SMA and EMA Moving Averages: The Most Popular Tool and Its Limitations

A moving average is the average price over the last N periods, shifted with every new candle. SMA is a simple average; EMA gives more weight to the most recent prices. Both smooth the chart and both lag by definition.

What it is

Moving averages are the oldest and most widely used technical analysis tool. Popular periods are 20, 50 and 200, and the media follow the 200-day average as the “line between bull and bear markets”.

A moving average doesn't predict price. It shows where price has been on average. When price is above a rising average, the market is in an uptrend, but that's just another way of writing down what's already visible on the chart.

How people use it

trend direction: price above or below the average
crossovers of two averages (e.g. 50 and 200) as signals
the average as dynamic support or resistance
the daily 200 average as a level watched by many participants

What the research and evidence say

Moving-average crossover rules are the most thoroughly researched part of technical analysis. Brock, Lakonishok and LeBaron (1992) showed their value on the Dow Jones from 1897 to 1986, but later work (Sullivan, Timmermann and White, 1999; Bajgrowicz and Scaillet, 2012) indicates that once the number of rules tested and costs are accounted for, the edge disappears on more recent data. Averages work best as a trend filter over a horizon of weeks and months, not as an intraday trigger.

Weaknesses

the lag grows with the length of the average
in consolidation price crosses the average many times
an average that “works” on history is often the result of choosing the period after the fact
it says nothing about volume or participant activity

What order flow shows

A moving average tells you where the average price was. VWAP – the volume-weighted average since the start of the session – tells you where the day's participants traded on average, which is why it's a reference point for many institutions. Order flow adds whether real demand appears at the average: whether buyers defend the level with aggressive orders, or price passes through it without resistance.

Indicators vs order flow

Every indicator and every pattern in this section is built from the same data: a candle's open, high, low, close and sometimes volume. They're derivatives of price, showing the effect rather than the cause. Order flow looks at what price is made of: who was buying and selling aggressively (delta), where limit orders were waiting (the order book) and at which prices the market accepted trade (the volume profile).

On CME futures this data is real and comes from one central market. Spot forex doesn't have it: platforms show tick volume, the number of price changes at a single provider.

To be fair: order flow isn't the Holy Grail. It takes learning, it can be noisy, big players hide some of their orders, and an edge only comes from a process and risk control. But it gives you information that simply isn't on a candlestick chart.

How to use it sensibly

use averages to describe context and the trend on a higher timeframe
don't pick a period based on which one “held” best on history
intraday, add VWAP and levels from the volume profile

Glossary: Moving average (SMA and EMA)
Glossary: VWAP
Atlas of biases: Overfitting
Atlas of biases: Herd behaviour
Order flow: VWAP and deviation bands: what they tell you and what they don't
Also under the microscope: RSI
Also under the microscope: MACD
Also under the microscope: The stochastic oscillator and Stochastic RSI

Frequently asked questions

What's the difference between SMA and EMA?

SMA is a simple average over N periods. EMA gives more weight to the most recent data, so it reacts faster, but it also gives false signals more often.

Why is the 200 average so important?

Because many market participants and the media watch it, so some of the reactions there come from the attention itself, not from any property of the average.

Which moving average is best?

There isn't a single best one. Every period is a trade-off between lag and noise. The worst choice is a period fitted to history.

Educational content, not investment advice. No signals, entries, stops or position sizes.