What parameters of Simple Moving Average crossover to select

FX Helper
Posts: 1447
Joined: Mon Apr 01, 2013 3:55 am

What parameters of Simple Moving Average crossover to select

#1 Postby FX Helper » Mon Dec 28, 2015 5:54 am

or how the change of a single parameter by 20 can entirely change your balance from a positive value to a negative one

Description of the strategies based on indicators
Indicators are perhaps the most popular elements of the trading strategies. If you do not take into account the followers of Price Action and the people who trade solely on the news releases, then almost every trader uses one or another indicator. Among all indicators, the most popular are moving averages.

What do the moving averages display?
Moving averages give information on two very essential factors:
1. The direction of the trend
2. The value of the instrument

The direction of the trend is a parameter that indicates what type of trades should be opened by a trader, buy or sell. In an uptrend, it is recommended to open the buy orders only, while the downtrend is for the sell orders.

Value is the parameter that shows the real price of the asset.

Shares of a famous company cost $73. Employees do their wonderful work and produce quality products. Suddenly, the Board of Directors decides to fire the CEO. After the publication of the news, the price just in 1 day falls down to the level of $59 per share.

Does all this mean that the company began to work 1.23 times worse? Does it mean that suddenly overnight the products have become less popular? Of course no. This is the difference between the price and the value. The value of the shares was not changed for investors and was equal to $73. However, the price for traders did change and became equal to $59.

The moving average shows the value of the asset. It takes into consideration certain range and then calculates the average for the selected period. In our example, the stock price would certainly rise in the near future, as the macroeconomic performance of the company remained the same.

Why this strategy might be efficient
The #1 difficulty that appears when a trader uses strategies based on indicators is the selection of the indicators’ periods.

Some traders are very eager to open the trades when a moving average with a period of 20 crosses the one with a parameter of 100 on the daily charts, while others prefer to put to use the averages with periods of 15 and 50 on the 4-hour charts. To answer the question what period to use, assumptions and preferences make no sense, so we refer directly to the testing.

Description of the strategy
Buy signal: a short moving average crosses a long one upwards.
Buy-signal-English.png (512.51 KiB) Viewed 49352 times

Sell signal: a long moving average crosses a short one upside down.
Sell-signal-English.png (466.04 KiB) Viewed 49352 times

Testing results
We tested the above-described strategy:
• on EURUSD currency pair
• on the D1, H4, and H1 time frames
• on ECN Feed historical data
• on the period from July 27, 2003 until July 27, 2015
• on the parameters of 30, 40, 50, 60 for the short moving average
• on the parameters of 80, 90, 100, 110, 120, 130, 140, 150, 160 for the long moving average
The results of the testing are given in the table below:

D1 time frame
D1 table.png
D1 table.png (41.19 KiB) Viewed 49352 times

Because the most important statistical parameters are the profit and the maximum drawdown we did not include the other factors in the table for its simplicity. You can download a more detailed report using this link: [url]Report[/url]
Among all of the periods that were tested, the SMA (50) and SMA (160) duet gave the best result: 11,442 pips of profit in 12 years (953 pips a year on average). The worst outcome belongs to SMA (40) and SMA (80): 3,492 pips (291 pips a year on average).
In addition, the 50 and 160 SMAs had a quite good maximum drawdown of 2,372 pips. Maximum drawdown is even more important value than the profit because it shows the potential risks and allows calculating an appropriate loss for trading.

Other useful statistics:
1. Total trades: 18
2. Profit trades: 13 (72.22%)
3. Loss trades: 5 (27.78%)
4. Profit trades consistency: 5
5. Loss trades consistency: 2

Putting it all to use
1. Let’s say, you have opened a deposit of $1,000.
2. You pick the strategy based on two moving averages crossover
3. You set the values for the periods of indicators at 50 and 160 (because the results of testing)
4. You open the trades with a lot of 0.01 (each pip equals 10 cents)
5. In a year you will most likely have $1,000 + $95 = $1,095
6. Your potential risk were at $237. It means that in the worst trading day of this year you looked at the chart and saw that the market had moved against you at the distance of 237 pips.
7. 237 pips = $237 = 23.7% of your deposit. Experienced traders like Alexander Elder suggest to risk with no more that 6% in a month (you can download the table that allows calculating risks here: [url][/url]). If a trader sticks to this obligatory rule, in a year (s)he will lose no more that 52.41% of the initial deposit. With this said, we understand that the maximum lot for this case is: 0.01 * 52.41% / 23,7% = 0.02
To sum up, you can open the trades with a 0.02 lot, you will risk with a bit less than a half of your deposit, and most likely, you will earn $190 by the end of the year (with a $1,000 initial deposit).

