The moving average is probably the most widely used tool on any FX chart, and also one of the most oversold. It does not predict anything. It summarizes what price has already done. Once you accept that, it becomes a useful, honest instrument for describing a market's recent behaviour.
What it actually is
A simple moving average (SMA) adds up the last N closing prices and divides by N. A 20-period SMA on a daily chart is the average close of the last twenty days. Each new day, the oldest value drops off and the newest is added, so the line "moves." Because every price in the window has equal weight, a sudden jump affects it slowly, and when the old value drops out of the window the average can shift abruptly.
An exponential moving average (EMA) weights recent prices more heavily using a multiplier of 2 / (N + 1). A 20-period EMA uses about 9.5 percent of the newest close and passes the rest down from the previous EMA value. It reacts faster than an SMA of the same length, which is the whole point, and also its main weakness: faster reaction means more false turns.
Lag is not a bug
Any smoothing filter introduces delay. A longer average smooths more and lags more; a shorter one tracks price closely but wiggles. There is no setting that is both smooth and instantaneous. Many traders switch from a 50 to a 21 to a 9 hoping to fix the lag, then complain about whipsaws. They have only traded one problem for the other.
What moving averages are decent at
- Describing trend direction. If price is above a rising 50-period average, recent behaviour has been upward. That is a description, not a promise.
- Giving a dynamic reference. In an orderly trend, price often pulls back toward a moving average before continuing, which is why some traders treat it as a loose zone, not a precise line. Same principle as in support and resistance zones.
- Filtering noise. Reducing a jagged chart to a smooth line makes it easier to see the broad shape.
- Standardizing rules. A rule such as "only look for long setups when price closes above the 100-period average" is objective and testable, which matters for honest backtesting.
Where they mislead
- Ranges. In a sideways market, price crosses the average back and forth. Every cross looks like a signal, and each one loses a bit to the spread (see the true cost of trading).
- Turning points. By design they confirm a move after it has begun. You will usually enter late and exit late.
- Self-fulfilling talk. The 200-day average is popular, and sometimes price reacts near it. But it does not always, and plenty of traders remember the reaction and forget the many times price sliced through.
- Parameter hunting. If you test fifty combinations and pick the best one, you have probably found noise that fit the past.
Crossovers: what the evidence suggests, and what to be wary of
The classic system buys when a fast average crosses above a slow one and sells when it crosses below. On trending instruments this can capture large moves; on choppy ones it produces long strings of small losses. The profile is typical of trend following: a low win rate, a few big winners, and the discipline to sit through losing streaks. If you cannot tolerate a 35 to 40 percent hit rate, this style will test you (see losing streaks). You do not have to take the example numbers at face value; test it on your own pair and timeframe, including costs.
Pairing with volatility
An average shows direction but not the size of normal fluctuations. Combining it with a volatility measure helps. For example, measuring the distance from price to the average in units of ATR tells you whether price is stretched or close to the mean. See ATR explained. Placing a stop beyond a multiple of ATR, instead of an arbitrary pip count, keeps your risk in proportion to current conditions.
A simple, honest way to use one
- Pick one length that suits your timeframe (many swing traders look at 50, many intraday traders at 20 or 21) and do not keep changing it.
- Use it to classify the market: price above and sloping up, below and sloping down, or flat and tangled.
- Only take setups from your written plan that suit that classification.
- Set stops from invalidation points and ATR, not from the average itself.
- Record in your journal whether the average helped, then review after thirty trades.
A quick worked illustration
Suppose the last five closes on a pair are 1.1200, 1.1210, 1.1190, 1.1220 and 1.1230. The 5-period SMA is (1.1200 + 1.1210 + 1.1190 + 1.1220 + 1.1230) / 5 = 1.1210. If the next close is 1.1250 and 1.1200 drops out, the SMA becomes 1.1220, which rose 0.0010 on a 0.0020 gap between the new close and the dropped one. That arithmetic is all a moving average ever does; the meaning comes from how you choose to use it. These numbers are illustrative only.
Summary
A moving average is a smoothing tool with built-in delay. Use it to describe and organize what the market has done, accept the lag, and never expect it to call the next turn. Treated that way, it earns its place on the chart without needing any magic.
Educational content only; not trading advice.