Home Creators Posts Import Register Favorites Logout
haven't archived this post yet. have a subscription? use the importer!

Downloads

  • awesomemacd-files.7z
  • AwesomeMacd.7z

Missing 2 files, as this post has not yet been imported.

Content

Intro

The awesome macd strategy is available in Freqtrades github strategies repository.

I tested this code to see if the original developer has found a way to earn profits through automatic bot trading with this strategy.

Downloading the strategy

You can download this strategy from the Github repo of the Freqtrade bot at the following location: freqtrade-strategies/user_data/strategies/berlinguyinca at main · freqtrade/freqtrade-strategies · GitHub

And if you want to download this file directly to your computer, then you can use the following command directly from within your strategies directory:

wget https://github.com/freqtrade/freqtrade-strategies/blob/main/user_data/strategies/berlinguyinca/AwesomeMacd.py

Lets analyse this file and see what this strategy does before we are going to backtest it.

The strategy

The author of this strategy code is Gert Wohlgemuth, but he was inspired by another author named Steven the Wissen. Steven gets an honorable mention as well here.

Let’s begin with the default settings of each and every Freqtrade based strategy here: the take profit and stoploss settings:

  • The only ROI point where the bot takes profit is at 10%
  • And the stoploss point is 25%
  • The strategy was originally designed to work on the 1-hour timeframe, but we will see later in this article if that is actually the best timeframe to work on.

Indicators

This strategy will be using the ADX, Awesome oscillator and MACD indicators as you can see below in the code:

Buy / Sell signals

Then we come to the actual strategy part of this file.

This is the part where buy signals are generated and this the part where the bot sells its position.

In the buy section we can see that the bot will enter a long position on an asset were the Macd indicator is above the 0 line, the awesome oscillator is above the 0 line.

However, the shifted value of the awesome oscillator should be below 0. In other words, Yesterday’s awesome oscillator should be below 0 and todays AO should be above 0 to create a signal for today.

This means that if the AO signals when MACD is below 0, then there will be no buy signal. Only when MACD is above 0 while the (AO>0) & (AO.shift<0) there will be a signal.

Note that I explained these buy and sell signals incorrectly in my video. I would like to thank TheDjarEl for his correction!

The exit signal is the reverse of the entry signal. So, in this case the MACD and AO of today should be below 0, but yesterday’s AO should be above 0 for this signal to be valid.

Take not that the ADX indicator is never used in determining buy or sell signals so maybe this is a remnant of an earlier version of this strategy.

The strategy on a chart

You can see the used indocators of this strategy on this chart below, together with the buy and sell moments.

According to the strategy, the bot should buy when the MACD is above 0 AND when the current AO is above, and the previous AO is below 0.

That are roughtly these points. And remember here that the bot is programmed to take profit when there is a 10% profit.

You can clearly see that this is a trend trading strategy where the trend is determined by both these indicators.

Backtesting

Now that we know this, lets backtest this code and see if this strategy actually has the best performance at its configured 1-hour timeframe or if another timeframe has better results and what those results are.

After backtesting, the 1-day timeframe is the most responsive to this strategy.

Maybe because on this timeframe the number of trades is kept at a low rate, and it gets the most out of its trend following nature.

Let’s see if we can optimize the ROI and stoploss settings of this strategy.

Hyperopt session results

Indeed, after some changes to the ROI and stoploss settings MACD became more awesome. However, this comes with a drawback because the win rate dropped with 20% to only 55%.

Nonetheless a good win rate because statistically every second trade is a winner with also high profits.

The second most remarkable thing, which I like here is that the drawdown went from 21% to 16%.

Now that is a good thing because we want to keep our gains en do not want to give these back to the market.

Strategy league

If you look at the strategy league, you can see that the MACD indicator is very popular.

Three of the top 10 strats have this indicator as some sort of confirmation to enter or exit trades.

The Awesome macd entered this league on the 7th position.

I think that it’s limited drawdown, which also results in a medium risk of ruin, adds to this high position.

It’s ability to gain much fortune is not that high, considering these higher scoring strategies, but a potential profit of 741% is not that you can find on any bank. But consider the risks ofcourse!

And we also know at this point that this strategy has a good probability to make you money, and keep it as well, even If you use the unoptimized version with its 75% win rate.

Then you could consider using this code for your trading bot.

But as always Do your own research first before making any actual decisions about this.

Comments

Ding Dong Bell

for some reason i can't open the python file after the extraction. its saying "I dont have permission to open the file" ..is anyone else facing this issue ?

Koprix

same for me

Koprix

can you please re-add the file? looks interesting this strategy :) many tks in advance

Koprix

try to use rar and works, with 7 zip asked for password

Dragos Vintu

the file on the platform says it has no permission to run = i'm on a macos

Dragos Vintu

after I unzip it it says that I do not have permission to view it. I wanted to open it first to see what is going on under the hood.

Dragos Vintu

Right now running paper trading for the last 2 weeks with flawlessVictory 15m, SMAOffsetProtectOptV1 30m, Obelisk_Ichimoku_Slow_v1_3 1h, Obelisk_Ichimoku_ZEMA_v1 5m. The true winner by far is Obelisk_Ichimoku_ZEMA_v1 5m. Will hyperotpimize it using the data from the last 360 days and will update you on this. Waiting for a new toy that will have the power to process all that data. My M2Pro 32 GB of ram was not enough. But looking forward to having Macd up and running to see it in paralel.

Dragos Vintu

The MACD is up an running. Out of the strategies I am paper trading right now Obelisk_Ichimoku_ZEMA_v1 5m is leading with a 4.5% / week but MacD is at 2%/day. Will hyperoptimize today the Zema with a Rogue Strix with 64 GB of ram(because the ram was what was lacking) and let them both run for one week paper trading. will come back with updates