CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. 64 % of retail investor accounts lose money when trading CFDs with this provider. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.

DIR - 100 % automated trading - by IlIlIlIlI

Hello,

I want to give you some information about my Darwin DIR.

The objective of this strategy to create permanent profit with acceptable drawdowns.
The trading is made only by an EA, which manages more than 20 currency pairs with the same trading logic.
The basic trading model is a trend following model.

All positions have a dynamic intial stop and take profit calculated from the chart. There are several conditions defined which have to agree before a trend is accepted for entering a trade. a tight trailing stop is used when a position comes into profit. The trading logic is used for accasionally manual trading since 2017.

It is possible, that more than one trade is entered for one asset. That happens only if the signal to enter a trade is repeated. Price levels are not used. The number of trades possible on the same asset is limited, but can be changed.
Example (picture from a MT4 backtest):

The EA trades the 1D timeframe, better said, 24 hours timeframe. That has the advantage that when positions are to close by differences in the trend signals it mostly happens during trading hours with high liquidity and not at midnight when the liquidity is usually low and spread and slippage are high.
The timing to entered a trade and to set the trailing stops is calculated from the 1 hour chart.

The following pairs are currently traded (when the column acitve shows ‘Y’), the results are calculated from January until August 2019 from the demo environment which is about 95% reliable.
All pairs offered by Darwinex are checked at least once per month and are added if the EA can produce a stable profit with them or removed if the daily chart doesn’t show reliability for trading a pair (like currently USDCHF).

SPA35, WS30 and SPX500 are not traded as long they are not offered for 0.01 lots (1 microlot). The max. DD could be significantly extended by trading them with the available lotsize of 0.10 lots (1 minilot).

Entering a trade is not manually controlled, it only depend on the signals generated by the markets behaviour. So there will be times where the trading activity is low or the EA is very busy as it is right now.

As the EA is trading the daily chart, I don’t expect the perfect attributes here. Most attributes give better values if the trading is more short term, I assume.
While the EA had a low trading activity significantly under average in August, it is now hyperactive since a week. The VaR is expected to stay in a range of 8 to 15 %, during the current massive trading activity it could also be slightly higher.

The EA is not written to paint nice green attributes, but to make solid profit with acceptable darawdowns. So some intraday drawdowns must be accepted just as ‘noise’.

My expectations for this strategy on the Darwin DIR:

  • a continuously monthly return between 2 and 12 %
  • return for the year 2019 should show more than 25 %
  • yearly or any 12 months performance should be higher than 50 %

And this is the picture this Darwin starts his journey at Darwinex:

5 Likes

no pressure then :sweat:

best of luck

2 Likes

Wish you the best @IlIlIlIlI for your DARWIN :slight_smile: !

2 Likes

Can you talk a little about what optimisation you did?

Great results.

2 Likes

Thank you for your compliment to my Darwin.
I hope I got your question right and I can add this information:

I didn’t use any typical optimization as the MT4 optimizer. That’s a point which makes me confident that the strategy cannot be overoptimized and will be successful in different or more difficult market situations.

For every pair I use the same value (in pips) for take profit and stop loss, where I use a minimum value and a volatility depending value derived from an indicator. The only exception is WTI because the basic calculation will generate unrealistic values, but also there I cut it only by a percentage factor to a realistic level and it works fine. The factor is derived from studying the chart and not from the MT4 optimizer.

Pairs, where I would need a different value than my standard values to create substantial profit, are currently not traded. I have enough assets for this account size. :slight_smile:
I didn’t even try to get an individually optimized value for sl or tp for every pair. I check every unused pair offered in MT4 at least once per month.

Additional I use a tight trailing stop when the position has enough profit that it would not become a loser even if closing it on the trailing stop is executed with slippage. That creates the high profit rate.

More than 80 % of the positions are closed on this trailing stop, less than 5 % on the take profit, less than 2 % on the initial stop loss and the rest is closed with profit or loss when one of my trend indicators from the 24 hours chart looses the trend direction.
You can also see that for positions in the graphics shown under the La attribute and for the single trades under ‘assets & timeframes’ -> ’ Max. Positive/negative excursion per trade, including open trades’.

So you have the same parameters for 20+ assets but you backtested only on 8 months of hystorical data…
What happens if you backtest on 2018 or 2017?

1 Like

Yes, I backtested only on 2019.

I do permanent backtests starating January 2019 to watch the developement of the assets.

I don’t know what would have happened in previous years because I did not test them. My opinion is that previous years are not very relevant for 2019 until now, what you can also see with many strategies being successful until beginning of 2018 and are suffering since 1 1/2 years.

The other point is that a backtest for previous years could be not very relyable because I suspend assets from trading when I see that my strategy could not find a stable trend on the daily chart, like currently on EURUSD.

But you think that parameters working on 8 months of 2019 will work in 2020 and 2021 …:thinking:

3 Likes

I used this parameters unchanged for sl and tp years ago on an older EA where I don’t use the trade entry logic anymore and the EA was made only for a single chart/assets.

1 Like

Have you considered how your ea will perform in the market on Monday during the conditions resulting from the Saudi oil attacks? Quite unprecedented conditions across most asset classes tomorrow i suspect.

1 Like

Do you think WTI will raise on Monday?

We have also an expiry date for WTI futures end of this week and are currently on the bottom before expiry - in my opinion.

