Watukushay FE - A Long Term Viable System

Hello Everyone,

I just earned my member status (not junior anymore !) and I wanted to start my first thread to share with you one of my trading systems, a system called Watukushay FE. This strategy is a likely long term profitable strategy which is absolutely systematic and fully automated. I coded this strategy as a gift for traders in December 2009 and it has been traded in several live real money accounts ever since.

The system uses a very simple trend following RSI technique which enters trades in the direction of a fast developing trend after a retracement has happened. The system adapts its SL, TP and lot size as a function of the 14 period daily ATR indicator's value and it should ONLY be used on the EUR/USD as other pairs do not show this particular inefficiency in such a significant way. (other pairs like the USD/CHF and GBP/USD give positive but yet much less profitable results)

This strategy has been evaluated significantly in the following ways and offers more evidence of profitability than ANY commercial system available :

 

1. Ten years of backtesting results EUR/USD Jan 2000 to Jan 2010
2. More than 8 months of live testing results on an account with real money. The account can be monitored via a myfxbook verified link, additionally the investor access information is available if anyone wants to directly monitor the account live.
3. Live/Back testing consistency analysis showing that live testing and simulations match (this was done in early June so it only features the first 5 months of results)

There are however several things you need to know about this strategy which are very important :

 

1. It is NOT a holy grail
2. There are profitable periods and draw down periods (below you can see a monthly profit/draw down analysis for a simulation of the past ten years on a risk = 1 setting)
3. Draw down periods can be LONG, they can last 100-200 days
4. The depth of draw down periods and the yearly profit depend on the risk level used

Now that you know the above feel free to learn about this system at the website I created for it at http://watukushayfe.blogspot.com. Please feel free to leave any comments, questions or suggestions. Remember that the system can be downloaded absolutely FREE of charge !

Thank you very much for your interest !

Best Regards,

Daniel

PS : If you would like the source code please send me a PM with the reason why you want it for and I'll be glad to send you a copy.

Interesting concept.

I noticed it seems you optimized the RSI values used. Is there any danger of some market-changing paradigm affecting the system?

Hello Ronald,

Thank you for your reply I understand your concern regarding curve fitting of the strategy, however the RSI values were optimized in a coarse and uncorrelated manner and RSI values around those values which were set are also profitable so even if small changes in the market that make the long term optimum deviate from these values happen the system should remain profitable.

Of course, there is the possibility of something happening that changes market behavior tremendously and makes the system obsolete but I have done everything in my power to make this the least possible case with coarse optimizations and the use of ATR adaptive techniques around the TP, SL and lot sizing

Thank you very much again for participating on this thread !

Daniel

upon my eyes reading the chart, it DOES look like a typical normal "significantly positive" distribution of 'monthly return'.

will look into, many thanks for your sharing.

Thank you very much for your post Charvo

I am glad you like the system. I believe that in order to trade successfully you need to know all the ins and outs of the systems you develop and use (either manual or automated) and for this reason Watukushay FE was built to provide reliable long term simulations and for these simulations to let us know the system better.

Please feel free to ask any questions or leave any further feedback you may have, thanks again for your participation

Daniel

Daniel, no problem, i should be the one who thanks.

i've read your blog, and this system is really simple (in a good sense).

have you ever tried to see what if you reverse your entry rules? in other words, sell when you should've buy, and buy when you should've sell?

i have a strong feeling that this opposite way is also working? since i cannot access your coding, i recommend you try it out. if your test may confirm my guess is right, i'd like to present an idea to expand the profit potential.

Hello Charvo,

Thank you for your comment The opposite logic does not work since it has negative statistical edge, due to this reason it loses money more rapidly that a random system would (since each trade has a higher probability end up as a loser than as a winner when you reverse the logic). I hope this answers your question

Daniel

Thanks, i just want to confirm it.

now my dumb idea here is:

would you please consider trying to add a 2nd entry parallel to the normal entry?

2nd entry is same entry price, SL, however, once floating profit is same as SL, you break-even this 2nd pos, and just let it run from ever since. (you might ask how to TP/exit then? that is really not an issue...)

i know that this is fully auto EA, but but, let's loose our mindset a little bit. watch the green floating equity curve when backtesting this idea, there might be surprise.

you can pm me with your email (your pm is closed), i can send u an EA of mine to show the idea.

Hello Charvo,

Thank you for your suggestion It is actually very common for people to suggest this kind of setups with systems like Watukushay FE. However the net result -for this specific system - is actually negative since the number of trades that hit the SL is not compensated by those that achieve to get higher profitability from a trailing stop or a similar setup that attempts to profit from much larger moves.

Although Watukushay FE exploits a trend based inefficiency attempting to follow "large moves" does not give good results. Overall the current implementation performs better than any technique similar to the one you have suggested. I hope this answers your inquiry

Daniel

thanks for your reply.

in fact, i think i didn't make myself clear. i'm not suggesting "similar idea" (be, trailstop, etc). my intention is to capture those yearly/monthly trend. not just a large move of 300 or 500 pips.

i backtest my idea with some tweaking vs your EA:

starting cap: 1000$, fixed lot = 0.01 lot (1 pip, 0.1 $), 2010/1/1 ~ now;

my data quality is messed up, not 90+% quality, but that doesn't matter (yes, it doesn't matter, though you may disagree :-) )

1st is my result

2nd is yours

again, there must be discrepancy from your backtest to my 2nd test due to my data quality, however, this is still a fair game between my 1st and 2nd tests.

1st test has inevitable drawdown, that's the price paid for the profitable positions, at the end of the test, those long term positions have to be closed, but in real world trading, you may keep holding them.

1st 's green floating equity frequently touch 1200$ line or even 1300$, but 2nd's equity never had such chance.

Hello Chavo,

Thank you for your reply Ok, I think I now understand what you would want to do a little bit better however I do not understand how this would increase the profitability of the system if you would never close the open positions. In order for this to provide an improvement there must be a closing criteria which allows the closing of these positions to be done, otherwise it is not possible to accurately estimate long term profit and draw down targets and as the market cycles all positions in the very long term will hit their SL. In order for this to be worth implementing there are two important things that would be relevant :

 

1. An exit criteria needs to be defined, otherwise profit is simply not being made and an accurate evaluation of long term draw down and profit (like that on the website and shown on the first post of the thread ) cannot be done.
2. There is the concern of swap interest which can accumulate negatively in an important amount when following this method if positions are to be left open for many months or even years.

Of course Watukushay FE has the potential to accumulate favorable positions in the direction of the weekly/monthly/yearly trends but these positions would need to be closed somehow in order to accurately evaluate the possible success of this strategy. If your method does not contemplate the closing of positions then I would not consider it worth implementing since I cannot accurately evaluate its potential and it seems pretty clear that in the very long term all positions would hit the SL.

Thank you very much for your contributions

Best Regards,

Daniel

the unfortunate part is that the exit cannot be defined in a complete way, but you might set 2000 pips.

i used to have same concern on "how to let pc define the exit", but now i don't. since in real life, i won't let my pc or vps run 2 month (or even 2 days) without my eyes checking, so i'm not really worried about exits for long-term position.

it is like you see 1 million$ on ground, you decide if you want to pick it up, not a difficult choice.

i suggest you to try the idea only because i feel reluctant to re-programme your logic though i know how to. and finally, i still did it.

sometimes, profitable idea is automatic; sometimes it is not, unluckily, my idea seems not this time.