Steve Hopwood's Epsilon Megastar fib auto-trading robot

There will be no further development of Epsilon Megastar. Development led to pathumracled bringing Simone Guy's thread to my attention. The robot for trading her system can be found at http://www.forexfactory.com/showthre...79#post3779479

____________________________________________________________________________________________________
Welcome to the stellar world of Epsilon Megastar.


The name of this trading robot arose in one of my earlier threads. We were laughing about the way in which EA's are presented for sale on the www, and I came up with a spoof robot that I described with similar hyperbole. This has cropped up from time to time and gained instant acceptance as the name for this robot, when I announced it would be coming in the swing trading thread.


I shall be very, very disappointed if this robot does not turn out to be a winner. Yes, there are loosing trades but the trading system uses pure price action to time both trade entry and exit.


There is an old adage in forex trading: 'cut your losers short and let your profits run'. The first part of this is easy – stop losses do that for us. The second part is much, much harder. We are all familiar with the market powering through our take profit and carrying on into the stratosphere – irritating, to put it mildly. The strategy that this robot trades takes care of letting profits run. This is the most simple and direct trading strategy I have come across. Miguel sent it to me, and we are in his debt for this.


In the 'TradingStrategy-QuantumGlobeInc.zip file attached to this post is the trading strategy document, 'TradingStrategy-QuantumGlobeInc.pdf'. This doc looks daunting at first but is actually easy to read, and well over half of it is taken up with trading examples.


To understand what is going on here, having studied the strategy pdf you need to be able to explain ICWR to someone who has never come across it. Read the pdf, study a chart, and take an imaginary person through the information needed to understand ICWR. If you cannot do this, then your best interests are served by reading the pdf again, and again if necessary until you can do this 'teaching' exercise.


Newbie traders, you need to understand these concepts:

 

1. 'swing high' or 'swing low'
2. trending markets
3. retracements
4. ranging markets

There is another doc in the zip for you if you are in doubt about any of these concepts – 'Basic Stuff.pdf'. If you are new to Fibonacci support and resistance levels, then view the video at http://myspace.vtap.com/topic.html?d...e&usrn=BoBokus for the simplest fib trading method and explanation I have ever seen.



Read the user guide to find out how to set up your charts, fibs, and the robot. This one is already trading on my live account. There should not be any bugs as I took time over its coding and have not released it until I am happy that it will work 'out of the box'.


There are a few 'tweaks' to the method, with regard to trade entry and exit; read about them in the user guide.


Future development will come from:

 

1. Looking at sending trend-direction pending trades during the corrective wave, at the 50 or 61.8 (falling trend) and 38.2 (rising trend)
2. Hedging losers instead of using a hard stop

These will be matters for discussion as our use of the robot develops.


Posting restrictions
I have restricted the ability to post in this thread to: traders with at least 1 voucher; and traders on my 'buddy' list. I have done this because I do not want the thread cluttered with 'newbie' questions about the basics. New/inexperienced traders, this is not intended as a direct insult. What I want here is discussion with traders who fully understand what it is they are doing; when you newbs have been around a while, you will come to understand what a difference experience makes to the posts a member writes and blanche at the recollection of some of the questions you asked. I do. http://cdn.forexfactory.com/images/s..._the_floor.gif Go to your profile page; there is a link to the FF page that describes the voucher process.


Traders with experience who have not yet been vouched for, can get around this restriction be being added to my 'buddy' list; all you have to do is pm me to ask me to do this.

For new/confused traders, I have set up a partner thread to this one where you can ask your questions - http://www.forexfactory.com/showthre...89#post3727389

A note to coders
I retain the copyright to this code, but that is a de-facto state of law. So far as I am concerned, it is open source. Coders should feel free to adapt it, pinch it, use it in their own robots. Please just respect the spirit of FF and refrain from using it in a commercial product.

Updates section.
The user guide is up to date.

The robot now has the ability to draw two higher time frame fibs as an aid to manual traders.

Time to play!
Just wanted to let u know the 'basic stuff.pdf' isn't in the zip file

Yes it is. Ridiculous idea that someone so well organised as me could possibly leave something out. Wash your mouth out with soap and water.

Given the length of time it takes to upload the zip, and so presumably to download it, I think I had better separate the trading strategy pdf into its own zip.

Thanks for the heads up.

Another observation thusfar, watching it in visual backtest (i know u hate it but i want to see it work).

The scale in feature seems to only work if the trade goes against you. Not sure if that was your intention or not. If you get a good entry and are instantly in profit, it doesn't scale in every 10 pips like it should (my interpretation of the pdf anyhow).

Something to look at, or tell me I am wrong

No ECN? Total bummer. I like to test on the same broker I use live.

Thanks Steve - will get this one fired up.

