A moving average smooths a price series by averaging the last n closes and recalculating as each new bar prints. It is the most widely used trend reference in technical analysis, mainly because it is simple and everyone can compute the same number.
Simple vs exponential
- Simple moving average (SMA): the plain arithmetic mean of the last n closes. A 20-period SMA gives each of the last 20 bars exactly one-twentieth of the weight, and a bar's influence vanishes the moment it leaves the window.
- Exponential moving average (EMA): a weighted average that gives more influence to recent closes. Each new value is the previous EMA plus a fraction (2 ÷ (n + 1)) of the gap between the latest close and that previous EMA. It responds to new moves sooner and never fully forgets older bars.
Because of that weighting, when price turns the EMA usually bends first and the SMA follows.
How to read price against an average
- Price above a rising average describes an uptrend on that timeframe; below a falling average, a downtrend.
- The distance from price to the average, as a percentage, describes how stretched the move is. "The 20-EMA sits 11% below price" means price has run well ahead of its recent average.
- Averages lag by construction. They confirm what has happened; they do not anticipate turns.
In MoonWire analysis
Our multi-timeframe reads use the 20-period EMA to set trend direction on each timeframe: a close more than 0.3% above it is "up", more than 0.3% below is "down", and anything inside that band is "sideways". The reads also quote the 20-period SMA and EMA side by side with their distance from price — for example, Ethereum's 3 September 2026 read put the 3-day SMA20 14.24% and EMA20 11.38% below price.
The market regime scoring uses a moving average too: Bitcoin's close relative to its 20-day SMA is part of how a risk-on or risk-off call is graded.