The strategy holds a long position in WTI since Tuesday which suffered in the last few days, but a large part of the DD on it is covered by successful trades in other assets.

The advantage of diversification in different assets cut this DD significantly.

Besides the market for WTI was open in time, the XTIUSD CFDs were not traded for more than 15 minutes in the announced time window at 18:00:01 NYT.

That looks like a very bad CFD provider (LMAX? Saxobank?) trading obviously against his clients, as this XTIUSD position was not closed at the best price (about 2 USD higher than the take profit) or another price higher than the take profit as expected and seen in the more than 15 minutes on the market when the provider didn’t trade the CFDs.
Now I hope they will be closed on their original take profit soon to lock the profit.

The EA trades as usual, this Darwin is 100% automated trading made by an EA.
I would only intervene if there is a technical issue that the EA could not trade as it should.

Edit:
The positions are now (10 minutes after posting) closed when the EA raised the trailing stop close to the take profit and the profit is locked on the account and the Darwin.

On the second days after officially published my Darwin DIR found the entrance to the ‘On Fire’ filter for ‘yielding the best returns with moderate drawdown, in the last 3 months.’

1 Like

Thank you for your congratulations. I also want to see a new all time high every month :wink: - and I don’t know what is harder to reach.

It is nice to see you introduced your Darwin from the early stages, this is a good sign. few bother doing this.

2 Likes

I think that is what a potential investor need for getting confidence in my (or any) Darwin - to have the chance to get the information live. That is an advantage migrated accounts can never deliver.

If he likes my Darwin, he could put it on his demo portfolio to watch it or - with more confidence - try it out with his personel minimum investment amount as I did …

… and then he would see trades made in the tax report after the month ends. So conficence can grow.

My small but profitable real portfolio does not need an own thread :slight_smile: as long as there is only my Darwin bought.

5 Likes

Hi nice start - well done - I note this is a 90% win rate system, with losses double that of winners ( pips) which always makes me a little uneasy, in that it maybe avoiding closing losers early enough. I also note that the maximum positive excursion seems to be steadily falling in value since starting - especially in September. Please don’t take this a criticism, just observations at this stage…and impossible to really judge with such a short history to date. Do you plan to add any more to your own investment, to show confidence in the system? Apologies if any of this has already been discussed and good luck!

2 Likes

Thank you for your compliments to my system.
Your observations are right, but as long as the positive excursions per position are larger or nearly the same size as the mngative ones I don’t read it as a negative characteristic of my strategy.

September is a difficult month for this strategy as for most trend following models like mine, just this week is additionally overshadowed by the Saudi oil attack which made all markets nervous on Monday and today is an expiry date for quarterly futures. But September has 6 more trading days, with hopefully more mormal trading activities than today, and we’ll see how the month will be finished.

The high profit rate is the result of the tight trailing stops which are set when the trade comes into profit. That’s an essential part of this strategy and responsible for a large fraction of its profit, especially of the currency pairs traded. It helps significantly to keep the DD low when the trading activities are splitted on several assets (currently six positions in different assets), mostly with the advantage of diversification.
Since the programming of the EA was done about 3 months ago, I run it on a demo account in the same version with identical parameters as on the underlying account of DIR with profit on the demo account’s equity of more than 30 % and a DD less than 8 %.

Currently it is not planned to add funds to the trading account. It is not necessary to run this strategy. I will change my mind if the VaR runs over 20 % and wants to stay there, my range where I want to see it is 8 % to 15 % with some temporary tolerance at this borders.
I also think some investors are trapped by a higher account balance, not regarding that the trader’s risk is not shown in the account balance alone, you also have to regard the VaR to get a vague image about the real risk of a trader. And I prefer convinced investors on my Darwin. :slight_smile:
My planning is more to setup a second Darwin in the next months, the last strategy I tested on the trading account of my (always closed) Darwin HRE is in live test on a demo account with good results. On the closed HRE there were more strategies tested and running. That is the main reason why it was closed - all attribute values and developement would be completely unusuable for the running strategy for a very long time.

1 Like

First I want to welcome my first investor (but me) and say ‘thank you’ for his confidence.

The Darwin had to take a drawdown of more than 16 % since middle of September which pushed the return since published in the red zone. This is the worst result of the year 2019. So the 4th quarter will start with moderate losses.

More than 70% of the drawdown was caused by trading CFDs, while only 35 % of the backtesting profit was made by CFDs.

So the following measures were taken to improve the results in future:

  • all CFD assets will be traded only with 1 microlot per position, currency pairs stay unchanged at up to 4 microlots per position
  • AUS200 CFDs are suspended from trading.
  • the currency pairs CADJPY and NZDCHF are suspended from trading.
  • USDCHF is reactivated and will be traded again as the chart shows more stability for a trend and on backtesting there were some positive trades in September which proved that

The VaR (currently 16.90 % shown under the Rs attribute) raised up to more than 17% during wild trading until the beginning of October, but with the reduced trading volumes I’m confident that it will come down below 15 % in the next two or three weeks. So there is no action required to bring it down.

All changes were impemented to the EA on the weekend which was easy as the Darwin was flat.

1 Like

Trading is paused until 17th of October.
Today’s market behaviour is too chaotic for an EA trading the daily chart, and October futures expire next week so I think it would stay extreme.
Open positions stay under management of my EA.

I’ll see on my demo accounts would would have happened in these six days.

1 Like