Big Sandwich System | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

The System Behind the System, by Merlin Jeffries 16 replies

Who has made BIG money with a BIG trading account consistently? 14 replies

Effilang's Sandwich... 33 replies

Big jump or big drop ? 3 replies

Would you trade this system? New system #2. 10 replies

Big Sandwich System

  • #1
  • First Post: Edited Jun 18, 2007 8:45pm Jun 17, 2007 11:55pm | Edited Jun 18, 2007 8:45pm
I have had quite a lot of interest in my Big Sandwich System, so instead of repling to everyone, I will share the principals of the system here.

What I will reserve is the disclosure of the exact mechanical system because it works on various markets and I don't want to give away my buy & sell prices for obvious reasons.

Not withstanding the above, the system below has been profitable for me whether discretionary or mechanical. If anything, the discretionary system is superior in terms of performance. I trade the system mechanically for two reasons:

1) workloadrofit ratio is better
2) I call it the "drifting boat effect". If you are fishing with your back to the shore, you could gradually drift away without even noticing. The same goes with discretionary systems, over the months you could drift away from the core system, venturing out into the unknown.

There are no "X" factors or secret hocus pocus. I will not post one or two trades to "prove" that my system is profitable. You are welcome to DEMO TRADE this system, and if you follow the principals and it hasn't worked at the end of a 3 month period, I will declare myself a fraud and leave the forum. I have posted my account statments on this forum, and I am willing to do so on request (within reason i.e. not every single day).

Ok, enough babble from me, here is the principals of the system.

1.0 LOGIC OF THE SYSTEM
To me this is the most important aspect of a system, and probably the most overlooked in relation to system creation; "WHY should it work"?

a) I believe that this system works because on daily charts the markets spend more time ranging than they do making new highs/lows. This means more frequency.

b) the range represents a zone of uncertainty which will not be substantially broken until further information is brought to the market. This is our edge.

2.0 SYSTEM PRINCIPALS
2.1 Create a channel. Use handdrawn lines or Bollinger bands or anything you like. My preference when trading discretionary is the hand drawn trendline. (see image as these channels are a little different to conventional channels)

2.2 Place a buy signal at the bottom of the range, and a sell signal at the top of the range.

2.3 Place a stop which is the width of the channel. $1600 is the limit for any pair.
NOTE: You will notice my large loss on the USDCHF on my statement. This is because the stops are wide, but the win rate is high to compensate. I will also show you how that loss was not really a loss at all!

2.4 Once price has crossed the centre of the range, set the stop to breakeven.

2.5 When price reaches the opposite end of the channell, then place your order in the opposite direction as per 2.2 above.

2.6 Exit your original order at the 50ema on 1hr charts.

They say "a picture is worth a thousand words", so here is two of them. Please note that as per my posted account statements, these are trades that I am actually in.


A class 1 trade is with the trend, and class 2 trade is against the trend. I take both, but I keep a tighter lead on class 2 trades by entering on a breakout on a 60min chart, and exiting with some discretion if the price starts to turn against me as it closes in towards the opposite side of the channel.



***PLEASE DEMO TRADE ONLY. ALL CARE BUT NO RESPONSIBILITY TAKEN. THIS IS NOT AN INVITATION FOR YOU TO TRADE, BUT TO RELAY HOW I TRADE****
  • #2
  • Jun 18, 2007 2:51am Jun 18, 2007 2:51am
As an additional note, I am still long in the GBPUSD with a 50ema trailing stop on the 1hr charts.

I have also sold the GBPUSD at 19829 as per the system.

This means that I am in a nett netural position with a 225pip profit, so you can see how that big loss on the USDCHF was not really a loss at all!

From here only 2 things can happen:

1) the gbp moves up and my position is neutral until my stop is hit, then my long trades goes into even more profit

2) the gbp moves down through the 50ema locking in profits and adding profits to the sell order
  • #3
  • Jun 18, 2007 1:30pm Jun 18, 2007 1:30pm
Greetings Mongoose,

Thank you for sharing your system. Please consider these observations:

Your pictures did not come through for some reason.

Your method is a bit unclear (possibly this is on purpose). If you can clarify, here is what isneeded:

1. What time frame do you use? Daily?

2. What pairs do you use on this?

3. What time do you monitor the market for the "range"?

4. When do you enter your orders?

Thank you.
  • #4
  • Jun 18, 2007 7:30pm Jun 18, 2007 7:30pm
1. What time frame do you use?
Daily and 1hr

2. What pairs do you use on this?
GBPUSD / USDCHF / EURUSD / USDJPY
Mechanically it works well on all the volatile pairs. The AUDUSD is too trendy so ones like that can be dropped from the list.

3. What time do you monitor the market for the "range"?
Daily

4. When do you enter your orders?
Every day.


The entry is largely uninportant, so any logical form of channel will yield similar results to my mechanical method.


