UPDATE AS OF 2019
Believe it or not, 2019 is here. We will start this year by an article fully-about our own Millionaire Signal (https://www.mql5.com/en/signals/446507) and improvements we have prepared going forward. FX Sumo team believes that these are instrumental and fundamental changes, which need to be implemented in order to put our product next (or even above) those of the very best offered through MQl5 platform.
Before we present our prepared updates, let's look back and see what has lead for need of these.
Millionaire Signal was started at the start of July, 2018. Since the very first days, it was based on 2 fully-automated grid-type trading strategies. For first 4 months signal's performance was top-notch, showing ~17% growth, while keeping DD ~4%. Because there were not huge trendy moves in the currency markets, our signal managed to fully-exploit these market conditions and outperform most of other comparable strategies. Unfortunately, these ranging conditions quickly changed at the start of November, caused by Brexit-nonsense, new nonsense from belowed Mr. Trumph and his religious fight versus evil Chinese (joke). That lead to erruptive and massive moves in various currency pairs, which caused Millionaire Signal to struggle quite a lot. Despite the signal ending month of November with 4,5% growth, our DD had peaked at 27%, caused by grid-cycle on EURUSD. Our team reacted and shut down the signal for December.
What has changed since November? - well, let's go point-by-point to understand our mindset and the necesity of the steps that we took:
- The risk of the Ea that caused our signal to experience 27% DD has been decreased by half. These 27% DD was expected based on the available backtests, however, the situation was quite scary and made us feel absolutely-uncomfortably.
- as it can be seen in this backtest, this particular Ea has historically experienced DD as high as 54%, thus, these 27% should not be a surprise. Despite that, FX Sumo team felt uneasy with the idea that one particular algorithm could have such enormous affect on total results of our signal. By decreasing the risks of this particular Ea, these recently experienced 27% of DD should hold as medium-term benchmark that we should not breach (historical tops of 54%/2=27%). Obviously, one cannot just reduce the risks of a particular Ea and hope that result in terms of growth will be maintained - other steps need to be taken in order to compensate for that.
- Since October, 2018 FX Sumo has tested on live-trading conditions various other algorithms that could be added to the existing duo. We had to look out not only for algorithms that individually would be profitable and sustainable, but also would blend in with existing ones. By "blending-in", we mean 2 things:
- a) low correlation with existing algorithms and their results - Markowitz Porfolio Theory:
For those that look for an interpretation, here, by "individual assets" we understand different algorithms, volatility=drawdown, expected return = growth. What this basically says is that there is mathematical way of combining different financial assets under one that would have lower volatility and higher expected return values than any of the included assets individually.
- b) we want not only to financially diversify, but fundamentaly diversify the underlying strategies. The amount of currency pairs, as well as traded-timeframes, will be increased. Not only that, added strategies should have fundamentally different trading approaches, which would be more acceptable by mainstream followers - strategies with usage of SL/TP levels (Misjudgement of hedging, grid trading and use of SL).
- Here, adding strategies with active usage of SL/TP can be quite tricky due to possible over-optimization risks. These strategies tend to be based on breakouts, which in past simulations look obvious and easily-predictable, however, that's not the case in real trading conditions. We have forward-tested the strategies that we are going to implement in Millionaire Signal - so far, forward tests match backtests and results are positive, which gives us room for optimism. Backtests of "mainstream strategies" that will be added to Millionaire Signal can be seen below:
- Yes, all 4 of these are different breakout strategies, specifically designed for those particular currency pairs and timeframes. Should they work in real trading conditions as well as in backtests? - time will tell. Either way, risks for each of these Eas individually will be relatively low, thus, affect of having each of them "on-board" shoudn't be too high.Currently we have 6 different Eas - 2 old grid-type of Eas and 4 new breakout Eas. The way we look on this set - 2 old Eas should continue to bring-in approximately 3-4% growth (with reasonable risks) on monthly basis, while the set of 4 added will be the one that will define how efficiently portfolio will perform. Either way, the oldest 2 should be able to create enough of ground to tolerate even off-months for the "4-horseman set".
- While testing the newest 4 Eas on live-trading conditions, we started to question whether the existing 2 Eas will generate enough of "expected returns" to off-set any potential problems with the 4. When November had ended, we realised that it could not be enough, therefore, we started to look for additional low-risk grid-type of Ea that could help us to sustain the needed ground. We believe, we managed to find an amazing addition to our 2 corner-stones:
- So what will we have prepared on 7th of January, 2019? - set of 5 new Eas - 4 of which will be based on breakouts, 1 will be based on grid-type of trading. Do You remember the point about correlations? - take a look on this:
Please, zoom in the browser to better see the picture - couldn't enlarge it any more.
In theory, all of this should work out wonderfully for Millionaire Signal in 2019, as we have tried to diversify the trading as much as possible. Will it work atleast somewhat similarly in real trading conditions? - only time will tell, but we have our reasons to be optimistic. We feel that throughout month of December we have taken the needed theoretical steps to improve our product, and we hope that this will materialize in real results.
That's all from our side. If You have any suggestions or You disagree with some of the ideas, feel free to leave Your comment below!
Best of luck to all of us,