What is important to understand about the EMA is that it does not work all the time. Asset prices trend only 30% of the time. Your risk management will play an important function in the success of an EMA crossover trading strategy. The goal, when using a trading strategy such as the EMA, is to generate robust gains when the price of bitcoin begins to trend.
If you are like me, you have experienced frustration trading bitcoin using a trend following indicator such as the EMA. I am sure that there have been numerous situations where you enter a trade when a trend signal is generated only to find that the price reverses and quickly moves against you. One of the reasons this occurs is that moving averages, in general, are lagging indicators.
Unfortunately, a trend is not a straight line. A trend can take an upward or downward path that can have several twists and turns.
The Chart of Bitcoin
As you can see from the chart of bitcoin (above) over the past 9-months there have been times when bitcoin experienced a trend and times when prices consolidate and move sideways. In fact, nearly half the 9-months in the chart above of bitcoin show that prices moved sideways (green circle).
What is the Exponential Moving Average?
The exponential moving average is a moving average that places more emphasis on recent prices. This is accomplished by weighting the moving average, so it responds more quickly to newer information. The formula that is used to calculate an EMA just involves using a multiplier to alter a simple moving average.
Moving Average Calculation
You can calculate as simple moving average by generating the average of a certain number of time periods (for example days). When the next day is added the first day in the calculation is dropped. A moving average provides a visual representation of the longer-term trend of an instrument. Faster moving averages, with shorter look-back periods, appear to be choppier relative to a slower moving average, with longer look-back periods. Because a moving average is a backward-looking indicator, it is described as lagging.
The calculation for the exponential moving average is performed by your charting software, but its helpful to know what is happening behind the scenes. After the simple moving average is calculated the weighting multiplier is calculated. Then, using the current price, in conjunction with the multiplier and the previous period’s EMA value, the EMA for each day, is calculated.
Generating a Trading Signal
The benefits of using an EMA compared to a simple moving average is that you are likely to receive a signal that is more in tune with current price action. An EMA will turn before a simple moving average (SMA), which can provide you with a slight edge. When you create trading signal using the EMA you need to understand that regardless of how much you weigh current price movements, moving averages are lagging indicators.
Your Moving Averages Will Define the Number of Trades Produced
The best strategy is the one that works best for you. If you want to day trade, you might consider using intra-day moving averages, along with short term daily average periods. If you are looking for a long-term trend, you should increase the number of days or even weeks used to calculate the EMA.
The chart of bitcoin has an overlay of the 5-30-minute EMA and the 20-30-minute EMA. In using a 30-minute time horizon, over approximately a 6-day period, 9-crossover down signals were produced along with 7-crossover buy signals. During that period a short term down trend in daily prices describes the current market environment.
- During this period 8 of the 9 sell signals that were created likely made money while one of them (the red arrow circled in green), was likely unsuccessful.
- The reason for the success is that the intra-day EMA is good and predicting short term trends when the larger trend is moving in the same direction.
The sell signals work in a down trend. You can see that the buy-signals, which are the green arrows in the chart, could work, but the amount of gain you would experience on each trade is small.
If you are looking to capture a medium-term trend, you can switch your moving averages to longer periods.
- The 20-day EMA and the 50-day EMA, represent a month and nearly a quarter.
- During the last 18-months, there were 7-sell signals and 5-buy signals.
The trend in bitcoin prices following the surge in value in December of 2017, was down. In early 2019 prices have appeared to have bottomed. The sell signals were more profitable during the down trend. It looks like 6 of the 7-signal would have likely been profitable based on your risk management criteria.
The long term EMA represented by the 50-day EMA, which is nearly a quarter, and the 200-day EMA, which is approximately a year, does not generate many crossover-buy or crossover-sell signal.
- The sell signals were produced in April and May of 2018.
- The buy signals were generated in May of 2018 and April of 2019.
Defining a Trend
Its up to you to define a trend, as the long-term trend will be your friend. You want to trade in the direction of the trend if you are using the EMA crossover trading strategy.
Some technical analysts view a trend as a period when the price of an asset is below a specific moving average.
The EMA crossover strategy is geared towards finding the middle of the trend. Since it uses backward looking data you will receive a signal only after something has already happened. A trend can be defined as a price pattern where the prices move in the same direction and this perpetuates over time.
