What is the difference between an EMA and an SMA?

A simple moving average weights every price in its lookback equally. An exponential moving average weights recent prices more heavily, so it responds faster to new information, though it remains a calculation performed on past prices.

Do EMA crossover strategies work?

They underperform structurally. A crossover requires enough movement to pull two averages past each other, so the signal necessarily arrives late, and in ranging conditions it produces repeated late signals in both directions.

Which EMA period is best?

No period is inherently correct. Common lengths matter partly because many participants watch them, which is a real but weak effect. Consistency with a settled period is more useful than optimising on historical data.

How should moving averages be used?

As directional bias and as a filter on trades you are already considering, rather than as entry signals. The slope and price position relative to a longer EMA carry more information than crossings.

EMA Trends: Context, Not Signal | SWTM Futures Course

Exponential moving averages are among the first tools most traders meet and among the first they misuse. The tool is genuinely useful. Almost everything commonly taught about how to trade it is not.

What an EMA actually computes

A simple moving average treats every price in its lookback equally. An exponential moving average weights recent prices more heavily, with the weighting decaying as you go back.

The practical effect is that an EMA responds faster to new information than an SMA of the same length. It is still, by construction, a function of prices that have already happened.

A moving average cannot lead price. It is arithmetic performed on price. Every property it has follows from that.

That is not a criticism — it is the definition. Problems only begin when a lagging summary gets used as though it were forward-looking.

Where crossovers fail

The most commonly taught EMA method is the crossover: fast EMA crosses above slow EMA, buy; crosses below, sell.

The reason this underperforms is structural rather than bad luck. A crossover requires enough directional movement to pull two averages past each other, which means the signal necessarily arrives after a portion of the move has already happened. In a strong trend, that is survivable. In a range — which is most of the time — price oscillates around both averages, generating repeated crossovers in both directions, each one late.

The result is a method that performs well in exactly the conditions where you would have profited anyway, and bleeds in the conditions that dominate the sample. Backtests look acceptable because trending periods carry them. Live trading feels different because the ranges are where you actually sit.

Using EMAs as context

The defensible use is directional bias rather than entry timing.

Where price sits relative to a longer EMA tells you something about the prevailing condition: consistently above and rising suggests buyers have had control over that window; repeatedly crossing back and forth suggests no one has. The slope carries more information than the crossing.

Used this way, an EMA is a filter on the trades you are already looking for, not the reason for taking them. It answers "does this trade agree with the recent condition" — a question worth asking, and a much smaller claim than "is this an entry".

Why period choice is mostly arbitrary

The 9, 21, 50 and 200 EMAs are common because they are common. There is no market-structural reason those numbers are correct, and their significance is partly self-fulfilling — they matter somewhat because enough people watch them.

That is a real effect and a weak foundation. It also means optimising the period on historical data is usually fitting noise: a length that performed best over a past sample carries little information about the next one.

Consistency beats optimisation here. A settled, unchanging period you understand the behaviour of is more useful than a tuned one you will change again after the next drawdown.

How this is taught in the room

Above is the honest version: what an EMA computes, why crossovers underperform, and why period optimisation is mostly self-deception. What members get is how we actually use them — which lengths we keep on the chart, what an EMA is allowed to veto versus what it can never trigger, and how the bias it provides is combined with structure and level work rather than acting alone.

This module exists partly to stop newer traders relying on moving averages the way they usually first learn to.

Platform

Charted on TradingView.

← RTH Fibonacci Order flow →

EMA Trends