Have we unraveled the secret to sustainable gains from 3X leverage tokens? (Patreon)

Building a bot or signal that can profit using leverage tokens on Binance or FTX can be a real struggle. 

These tokens are like the final boss playing in level god in one of these Dark Souls games, and without cheat codes! 

If you are not into computer games, it's ok, let's just agree that at least it's much more difficult than trading almost any spot token. They belong to a different league.

So should we stay 100% away from them? 

If you ask me I would say yes, stay away as they are very tricky. 

You can easily enjoy huge gains from them but as many things in crypto the problem is to retain those gains. 

Switching on a bot that trades 3X tokens in the beginning of a bull run can be really sweet. You can make 1, 2 or even 3 a day if you push it to the limit. 

It's not a challenge to make big gains, the challenge is how not to lose them soon after.

But something has changed since the last time we were in a bull market and we were trading them: we now have an arsenal of new tools and weapons that we can use to study what to do and what not to do in much more detail than we were ever able to do before.

And that's why I've taken this as a personal challenge and have defined two goals I want to achieve.

1) Goal 1: build a bot that doesn't use any external signal, only 3commas deal start conditions that trades 3x leverage tokens and that can profit great, get some small losses when the trend becomes bearish and it should automatically stop opening new deals for as long as we remain bearish. 

2) Goal 2: build a signal that can trade 3x tokens using it in combination with the 3commas deal start conditions and tools available in the platform.

The objective is to reach something around 1% a day with a big BUT: but without getting hit hard by a big market correction.

Goal 1: DEMETRA DCA bot, a fair attempt with pretty decent prospects

For the first objective we are going to use Bollinger Bands 5m. 

On 3X tokens all levels are a bit exaggerated, compared to spot because the exchange is dynamically adjusting the leverage of the token : fluctuating between 1 and 3 depending on the volatility of the market.

The guys that have been in the channel for many months know that we've been ONLY using DCA. We have never used only a BO and a stop loss for risk management. 

Well I decided to give it a go, even though DCA is still my favourite way of managing risk. But maybe for 3X tokens BO + SL could be a more adequate approach. Let's see.

I'm going to use and order of size 100, TP equals 2 times the SL. This is pretty standard technique which together with a good entry and confirming the trend is what many traders sell as the recipe for success.  

We are going to add some tokens to our DCA bot. You can copy into your tradingview these list: https://www.tradingview.com/watchlists/55971852

They contain the following tokens:

I didn't pick these tokens based on any criteria. You can use any but they need to be 3X. We will use binance ones as FTX doesn't offer TradingView technical ratings for deal start conditions neither RSI or Bollinger Bands on 3commas unfortunately.

And here's the recipe step by step:

• We need a short timeframe entry signal. For this I picked Bollinger Band 1std on the 5 min. This entry condition will fire very frequently. We want to catch as many entries as possible. The next steps will help us filter bad conditions to trade on larger timeframes.
• We need a trend confirmation. For this I'll go with TradingView BUY on the daily. We need a strong confirmation that we are on an uptrend.
• We need to make sure that we are not overextended, that we are not buying the top. For this I'm going to go with 2 RSIs together. RSI 1H < 75 and RSI 4H <85 Why? the RSI 1H is faster than the 4H and will hit higher numbers before the 4H does, we need to allow the 4H to get exhausted before stopping to trade. 

If your bot only traded with BBs, you will be happy as long as we are in a bull run, then on corrections it will enter constantly bad deals. That's why we need much more than just the short time frame deal start condition. We want to do anything that's possible to future proof our well earned profits.

Ok, we have everything we need to place deals. We need now to take care of risk management. We said we won't use DCA, we are going to go with stop loss instead.

We need to then make sure that the bot has max and active safety orders set to zero. The size of SO can be anything as it will get ignored.

You can have a look at the settings here:

https://3commas.io/bots/5530028/shared_show?secret=5d05337fc9 

How different is to run a bot like this compared to a typical DCA bot on spot tokens?

Very different. It's a bit like riding a crazy horse. Good, bad and amazing days. Depends on the market, but in theory with profits on a uptrend. 