When prices are consolidating or moving sideways there is no visible trend. There is a theory that says that prices trend approximately 30% of the time and trade sideways 70% of the time.
Risk Management with EMA Crossover Strategies
Since you are trying to capture profits during periods when the market trends, the risk management that you employ will play a large role in generating gains using a moving average crossover strategy.
Here are some assumptions.
- If you will only capture a trend 30% of the time, you need to make more than $7 for every $3 you risk.
- If you make 10-trades and win 3 earning $7 each time ($21), it will be offset by 7-trades that lose $3 per trade ($21).
The risk management profile, that you can consider using when you trade a long term EMA trading strategy, should hold assumptions. It could be based on the idea that you will lose more than you win but will generate more profits on winning trades than losses on losing trades. This is a successful formula.
Some trend following risk management creates a trading strategy where you are always holding a position in the market.
- For example, assume you generate a transaction as a sell signal at the red arrow in April of 2018.
- You would exit and reverse on the buy signal in early May 2018 and then reverse again on the sell signal in late May of 2018.
- The first two trades were losers, as Bitcoin traded sideways.
- The next trend was substantial, allowing you to generate gains of 40% unleveraged as prices moved from 8,200 to 4,900 and then reversed.
- You would still be in a long position in Bitcoin as of mid-July 2019 even as prices retreat, and you give up gains.
You could also use a different methodology of risk management. If you believe that the markets will only trend 30% of the time you must win $7 for every $3 that you lose, you can generate a multiple of this number. This is the same as $2.13 for every $1 you risk ($7/$3). If you set your risk parameters which allow you to lose 10% on a position, you should target your reward at 22%.
Another very popular risk management technique that can be used when you are using exponential moving averages is a trailing stop loss.
The chart above provides an example of how you might consider using a trailing stop loss.
- You enter a sell signal at the red arrow as the 20-EMA crosses below the 50-EMA near 1,310.
- When the price of bitcoin hits your take profit range (22% of 1,310 = 1,022) you can move you stop loss down allowing you to continue to track the trend.
- You should only consider moving the stop loss when the price moves your way. For example, you would not increase the stop loss to a higher price if the closing price of bitcoin on that day moved higher.
Each day you can adjust your stop loss. When the price of bitcoin closed at 7,000 in February of 2018, the new stop loss would be 7,700.
Trading with the Trend
Since you want to trade with the trend you want to define what kind of trend you are looking to capture. If you are looking for a short-term trend, you want to evaluate the daily trend and consider trading intra-day signals.
If you want a medium-term trend, you might evaluate the quarterly and annual trend and generate signals with a medium term moving average crossover-strategy such as the 20-day moving average and the 50-day moving average.
Automating Your Trading
While it is possible to completely automate your trading strategy, you can also use discretion when trading the EMA crossover strategy on bitcoin.
An automated strategy would provide specific criteria to an expert advisor (or similar), allowing you to execute trades that incorporate your strategy. The best way to proceed if this is an attractive option is to back-test an EMA using a demonstration account. A demo account allows you to use demonstration money to see if your strategy works in real-time without risking real capital.
What you will likely find is that these strategies work overtime, if you are willing to stick with specific risk management criteria. You might find as you experience these trades in real-time that you are buying at a short-term high only to have the market reverse, providing short-term unrealized losses. The goal is to stick to a specific criterion and ignore losses until you are completely stopped out.
You can also trade an EMA crossover strategy using discretion. You might consider using other criteria such as momentum, to help you define a trend. You want to trade in the direction of a longer-term trend, which will help improve the success of the short-term EMA crossover strategy.
The exponential moving average is an efficient trend indicator. It provides users with a customized way to designate that a trend has started. An EMA is weighted and is more sensitive to current price action compared to a simple moving average. Short term crossover signals tend to work whe a longer-term trend is occurring in the same direction.
Risk management plays a key role in determining if an EMA crossover trading strategy will work. Since trends are theorized to only occur approximately 30% of the time, you need to “make hay when the sun is shining”.
For your risk management to work correctly you need to offset periods when you are receiving EMA crossover signals that are dead ends, with robust gains when a trend is in place. There are several different risk management strategies that you can use to make an EMA crossover strategy work successfully.
You can automate an EMA crossover trading strategy or use discretion. If you plan to automate your strategy you should consider back-testing different EMA time frames and then use a demonstration account to test to see if it works using your risk management criteria.