The rule
Two exponential moving averages, commonly 50 and 200. The fast one crossing above the slow one is a long condition, below is a short condition. Everything else in this lesson is about not taking every one of them, because taking every one of them is how the method got its reputation.
The filter that makes it usable
Only take crosses in the direction of the higher timeframe. On a one-hour chart, only long when the four-hour is above its own 200. This one filter removes most of the losing signals, because most of them are crosses that happen while a bigger chart is going sideways and the two averages tangle around each other repeatedly.
Entry, stop and exit
Enter on the close of the candle that crosses, or on the first pullback to the fast average, which gives a better price at the cost of some missed moves. The stop goes below the recent swing low, not below the average — an average is a calculation, not a level where anyone has orders. The exit is the opposite cross, or a trailing stop under each new swing.
Why it lags, and why that is the deal
An average of the last fifty candles cannot turn until the price has already turned, so every crossover arrives after the move began. That is not a flaw to be tuned away; it is what an average is. Shortening the periods to get earlier signals gets you more signals and more false ones, and every trader who has tried it has arrived at the same place. Accept a late entry and a real trend, or do not use averages.
The losses to expect
In a sideways market the two lines cross back and forth and each cross is a small loss. A run of six or eight such losses during a quiet fortnight is completely normal for this method, and the traders who fail with it fail there — they abandon it in the range, right before the trend that pays for the whole month. If you cannot sit through that, this is not your strategy, and knowing that in advance is worth more than the strategy.
In short
| Common pair | 50 EMA and 200 EMA |
|---|---|
| Long | Fast crosses above slow |
| Filter | Higher timeframe direction |
| Stop | Recent swing, not the average line |
| Weakness | Repeated small losses in a range |
Common questions
Which periods are best?
The standard ones, because they are what most people are watching. Searching for the pair that would have performed best on the last year of one pair is fitting, and the winner of that search almost never repeats.
Is a golden cross a buy signal?
It is a trend condition, widely watched, and it arrives well after the trend started. As one input alongside structure it is fine; as a standalone instruction it is the version that loses in ranges.
SMA or EMA?
EMA reacts sooner, SMA is smoother. The difference is much smaller than the difference made by the higher-timeframe filter, so pick one and stop reopening the question.
Trading forex and CFDs on margin carries a high level of risk. Most retail accounts lose money. Nothing on this page is financial advice or a recommendation to trade.