You increased your deposit by 19% and had relatively small risks.

It might be extremely hard to open 1-2 trades in a year for those traders who came to the market for emotions, activity and not the money. Moreover, the “Loss trades consistency: 2” statistical value means that during 1-1.5 years in a row you will have a losing trade.

Looking forward to the comments of traders who cannot imagine themselves opening 1-2 trades a year and then doing nothing, we have tested the same SMAs on the H4 time frame.

H4 time frame
H4  table.png
H4 table.png (41.06 KiB) Viewed 49352 times

First of all, the values for SMAs of 50 and 160 that won on the D1 time frame would guarantee you one of the worst outcome if you implement them on the H4 time frame: just 2,515 pips in 12 years. This proves the point that everything should be tested in Forex and that assumptions are very dangerous. If you are a big fan of the H4 time frame then consider to use the periods of 30 and 100: they gave a profit of 7,221 pips in 12 years. Nevertheless, the total profit will be 11,442 / 7,221 = 1.58 times less.
The maximum drawdown in its turn will be a bit higher.

Other useful statistics:
1. Total trades: 218
2. Profit trades: 84 (38.53%)
3. Loss trades: 134 (61.47%)
4. Profit trades consistency: 5
5. Loss trades consistency: 8

Opening 1.5 trades in a month compared to 1.5 trades in a year might be easier for most of the traders.

Much smaller income and bigger risks compared to the D1 time frame; huge amount of loss trades (8 of them were in a row)

H1 time frame
H1 table.png
H1 table.png (41.58 KiB) Viewed 49352 times

Finally, backtesting results on probably the most common time frame among the day traders

The testing result on the H1 time frame were even worse than on H4. SMAs with periods 60 and 160 gave us 4,870 pips in 12 years (405 pips a year), which is twice worse compared to the backtesting on the D1 time frame. Twelve last lines of the table show that in the long run you can even lose with this strategy in case if you use wrong settings for indicators.
We especially want you to pay attention to the results of the SMA (30) and SMA (120) vs SMA (30) and SMA (140). In the first case, the trader has a loss equal to -1710 pips whereas the changing of the second average’s value by 20 provided us with a profit of 1,357.
A bit more than 3,000 pips is a huge difference in result and is one more proof that backtesting is everything when it comes to trading.

Other useful statistics:
1. Total trades: 501
2. Profit trades: 195 (38.92%)
3. Loss trades: 306 (61.08%)
4. Profit trades consistency: 5
5. Loss trades consistency: 9


The lowest income, possibility to lose a significant amount of deposit.

This test proved that a trader should be a scientist and not a player. Scientists take assumptions and afterwards test them in order to find if the assumption was correct or not.
The simple moving average crossover seems to be a very easy and profitable strategy.
The testing results claim that the best option is to trade on the D1 time frame and select the indicators’ periods of 50 and 160. In this case, you will be able to earn 11,442 pips and increase your deposit on average by 19% every single year.

Forex Tester Software, Inc. is not responsible of your potential profits or losses. We test the strategies on Forex Tester and provide you with the statistics received on historical data. The decision to either use our results or dismiss them is completely up to you.
2 SMAs.zip
(40.01 KiB) Downloaded 1143 times

Posts: 2
Joined: Tue Oct 24, 2017 8:05 am

Re: What parameters of Simple Moving Average crossover to select

#2 Postby avrahambr » Tue Oct 24, 2017 8:56 am


The best results were in daily timeframe, but there were only 18 trades in so many years. 18 trades don't have any statistical value. so how can you say that it's a science?
This backtesting only proves that you cannot rely only on indicators. You should find something which is more robust.


Return to “Strategies backtesting”

Who is online

Users browsing this forum: No registered users and 1 guest