Moving Averages: what they are, and whether they work

23,368 of 121,111 out-of-sample tests beat simply buying and holding the same asset — 19.3%. On the other 97,743 it did not. That is below 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).

One test is one variant on one asset on one timeframe. We tested 72 variants of Moving Averages, so an asset that carries several of them contributes several tests — this is the family's record across everything we ran, not a count of assets. Those variants are not alike: individually they beat buy-and-hold between 10.3% and 29.0% of the time, and 19.3% is the pooled rate over all of them, not a property every version of this indicator has. The table below gives each one separately. "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.

A simple moving average (SMA) is the arithmetic mean of the last N closes. An exponential moving average (EMA) applies a smoothing factor of 2/(N+1) to each new bar and retains the rest of its previous value, so weight decays geometrically into the past and nothing ever drops out entirely. A weighted moving average assigns linearly declining weights instead. All three trade the same thing off: a longer window is smoother and slower, a shorter one is noisier and faster.

How it's read. The usual readings are price crossing the average, and a faster average crossing a slower one — the crossover. The slope of the average is itself read as trend direction, and long averages such as the 200-period are widely watched as reference levels irrespective of any signal.

Where it struggles by design. Every moving average lags by construction: it is an average of prices that have already happened, so it cannot turn before price does. In a sideways market price crosses the average repeatedly and crossover readings produce signals in both directions in quick succession.

Origin: In use in technical analysis since at least the early twentieth century; the exponential form was popularised in the 1960s.