead.
Can I choose the window myself?
Not on this page — it is derived from the length of the series so that the same data always produces the same result. Export the values and change the window in a spreadsheet if you need to experiment.
Which window should I want?
Match it to the cycle in your data: 7 for daily numbers with a weekly rhythm, 12 for monthly numbers with an annual one. A window that does not match the cycle leaves the cycle in the smoothed line.
Moving average or trend line?
A moving average follows local changes and shows turning points. A regression line assumes one constant direction across the whole period. Use the average to see what happened and the regression to describe the overall direction.
Is my data uploaded?
No. Smoothing and drawing both happen in your browser.