Gaussian Channel: what it's actually best at

364 of 1,849 out-of-sample tests beat simply buying and holding the same asset — 19.7%. On the other 1,485 it did not. That is indistinguishable from the 20.1% rate across all 382 indicators we test (one pooled rate over all 660,005 tests we have run, not a mean of the per-indicator rates) — the difference is inside the margin this many tests can resolve, so read it as ordinary, not better or worse.

Each test is one asset on one timeframe: 1,849 of them, drawn from 902 assets across up to 4 timeframes. Not every asset has usable history on every timeframe, so that total is the grid we could actually run — it is not 902 × 4, and we do not pad it with tests we did not do. "Beat" means a higher return than holding that same asset over that same window. Measured out-of-sample — on data the setup was not chosen on.

Picking the single best timeframe for each asset after the fact raises it to 26.7% (241/902 assets). That number is the one worth distrusting: choosing the timeframe once you already know the answer is how backtests flatter themselves. Every indicator, ranked by this number

The Gaussian Channel is a price envelope built on John Ehlers' Gaussian filter — a low-pass smoothing filter in which an exponential moving average is applied repeatedly (each application is called a 'pole'), producing a centreline much smoother than a single EMA of the same length. More poles means more smoothing and MORE lag, not less — the filter's own author notes that larger inputs give smoother output with increased lag, and Ehlers' lag relation rises with pole count. Ehlers' low-lag comparison is against a Butterworth filter of equal poles, not against a simple average. Bands are drawn above and below that centreline at a multiple of a volatility measure (typically the true range), forming the channel. The multi-pole construction is the whole point: more poles give a sharper frequency cut-off, so short-lived noise is filtered out while genuine trends still bend the line.

How it's read. Read it like any envelope: the centreline's slope is the trend filter, and price interacting with the outer bands marks stretched conditions. The popular trend-following use — the one most TradingView implementations trade — goes long when the filter turns up or when price closes above the upper band, and stands aside when the filter points down.

Where it struggles by design. Like every smoothed-trend rule it is late at turning points by construction, and in a sideways market the channel flattens and whipsaws — the filter cannot distinguish a pause from a reversal until after the fact.

Origin: Filter design: John F. Ehlers (published in his cycle-analysis work). The widely-used charting implementation was popularised on TradingView (DonovanWall's open-source script).