When the bot takes a bad deal you will lose 7.5 % of the size of the deal. If your bot, like my version above, uses 10 max active deals then you will lose 0.75% of the total amount you have assigned to this bot. If the bot takes profit you'll make 1.5%. 

The strategy has higher chances to make a good deal compared to a bad deal, so in average and with a bit of patience you should end up in profit.

Backtest results

I need to make sure you understand I haven't run this setup. Not even on paper trading. 

I just started it today and will keep you in the loop on how Demetra is doing. We will need 2 to 4 weeks to confirm if the backtests are fairly accurate and then decide to invest real money on it.

3X leverage tokens are VERY risky so do not jump blindly on this bot. You cannot paper trade it as most of the 3x tokens are not available on paper trading.

I encourage you to use Gavin's backtest script and play with the settings and explore the results.

Below are the results based on 72 days of backtesting:

The average daily profit when a applying the strategy to all tokens selected is 0.7% a day. That's about the same we are getting from BEBEK currently. The worst performing token made a total of 14% in profits and the best one 110% which is about 1.5% a day.

You can also get an idea how how is the ride on the best performing pair which was XRPUP by looking at the image below. It's not smooth at all and if you check the results below you can clearly see many periods of drawdown. Profit goes up and down in the short term, and up in the larger picture. In the table above you can see some tokens experience a very small drawdown of 6% and some others up to 30%. High drawdown and profit levels seem to be correlated.

Goal 2: BIANCA signal, no TP no SL, few large wins and rare losses

Bianca is a signal that uses the latest improvements on IFTRSI and combines the actual signal with the available 3Commas DCA deal start conditions on Binance.com 

Once this signal enters a deal it needs to decide when to close it using the SELL signal. 

It does NOT use any take profit level therefore it's possible to get a single deal to close at even more than 80%. 

Interestingly backtests don't show any drawdown below 7% during the past 2 months as opposed to what Demetra showed us with even a token with a 30% drawdown which is really great. 

This signal definitely seems more refined than Demetra and it shows an average daily profit of 0.87% which in paper doesn't sound like a lot more than 0.7% from Demetra but after a few months it can make a big BIG difference. 

Having low drawdown is also very attractive. 

I'm using the version of IFTRSI released today which includes a volatility range filter, which allows you to filter signals between a normalized area of volatility. 

For example you can say "only take deals when volatility is between 10% and 50%". 

This is really great because many tokens were not compatible with IFTRSI due to having a volatility not compatible with the strategy so now it's  possible to filter some of those scenarios. 

As you can see in the image above XRPUP, the best performing bot only took 3 trades in 70 days. 

One of them lasted for almost 5 days. 

The last deal made 85% profit ending up with a cumulative 107%. 

The BUY signal is very active all day but it takes many days to send a SELL signal. 

One annoying thing about this setup is that ir requires you to set take profit at 100% to actually disable it so that is the strategy deciding when to close the deal but unfortunately 3commas keeps sending errors that the take profit is too high according to the exchange. 

I wish there was an option to officially disable the TP instead so I don't get an error every 5 minutes on telegram! Alternatively you can set the TP to only 50% so that you don't get any error but you'll miss out those sexy 85% deals whenever they actually happen :) 

Here are some basic stats of Bianca for the past 70 days:

Even though I'm color coding in red the tokens that take long to close, in the case of Bianca, taking long is part of the design of the bot and not necessarily something bad.

I will continue testing and keeping an eye on these two strategies.  

If after 2 to 4 weeks I see good results I will definitely will be offering the Bianca signal. 

You can backtest Bianca :

• composite 10 max active deals
• no simultaneous deals same pairs
• BO 100 USDT
• no SOs
• TP 200%, no TTP (signal determines when to TP)
• deal start condition 1 : Manually/API (Bot won't open new trades automatically), 
• deal start condition 2: Trading View, 1 day, Buy,
• deal start condition 3: RSI-7, 4 hours, 80
• deal start condition 4: RSI-7, 1 hour, 80
• IFTRSI Volatility filter enabled 15min between 0 and 70, use TV ratings levels using 15min: -0.96 , -0.96 , 0.99, 0.999, and the rest all defaults
• Same tokens as Demetra

Wish her both good luck :)