After a little persuasion from a forum member I have decided to disclose the final part of the system. This member has convinced me that a despite a few unwarranted snarls and scoffs when I entered the forum, there are members that could offer improvements to the idea.

Inserted Code
inputs: dollarStop(500), emaLength(10), exitEMALength(50); 
{$1600 stop for the USDCHF, $1200 stop for the GBPUSD}
vars: upperEMA(0), lowerEMA(0), totTr(0), prof(0), tradeStr(""), middleEMA(0), breakEvenEngage(FALSE), numContracts(0);
 
upperEMA = xaverage(high,emaLength)[1] of data2; {data2 is daily}
lowerEMA = xaverage(low,emaLength)[1] of data2;
middleEMA = xaverage(open,emaLength) of data2;
 
numContracts = 1{intPortion(((50000 + NetProfit)*.10)/2000)};
{*********************************************************SELL SIGNAL**********************************************************}
if marketPosition > -1 and high crosses above upperEMA then sell numContracts contract at maxList(upperEMA,xaverage(close,30)) limit;
{******************************************************************************************************************************}

{**********************************************************BUY SIGNAL**********************************************************}
if marketPosition < 1 and low crosses below lowerEMA then buy numContracts contract at minList(lowerEMA,xaverage(close,30)) limit;
{******************************************************************************************************************************}

{*********************************************************EXIT SIGNAS*********************************************************}
if marketPosition = 1 and  high > upperEMA then exitLong("LX Target") at maxList(upperEMA,xaverage(close,exitEMALength)) limit;
if marketPosition = -1 and low < lowerEMA then exitShort("SX Target") at minList(lowerEMA,xaverage(close,exitEMALength)) limit;
{*****************************************************************************************************************************}

if marketPosition = 0 then breakEvenEngage = FALSE;
if marketPosition = 1 and high crosses above middleEMA then breakEvenEngage = TRUE;
if marketPosition = -1 and low crosses below middleEMA then breakEvenEngage = TRUE;

if breakEvenEngage = TRUE then 
begin
 exitShort("SX BE") next bar at entryPrice stop;
 exitLong("LX BE") next bar at exitPrice stop;
end;

setStopContract;
setStopLoss(dollarStop);
  • #5
  • Jun 18, 2007 8:32pm Jun 18, 2007 8:32pm
Mongoose, are you trying to be vague on purpose? If so, then I will not ask you any more questions about your system.

Basically your answers below are "daily." That doesn't help much. How exactly do you trade this?

Again, if you are purposefully being cryptic and vague then I won't ask anything further, but then again that makes me wonder why you posted here in the first place.

So you set some kind of a channel, look on the daily charts, enter a trade daily, and daily close or take profit?

I don't know anything about the code posted below, any chance you can post it as mq4 or ex4?

Quoting Mongoose
1. What time frame do you use?
Daily and 1hr

2. What pairs do you use on this?
GBPUSD / USDCHF / EURUSD / USDJPY
Mechanically it works well on all the volatile pairs. The AUDUSD is too trendy so ones like that can be dropped from the list.

3. What time do you monitor the market for the "range"?
Daily

4. When do you enter your orders?
Every day.


The entry is largely uninportant, so any logical form of channel will yield similar results to my mechanical method.


After a little persuasion from a forum member I have decided to disclose the final part of the system. This member has convinced me that a despite a few unwarranted snarls and scoffs when I entered the forum, there are members that could offer improvements to the idea.

Inserted Code
inputs: dollarStop(500), emaLength(10), exitEMALength(50); 
{$1600 stop for the USDCHF, $1200 stop for the GBPUSD}
vars: upperEMA(0), lowerEMA(0), totTr(0), prof(0), tradeStr(""), middleEMA(0), breakEvenEngage(FALSE), numContracts(0);
 
upperEMA = xaverage(high,emaLength)[1] of data2; {data2 is daily}
lowerEMA = xaverage(low,emaLength)[1] of data2;
middleEMA = xaverage(open,emaLength) of data2;
 
numContracts = 1{intPortion(((50000 + NetProfit)*.10)/2000)};
{*********************************************************SELL SIGNAL**********************************************************}
if marketPosition > -1 and high crosses above upperEMA then sell numContracts contract at maxList(upperEMA,xaverage(close,30)) limit;
{******************************************************************************************************************************}
 
{**********************************************************BUY SIGNAL**********************************************************}
if marketPosition < 1 and low crosses below lowerEMA then buy numContracts contract at minList(lowerEMA,xaverage(close,30)) limit;
{******************************************************************************************************************************}
 
