EMA Smooth¶

Math · Temporal

Exponential moving average of ANY series (not just price) — smooth a derived signal, an indicator, a spread. α = 2 ÷ (length + 1), seeded at the first value (matches EMA with adjust=off). Expects a finite series.

EMA Smooth is an exponential moving average that works on any series, not just price. It weights recent bars more heavily than old ones, so it reacts faster to change than a simple average while still filtering noise. Reach for it when you want a responsive smooth of a derived signal — an oscillator, a spread, a score.

How it works¶

Each bar the output is α × value + (1 − α) × previous, with the smoothing factor α = 2 ÷ (length + 1). It's seeded at the first value (equivalent to adjust = off), which matches how the price EMA is built, so behaviour is consistent and export-faithful. It expects a finite input; feed it a clean series. Lookahead-free.

When to use it¶

Use it instead of Rolling Mean when responsiveness matters more than an equal-weighted average — smoothing a fast oscillator you intend to cross, or a live spread you're trading around. The lower the length, the closer it hugs the raw input; the higher, the smoother and laggier.

Example¶

Responsive momentum line: wire a Momentum reading into EMA Smooth (length = 8), then trigger on it crossing zero with CrossesAbove — a cleaner momentum-flip entry than the raw, jagged line.

Tips & gotchas¶

Recency-weighted — reacts faster than Rolling Mean, at the cost of a little more noise.
Seeded at the first value (adjust = off), matching the price EMA — so exports stay faithful.
Feed it a finite series — long NaN runs upstream can poison the recursion.

Inputs¶

Socket	Type	What to wire in
In	`series` / `bars`	Source series. BARS input uses close.

Outputs¶

Output	Type	Plots as	Description
ema	`series`	Line · sub-pane	EMA-smoothed series.

Parameters¶

Parameter	Type	Default	What it does
Length	number · 1–1000	`10`	EMA length; α = 2 / (length + 1).

Reference auto-generated from the block catalog · category Math.