A moving average smooths price so you can see direction without reading every candle. It is a lagging tool: it describes what price has already done and does not predict what comes next. Used for that purpose, it is useful. This page covers the two main types, how they are calculated, which settings traders commonly use, and where they break down.
SMA vs EMA: the quick comparison
Both average recent closing prices. They differ in how much weight each bar gets.
| Feature | Simple Moving Average (SMA) | Exponential Moving Average (EMA) |
|---|---|---|
| Weighting | Every bar in the window counts equally | Recent bars count more |
| Reaction speed | Slower | Faster |
| Smoothness | Smoother, but drops old bars abruptly | Slightly noisier, with no sudden drop-off |
| Typical use | Widely watched levels such as the 50 and 200 | Short-term trend tracking and dynamic pullback zones |
| Main weakness | Lags more at turning points | More false signals in choppy markets |
Neither is better in general. The EMA's speed helps when a trend is clean and hurts when price is chopping sideways.
How each one is calculated
SMA
Add the last N closing prices and divide by N. A 10-period SMA on closes of 1.1000, 1.1010, 1.1020 and so on is just the plain average of those ten values. When a new bar closes, the oldest bar falls out of the window.
EMA
The EMA uses a multiplier: 2 ÷ (N + 1). For a 20-period EMA that is about 0.0952.
- Start with a seed value, usually the SMA of the first N bars.
- Each new bar: EMA = (Close − previous EMA) × multiplier + previous EMA.
Because the EMA is recursive, old data never fully disappears. Its fading influence is why two platforms can show slightly different EMA values depending on how much history they load and how they seed it. Small differences like this are normal.
Settings by timeframe: starting points, not answers
No setting is "best." The common ones work partly because many traders watch them, and that crowding is a reason to know them, not proof they have an edge. These are typical starting points to test, not recommendations:
- M5–M15 (intraday): 9 and 21 EMA for short-term direction, with a 50 as a context line. Expect many signals and many false ones.
- H1–H4 (swing): 20 or 21 EMA for pullbacks, 50 SMA or EMA for trend bias, 200 as a broad filter.
- Daily: 50 and 200 SMA are the classic pair. Crossovers are infrequent and late by design.
- Weekly: 10, 20 or 40-period averages, used mostly for context rather than entries.
Also note that a 200-period average on M15 covers about two trading days, while on the daily chart it covers roughly 40 weeks. The number means very different things on different timeframes.
Practical ways to use them
1. Trend filter
Price above a rising 50 or 200 average suggests an uptrend. Price below a falling one suggests a downtrend. Use this to decide which direction you are willing to trade at all, not as a trigger.
2. Slope matters more than position
A flat average with price crossing it back and forth is a warning. Before trusting any signal, ask whether the average is clearly sloping.
3. Pullback zones
In a trend, price often retraces toward a 20 or 50 average. Look for reaction there, but confirm with price action such as rejection candles or a structure break. "Often" is not "always," and strong trends can slice through these levels.
4. Crossovers
A fast average crossing a slow one is a lagging confirmation. It can keep you in a big move, but it usually arrives after a meaningful part of the move has happened.
Recommended for youMACD Explained: How It Is Calculated, Best Settings, Signals and When It FailsFailure modes
- Range whipsaws: In sideways markets, crossovers fire repeatedly and lose small amounts each time. This is the most common failure.
- Lag at reversals: Averages turn after price does. Late entries mean wider stops or worse reward-to-risk.
- News spikes: A single volatile candle can drag an EMA sharply, then reverse, giving a false signal.
- Curve-fitting: Tuning 13/34 versus 12/36 until a past chart looks perfect usually produces settings that fail on new data.
- Platform differences: Daily candle close times (for example, a 5 p.m. New York close versus a server-time close) change closing prices and therefore the averages.
- Self-fulfilling levels failing: Popular levels like the 200 can hold repeatedly and then break without warning.
Pairing them with other tools
A moving average answers "which way is price drifting?" It says nothing about how strong the move is or how stretched it has become. Traders often add a trend-strength or momentum tool, but stacking similar indicators just repeats the same information. Our guide on which trend and momentum indicators to combine without redundant signals covers how to avoid that. For specific companions, see MACD (itself built from EMAs), RSI and Stochastic, ADX settings for trend strength, and Bollinger Bands settings, which are built around a moving average. For the wider picture, start with Trend & Momentum Indicators Explained.
Should you pay for an "enhanced" moving average?
Standard SMA and EMA come free on every major platform. Some vendors sell adaptive or "smart" variants and tout reduced lag. Reduced lag usually means more sensitivity, which brings its own false signals, so the tradeoff does not disappear. If you are weighing a paid option, read Enhanced MACD, RSI and Stochastic Indicators vs the Standard Versions: Are the Upgrades Worth Buying? first, and confirm current features and pricing on the vendor's official site. Whatever you consider, test it on a demo account across both trending and ranging periods before risking real money.
A simple testing routine
- Pick one pair and one timeframe.
- Mark a trending stretch and a ranging stretch from past charts.
- Apply one SMA/EMA setup and record every signal in both stretches.
- Note the results honestly, including the range losses.
- Forward-test on a demo account before going live.
Forex trading carries a high risk of loss, especially with leverage. Nothing here is personal financial advice. If you are unsure whether trading suits your situation, consider speaking with a qualified financial professional.
Keep reading
RSI and Stochastic Oscillator Explained: Calculation, Timeframe Settings and Failure ModesRSI and the Stochastic Oscillator are the two momentum oscillators most traders meet first. They lo…
Which Trend and Momentum Indicators to Combine Without Redundant SignalsMost cluttered charts share one flaw: five indicators that all say the same thing. A moving average…
Enhanced MACD, RSI and Stochastic Indicators vs the Standard Versions: Are the Upgrades Worth Buying?The short answer Usually, no: not as a standalone purchase. An "enhanced" MACD, RSI or Stochastic i…