{*********************************************************EXIT SIGNAS*********************************************************}
if marketPosition = 1 and  high > upperEMA then exitLong("LX Target") at maxList(upperEMA,xaverage(close,exitEMALength)) limit;
if marketPosition = -1 and low < lowerEMA then exitShort("SX Target") at minList(lowerEMA,xaverage(close,exitEMALength)) limit;
{*****************************************************************************************************************************}
 
if marketPosition = 0 then breakEvenEngage = FALSE;
if marketPosition = 1 and high crosses above middleEMA then breakEvenEngage = TRUE;
if marketPosition = -1 and low crosses below middleEMA then breakEvenEngage = TRUE;
 
if breakEvenEngage = TRUE then 
begin
 exitShort("SX BE") next bar at entryPrice stop;
 exitLong("LX BE") next bar at exitPrice stop;
end;
 
setStopContract;
setStopLoss(dollarStop);
  • #6
  • Jun 18, 2007 8:45pm Jun 18, 2007 8:45pm
Mongoose, Nice charts. I like the idea of this method, however your charts are too large and are messing up the screen display. I have to delete them. Could you please re post them smaller maybe 600 x 600?

BTW for those who want to know, the program is EasyLanguage for Trade Station..

thanks
  • #7
  • Jun 18, 2007 9:03pm Jun 18, 2007 9:03pm
Daily chart showing 10 ema of highs and lows as channels

http://img503.imageshack.us/img503/3...teurusdne2.jpg


Discretionary method showing hand drawn channels as channels
http://img502.imageshack.us/img502/1...plifiedqv9.jpg

Exit Technique on 30min chart
http://img503.imageshack.us/img503/1...trationog2.jpg

The red lines below are the DAILY 10 ema of the highs and lows on a 30 min chart.

http://img502.imageshack.us/img502/5...ch30minku9.jpg

http://img502.imageshack.us/img502/3...sttradeyt0.jpg
  • #8
  • Jun 18, 2007 9:03pm Jun 18, 2007 9:03pm
http://img407.imageshack.us/img407/2...chgbpeczt1.jpg
  • #9
  • Jun 18, 2007 9:05pm Jun 18, 2007 9:05pm
http://img502.imageshack.us/img502/3...monthlywn5.jpg
  • #10
  • Jun 18, 2007 9:11pm Jun 18, 2007 9:11pm
Looks like you have enough rules and coding to actually try and backtest the system on tick data. When you do i think you will be dissapointed to find it loses money. All these types of systems lose money unfortunalty but its worth finding out the easy way and check it now rather then see your hard earn cash go down the pan.

Here are the reasons why

1) 90% of traders lose money so you are looking for a 10% edge thus you in a bad position before you even start.

2) Every instrument you have in your toolbox is a linear math derived from the history but forex is not linear so it it wont work.

3) If you backtest and optimise past data all you are doing is finding a linear curve fit solution to past data. It is enough sometimes to let you know a bad system has no chance for the future if it cannot be made to work at all in the past. It will possibly work for a few weeks into the future before the linear fit means nothing any longer.

If you dont believe me, then find me any EA which worked great last year and is still working this year and makes real money live. They are as rare as rocking horse crap.

4) If your serious about finding a system you must find a chaotic based probabilty system which thus is non linear and uses no dials, channels, ma's or stochy stuff whatsover. OR use self adaptive neural networks and/or Genetic progamming based systems as these are non linear. When futures brokers fixed odd bets, binary bets and options are set they only assume one thing. The market can go up, flat or down within a range of what it has done very recently ie ATR they dont expect anything else so why should you?

I give you an example. In 5 years worth of euro days 33% will be long over 20 pips a day, 33% will be flat within 20 pips a day and 33% will be short below 20 pips.

5) Every pair is complelty different to another there is no such thing as one size fits all.

6)There are 2 timeframes which do work though with reasonable results. High speed automatic ECN based scalpers running on tick data looking for 1 or 2 pips but they are out of reach of most all retail brokers due to their rules, speed of execution, stop placement and slippage OR trade long term ie daily and over and try and find the bigger trends. Although this can work its boring as hell and no fun so not many do it.

7) The best way to lose money is trade intraday without solid money man and on hunches based on other people advise/thoughts/signals.

Best of Luck
  • #11
  • Jun 18, 2007 10:05pm Jun 18, 2007 10:05pm
Best post on any thread to date. You are to be congratulated on cutting through the crap that abounds on FX sites.I've realised that daily charts focusing on price action is the only way to go,forget indicators,predictions, opinions and you might have half a chance at this game.
  • #12
  • Jun 18, 2007 10:11pm Jun 18, 2007 10:11pm
Thanks mongoose for posting your system. It looks very solid.
  • #13
  • Jun 18, 2007 10:15pm Jun 18, 2007 10:15pm
Quoting edge540
Best post on any thread to date. You are to be congratulated on cutting through the crap that abounds on FX sites.I've realised that daily charts focusing on price action is the only way to go,forget indicators,predictions, opinions and you might have half a chance at this game.

