A moving average is the average of a set number of recent prices, recalculated as each new period arrives. It smooths the day-to-day noise so that the direction of the price is easier to see. A simple moving average gives every price in the window the same weight. An exponential moving average gives recent prices more weight, so it reacts faster to a change. Both are among the oldest and most common tools in chart reading.
This guide explains how each average is calculated, shows the arithmetic with invented prices, and sets out how traders use averages for trend and support. It also explains why the most watched crossovers arrive late and sometimes give false signals. An average describes the past. It cannot tell you what the price will do next.
What does a moving average tell you?
A moving average turns recent prices into one line. When the price sits above a rising average, recent prices have been higher than older ones. When it sits below a falling average, the reverse holds. The slope matters as much as the position, because a flat average suggests the price has been moving sideways.
Traders use several lengths. Short averages follow the price closely, while long averages move slowly and show the broader direction. The lengths most often quoted, such as 20, 50 and 200 periods, are conventions that developed through common use. They are not special properties of the market, so choose a length that matches how long you plan to hold, and test it on past data rather than copying a number from a chart you admire.
Averages also show where the price sits relative to its own recent history. A price well above a long average is stretched, and a price well below one is depressed, although neither condition says how long it will last. The distance between the price and the average can be measured, and some traders compare it with the stock's normal range to see whether a move is unusually large.
How are simple and exponential averages calculated?
A simple moving average adds the closing prices over the window and divides by the number of periods. A five-period average of 10, 11, 12, 11 and 13 is 57 divided by 5, which is 11.4. An exponential average uses a smoothing factor based on the window length.
For a five-period average the factor is 2 divided by 6, which is one third, so each new average equals one third of the latest close plus two thirds of the previous average.
The table below uses invented closing prices. The exponential average starts from the simple average on day five. The figures are illustrative.
| Day | Close | Simple average (5) | Exponential average (5) |
|---|---|---|---|
| 5 | 13 | 11.40 | 11.40 (starting value) |
| 6 | 14 | 12.20 | 12.27 |
| 7 | 9 | 11.80 | 11.18 |
On day six the exponential average is higher than the simple average, because the recent rise carries more weight. On day seven the close falls to 9, and the exponential average drops further than the simple average. The difference is small in this example, but on a real chart it can decide whether a signal appears a day earlier or later. Neither average is more accurate. They respond to change at different speeds, and the choice between them should follow from the rule you are testing.
How do traders use averages for trend and support?
Many traders treat a price that holds above a rising average as evidence of an uptrend, and a price that holds below a falling average as evidence of a downtrend. A moving average can also act as a level of support or resistance.
Price often pulls back to an average in a trend and then resumes, although the average does not stop the price every time. The more often a level has worked in the past, the more attention it deserves.
Averages are also used as a filter. A trader who only takes long trades when the price is above a long average avoids many falling markets, but also accepts late entries after the trend has already moved. That trade-off is the central issue with all trend tools. The average tells you where the price has been, and every entry based on it is a judgement about whether the move will continue. A filter that removes many losing trades may also remove the best ones, and only a test on past data can show which effect dominates.
Why do crossovers lag and whip?
A crossover happens when a short average crosses a long average. A short average crossing above a long one is often read as bullish, and a cross below as bearish. The best-known example is the 50-period average crossing the 200-period average, which traders call a golden cross when it rises and a death cross when it falls.
Because both averages are built from past prices, the cross appears only after the move has already started.
In a sideways market, crossovers are frequent and mostly misleading. The short average crosses back and forth as the price oscillates, producing a series of small losses. This is called whipsaw. A trader who uses crossovers should expect many false signals in ranges, and should decide in advance whether to wait for a close beyond the average, a retest or a minimum separation between the lines before acting.
How do you combine averages with other tools?
Moving averages work best as one part of a set of tools, such as the trend, levels, volume and a momentum reading. An average tells you the direction and rough position of the price, while the other tools help with timing and risk.
A signal supported by several tools is usually easier to explain than one that depends on a single average.
A common combination uses a long average to define the trend, a shorter average to find pullbacks and a horizontal level to mark where a pullback should stop. A trader can then wait for the price to touch the short average inside the uptrend, and set the stop below the level. The rule is clear enough to test, and each part can be judged on its own.
Be careful about adding too many averages. Several lines that cross each other constantly produce a chart full of signals, most of which are noise. Two averages chosen in advance, tested on past data and kept fixed, are usually easier to trust than a crowded chart in which every crossover looks like a decision.
Common mistakes
Treating the lag as a flaw to fix with a shorter length. A shorter average reacts sooner but also whipsaws more. Test the trade-off rather than assuming a shorter length is better.
Trusting a crossover in a range. Flat markets produce crossovers constantly. Check the slope and the recent range before you act.
Changing the length after seeing the result. A length chosen because it fitted one past move will not be a reliable rule. Set the length before testing, then keep it.
Checklist
- Decide the window length from your holding period, and write it down.
- Check whether the average is rising, falling or flat before you read the price against it.
- Note whether the price has respected the average as support or resistance in the past.
- For a crossover, wait for a close beyond the lines and check the wider range.
- Compare the simple and exponential versions if the difference matters for your rule.
- Test the rule on past charts, including the periods when it failed.
- Set an exit that does not depend on the average alone.
Key takeaways
A simple moving average gives equal weight to each price in its window, and an exponential average gives more weight to recent prices. Averages summarise trend and can act as reference levels, but they describe the past. Crossovers are late and whipsaw in ranges, so they should be tested and combined with other evidence before they are trusted.
This guide is for general information, not financial advice. See the disclaimer.