Paul

Just a word of warning about this method to save some the frustration.
This complete rubbish (to use a steve word) on gbpusd
Great on eu however, at least for the last few months anyhow
Play with it manually or with the ea on gu and you'll see how bad it is, i guess gu just doesn't like fibs or solid movement
Guess its like trying to surf in a pond with speed boats going by (gu) as opposed to riding a nice ocean wave (eu)

For what it's worth

Backtesting doesn't work with this robot. Don't bother ever trying. You are wasting your time.

Guess how I know.

okey doke, seems to work pretty well other than the scale in issues, but will keep an eye on those in forward test

No, there is no ECN. My opinions of this blatant attempt to separate punters from their cash are well known.

I will repeat them if you want, but you will not like them. Most people do not like being described by association as idiotic, naive, criminal fodder etc.

If you are daft enough to insist on using one of these criminals to host your account, then use mptm to provide an instant stop loss.

Don't blame me when the inevitable happens and a trade is sent:

 

1. just before your system crashes irretrievably.
2. the market crashes irrevocably against you because of an unexpected event, and wipes your account.

Remember this unavoidable rule of forext trading: if it can happen, and it is bad for you, it will happen.

This is why the crims offer ECN accounts; they are simply waiting for this to happen.

Now, take this on board guys. I mean it seriously.

Mentioning ECN from now on earns the poster an instant ban from all of my threads. I do not care whom that individual is.

If umpteen people decide to call my bluff and flood this thread with stuff about ECN, I shall simply close it and go my own way.

That is how strongly I feel about this blatantly criminal attempt to separate the naive and stupid from their cash.

Logically, this method will work in trending markets and we all know this is the EU - currently trending all the way down to Hell in a handbasket.

It looks to me as though this method will keep us out of ranging markets. Is this not your experience?

for GU, definitely not my experience. Doesnt take long to see on a gu chart that this method just burns money like its going out of style. Maybe can be helped by not using such a high 'trend filter' as the daily, not sure.
But this combination of daily rsi to determine trend, and the fib levels is total nitemare on gu.
I've spent many hours looking at gu this week already, and the up/down movement of this pair just laughs at these fibs levels it seems.
Like I said, eu loved it, and hopefully some other trending pairs love it as well.
Why the author even mentioned gbpusd in the doc baffles me.
I hope you don't have gu on your live trading

Steve.....I'm speechless....Every day it's becoming an endless sea of ribbons, wrapping paper, ingenous preassembled toys and even non-fiction fore those of us who like to read!!!
I will stick with the word excellent. Thanks for sharing the fun........

whip

I'm loading as I type!

Steve's screenshot on the other thread does show GBPUSD. How does this pair look in your results thus far this week, Steve? Does it look like some pairs work and other's don't, or are all pretty even thus far?

Cheers tazz. This post will go into the user guide. Copyright on the trading doc is 2005; I assume that explains the difference between now and then.

I had got the robot on the GU and there were a couple of trades open, happily both in profit so I have closed them and closed the chart.

That leaves me with EU, EJ and GJ. Frankly, if I eventually get left with EU I will not cry. I only actually need one successful pair, after all.

Thanks for this tazz; much appreciated.

On demo 5 trades: 2 winners but an overall profit of about 50 pips because the good trades went a long way. This is easily explained by problems in the UK this week and cannot be used as a representative sample.

Live, just about at breakeven - not good for a pair on its way to the bowels of the earth.

All of this needs a caveat. I had a lot of trouble getting the trade closure code correct, so my results are skewed by this.

Hi Steve,
I see you've added an additional RSI filter, where it has to be greater or less than during previous bar to match the trend. Was the strategy proving unprofitable without that?

Once again Thank you Steve.

For what its worth, my 13 pair demo breakdown shows that over 4 days of testing, EurUsd, AudUsd and GbpUsd are the only pairs that have contributed favorably to the account. Of the other pairs, not all have had trades, but the xxxJPY pairs have been net losers so far. Not statistical results mind you, but all I have for now.

I am seeing that many trades are hitting the stop loss rather than seeing a close of the trade due to a reversal. A reevaluation of the stop loss setting might be in order for some pairs.

No. It just makes sense to me to not send a trade when the more immediate Rsi values are moving against it.

Initially, I had the same filter to trade closing, but this appeared to stop the closure process altogether. Now, I think these probs may have been caused by my difficulties with the closure logic rather than the filter.

Here is what I am thinking:

 

1. there is not a lot of point in taking out a trade when the market is already moving against it.
2. there is not a lot of point in closing a trade when the market is already moving back in favour of it.

After all, moves from out of the retrace levels in favour of a trade, or back to the opposite confirm levels against an existing trade are already quite large moves.

So, I use ltf Rsi to confirm trade entry, and candle direction to confirm trade closure.

Whether to use one method for both, and if so which one, is one of the purposes of this thread.

All ideas and viewpoints welcome.