Systematic Trading on Cryptos : Here’s How We Solved The Initial Problems
Hi, guys. Hi, everybody! Andrea Unger here and I'm here with Francesco. Hi! …To talk about cryptocurrencies. Yes. So, you might remember I was not a huge fan of trading on cryptos, but I explained why! It was a technology gap and data missing, historical data missing problem, which are both now somehow solved. The data are at our disposal - many players in the software world provided the historical data - and most importantly, today, we can also trade directly through MultiCharts on one exchange, which is Binance.
So, just to recap: as soon as the data were available, I did develop something to trade Bitcoin futures, because as said, there was not yet the possibility to trade directly on crypto. So, I chose the future on CME and I developed a very simple trading system, sort of an opening range breakout, which was taking the range, calculating 70% of it, adding to the high and subtracting to the low, and those were the entry points. When these were exceeded, the system was entering a position and it was then riding the trend. There was a stop loss. I introduced then a trailing stop - which I'm not a fan of, you know - but in this case, it was necessary because of a number of reasons.
There was also a take profit, at a very wide distance compared to the entry points, of course. And this system was successful. I've been trading it for a while, but I was not satisfied because of 2 main reasons.
The first reason was - and also Francesco is aware of this - that the margin requirements for the futures on Bitcoin are very expensive. And as soon as you start to trade with 2, 3, or even more contracts, you are eating out a huge part of your capital dedicated to trading. But ok, this was one minor problem, somehow.
The major problem are the bid-ask spreads, which are there all day long. You often find 50 points, which is very expensive, and this was creating problems even to the system, understanding if a level was triggered or not. So, it was creating a mess, and it turned out to be very expensive.
Costs were very high compared to what we could estimate. Systems were still profitable but not really satisfactory at all as I supposed. As soon as MultiCharts opened its doors to the connection to Binance, that was a new world. Francesco is a big supporter of this, he has always been pulling me into this world because of his curiosity, and I am grateful to him because I am really enjoying what we are doing today on cryptos. One important note before leaving Francesco to his introduction is that we have not chosen cryptos because of the huge returns we all are aware of, because those returns are sort of investments somebody could make.
You sit on the crypto and then you wait. The problem is: you don't know when to stop waiting. You don't know when to sit in. At least, I don't know! But I think Francesco is in the same position and I believe really most of us all have a problem in managing the positions. So, our purpose was to find markets that have enough volatility to be traded in a clean way with our basic models.
Sort of getting back to the ‘80s on commodities, when everything was working there. And this is what we try to do, and we found out very interesting approaches, following our development method, to trade on cryptos. But, it was not always a pleasant ride because obviously we also found some problems, but I believe Francesco, who is a biggest fan of this, and the biggest sponsor, could explain better what problems he encountered and what his situation is today after, I think, 6 months of real-life trade. So please tell us about your experience.
Yeah, developing trading systems on cryptocurrencies is in a certain sense, quite easy because a very simple logic tends to perform well, but I made some mistakes at the beginning. First of all, I underestimated the importance of a solid infrastructure. I'm referring to a dedicated server instead of a VPS. So, it's so important to have a very efficient PC because when the market is in a bull trend and for example, it breaks the high, a lot of orders arrive at the exchange and MultiCharts can have some problems connecting to Binance Some processes can still wait for a response from the exchange and the RAM, the use of the RAM becomes bigger and bigger, and so… When the data feed goes crazy, also the platform is sort of suffering because of this. …So, the server can crash.
And this is an issue because first of all you have to realize that! Maybe you are doing something else and… Doing something else? Aren’t you sitting 24 hours a day in front of the monitor? No, fortunately, this is a big advantage of systematic trading! And the other aspect is that when this happens, you just close all the positions and maybe you can lose a very good day of trading because, when the market spikes, sometimes it happens that at the end of the day you can make a huge profit. And, in my experience, I lost 2 or 3 very beautiful days of trading because of the infrastructure. So, it’s Murphy's Law applied to trading! When something goes wrong, it creates the maximum damage possible in your account. So, my suggestion is: you have to spend money to have an efficient machine.
That’s my first suggestion. By the way, I remember when Francesco was suffering these problems and he was not aware yet of the real problem. He was claiming against Binance and against MultiCharts, claiming that they were not efficient, that they had problems.
Then, he realized that he was the culprit, that he was the one who didn't want to invest enough money in technology. And once he made the next step and he got a dedicated server, all his problems suddenly disappeared. I confirm! Still the same software versions he had before so actually, Francesco was guilty of his losses. I have to admit it. And the other issue, the other mistake I made, was to use a monetary stop loss.
You know, when you are trading an underlying that can run, can make 200% a year, it is non-sense to use a monetary stop loss. So, I changed it to a percentage stop loss and that helped a lot with my trading systems. And the other important aspect is using the volatility positional sizing. So, when the market is so volatile, I suggest reducing the size of the trade and doing the opposite when we are in a low volatility environment. That's a warning Mr. Placci, if you read my book on money management, you would have done that from the very beginning.
I mean, yeah, but I have to say that in traditional markets it is not so simple to do it. But in the cryptocurrency field, you have a lot of granularity and you can really use volatility position sizing and money management techniques, and this is very important. This is another aspect I like of cryptocurrency trading. And it was not possible on futures because, again, there you have 1 contract or 1… so you jump from 1 to 2, from 2 to 3.
So, it's not possible to trade a smaller or a better position size for your trading. And the other aspect is that it’s a market that is available for all the capitals, because commissions are calculated in percentage terms. So, you don't have any disadvantage in having a small account. Obviously, you will earn less than a big account, but the performances are not affected by your starting capital. And so, I developed 15 trading systems, 14 - 15, and I trade mainly Bitcoin, Ethereum, and BNB. There is also a decorrelation between cryptocurrencies, different cryptocurrencies.
I didn't realize that but it is useful to diversify using different cryptocurrencies. One can think that they are all correlated but they are not. Using your capital for trading different cryptocurrencies can be beneficial for your portfolio.
So, we can take a look, for example, at my actual portfolio. Let me share my screen. OK. This is a report, maybe in R, from real trades taking into account commissions, slippage, and the funding fee. And the red line is the cumulative equity line, the equity line of the portfolio, while the yellow line is Bitcoin, which as you can see did not do a good performance.
And then, we have in green Ethereum and in blue we have BNB. I started trading later. So, this is the point where the equity line starts, in April.
I made a 60% return with a drawdown of more or less 30% of my capital. This depends, at least in part, on the issues I told you about before: the infrastructure, the wrong stop loss, not using position sizing. But recently the performance is quite good also considering that the market had a really important drawdown, more than 50%.
And this is… And this is what I was saying before. When you invest you sit there, you don't know what to do, if you don't have a plan when things go as they did in the last months. On a system, you are telling the machine what to do, depending on the moves, and you've got all the rules put in place already. Yeah. About three months ago, I thought, “Maybe it would be better to buy Bitcoins and do nothing” but, after the drawdown you realize that managing the trade, the algorithmic approach is very important because the risk-adjusted return is much more linear than buy-and-hold or other kinds of techniques, in my experience, obviously.
So, now I have… Let me check… OK, 14 active systems. They aren’t in position today, this is a workspace made for this video. And so here you can’t see the automated trading active because of that, but I have a position on the market today. We can take a look at the equity lines: this is the first one on Bitcoin. It's a good equity line considering what happened recently. Obviously, not all the equity lines are like this, but for example, here you can say that the system stopped performing well when the market started going down because I trade only on the Long side.
This is… I’ve also short trades in my systems. I know, I know, but it's very difficult to develop good trading systems on the short side. I have 2 or 3 but... It’s true. The market is biased to the long side of course, due to having a huge developing edge. But I prefer to keep the short trades active, even if in some cases they break even at the end of the story.
But I’m ready in case it's needed. Yeah, this is the equity line on BNB, another system. This is on BNB. So due to the drawdown, the equity line is flat now.
And Bitcoin, one of the worse trading systems that I have. This is another on Bitcoin, but this is in recovery now, this system. Then we have Ethereum, you see it's on a new high.
Ethereum again. And Ethereum… I also noticed that Ethereum’s responding better than Bitcoin to most of the models. At least in the last year, yes, I confirm that.
This is on Bitcoin, as you can see recently, the systems on Bitcoin are not doing well. Ethereum is not so good in this equity line, but I keep it. Bitcoin, this is a quite good equity line. Then we have Ethereum again which is flat. And finally, another system on Bitcoin, which is quite good.
So, nothing special, as you can see. But compared to the real… what the market made recently, I'm satisfied because if we take a look at the equity line of the portfolio, the whole equity line of all of the trading systems, it's satisfying to see a new peak in the equity line exactly when the market is down 50%. So, yeah, this validates my approach, our approach to trading with trading systems. Yes, I see this on one side as one further way to diversify my portfolio and on the other side, the whole crypto world is really interesting because it offers different opportunities. Not only what we have seen here, but also still trying to understand it through an algorithmic point of view, there are other opportunities. I know you Francesco are also running some cash and carry strategies, which are with zero risks.
And what do you mean, it's zero risks? Apart from the failure of finance, which I hope is not happening of course, and of course, we are not now deepening this here, but there are interesting opportunities, especially when the market goes bullish. Now that we are in a bearish phase these opportunities are a bit reduced, but as soon as it starts riding again they increase. And I know you did something like 20% a month or something like that? Yeah, I made, more than a 40% annualized return, so it was 20% of real return because the expiry date of the future was after 6 months so, in annualized terms, more or less 40%. But the real return was 21%, something like that. And I closed it earlier, so before the expiry, and I made it a month. So 20% in a month is… It's amazing! Doing mostly nothing.
Just sitting and waiting. Absolutely. So, this is one of the reasons why I enjoy being here in this world full of opportunities now that - I want to stress this again - it's open to a decent connection with all the software we wish, especially MultiCharts, on the trading system part, but also on other possibilities to get and manipulate the data that we are getting. It's something that is challenging and thrilling because, again, new opportunities and new adventures in the trading world are always welcome. It's a bit of a “boring stuff” normally.
But when there are these new possibilities, we all are getting excited and curious. It's really about the numbers, not about the money. Money is there, of course, we are doing this for a living, so we need to make money out of the markets.
But it's interesting to do something where you can challenge your brain and then see the results, which is what often is missing in the normal world, let’s say. That’s it, is there something you wish to add, Francesco? No, at the moment no, maybe we can make another video in the future talking about, as you said, “cash and carry”, which is very interesting. I did! In the meantime, I invite you, listeners, to share your experiences on cryptos in the comments below so that we can learn from you, or if there’s somebody who purchased Bitcoin below 1, maybe.
And generally, all experiences are interesting. As you see, we are facing this from a systematic point of view. It's not the investment in cryptos, it's trading the crypto world. But, let us know what you do, what you did, if you are all trading through Binance or if you're also using other exchanges to place your trades, or maybe you are more active on DeFi, which is not what we talked about today.
But, let us know! We are curious to share your information and to know about you. That’s it, guys! Thanks, Francesco, thanks for today's video. Mostly thanks to Francesco who pulled me into this, and keep on following us on the channel because there is more to come! Ciao from ... Ciao a tutti …and from Francesco, Ciao! Ciao!
2021-07-30 15:42