You are right. Believe it or not systems like these work better than most intraday systems. I have been trading over a year. I just don't post a lot.
  • #14
  • Jun 18, 2007 10:19pm Jun 18, 2007 10:19pm
"Looks like you have enough rules and coding to actually try and back test the system on tick data. When you do i think you will be disappointed to find it loses money."

Even better, I have been trading it live an posting my account statements on this forum.

"90% of traders lose money"

I would find it unreasonable to put behind me 12 years of successful trading and hang my hat up on a slogan.

"Every instrument you have in your toolbox is a linear math derived from the history but forex is not linear so it it wont work."

FOREX is commonly traded using Fibonacci numbers, interest rate data and other economical data with some success, especially in large commercial organizations. All these are Euclidean Geometry, vector spaces, two- and three-dimensional Euclidean spaces or in other words, linear math!
Of course a systems edge does get diluted and require changing to match the performance of previous years, but the principals of the system remains intact.

In fact even some non-linear subjects are often so complicated that they are modelled by linear approximations as a first step to unravelling their mystery, so it is untrue that a linear system cannot benefit the research on a non-linear phenomenon.

If you back test and optimise past data all you are doing is finding a linear curve fit solution to past data.

I have checks for robustness, which is how this system was created. Checks such as:
*Multiple markets testing (FOREX, futures, stocks)
*Out of sample data testing
*Overstating the size of losses and understating the size of profits
*Bell curves ...See my post http://www.forexfactory.com/showthre...03#post1469003

If you dont believe me, then find me any EA which worked great last year and is still working this year and makes real money live. They are as rare as rocking horse crap.

Too easy:
*Volatility breakout system
*Larry Williams Oops
*Original Turtle Trading System
*My systems such as 6EV, Bollinger Bandit, Boops (my variation of the Oops), Weekly Warrior

The success in these systems lies in letting the winners run, and cutting the losses, and IMO that is the cornerstone of success.

If you’re serious about finding a system you must find a chaotic based probability system

Now you have me interested as I have heard a lot about chaotic based systems, but never really seen one presented. It's like the new chocolate bar on the supermarket shelf; did someone release a book or something? Any further material on this chaos trading system development would be appreciated.


I give you an example. In 5 years worth of euro days 33% will be long over 20 pips a day, 33% will be flat within 20 pips a day and 33% will be short below 20 pips.

I fail to see your point, would you mind elaborating?

5) Every pair is completely different to another there is no such thing as one size fits all.

I understand, however I should mention that I only usually trade the GBPUSD as I am not interested in becoming proficient with too many pairs. I will say that due to the high correlation of pairs, I would expect that most of the majors would exhibit similar characteristics and yield to similar methodologies.

7) The best way to lose money is trade intraday without solid money man and on hunches based on other people advise/thoughts/signals.

I have found that to the contrary, intraday systems just don't add up. This is the point that I totally disagree with you on as I have done exhaustive research on the subject.

Look at this logically:
1. Daily ranges are bigger than intraday ranges = more profit per trade
2. Daily trading = less trades = less spreads to compensate for
3. Interest made on daily positions can overcome the spread
4. Most traders fail + most traders trade intraday therefore: trading intraday is probably inducive to failure


Thanks for taking all that time to reply to my post. As a system developer and tester I am rather interested in the Chaos Trading System ideas so further material would be of great interest to me.
  • #15
  • Edited Jun 19, 2007 12:18am Jun 18, 2007 11:36pm | Edited Jun 19, 2007 12:18am
You will note that as per the system I am now short in the EURUSD.

Whilst not part of the system, I do like to see a lower low and a possible lower high as well as shown on the picture below.

http://img140.imageshack.us/img140/9...dtrade1iu9.jpg

**PLEASE NOTE: THIS IS NOT AN INVITATION TO TRADE! THIS IS PURELY A DISCUSSION OF A TECHNIQUE OF MINE. YOU CAN LOSE SUBSTANTIAL MONEY TRADING FOREX OR ANY OTHER MARKET WITH ANY TECHNIQUE INCLUDING THIS ONE**



edge, I take offense that you refer to my work as "crap". I posted the system, code, charts, trades and so on for constructive criticism. If you didn't like it you always have the option of not returning to this thread.
I also find it contradictory that you scowel at intraday trading, and on the other hand you trade the "Daily Scalping" system on this forum and regularly comment on it.
  • #16
  • Jun 19, 2007 7:24am Jun 19, 2007 7:24am
Quoting Bolt

1) 90% of traders lose money so you are looking for a 10% edge thus you in a bad position before you even start.
We had a thread on this subject just a few days ago.

Where did you gather your statistics that show 90% of traders losing money?
  • #17
  • Jun 19, 2007 9:20am Jun 19, 2007 9:20am
I haven't even began to digest this big sandwich - so I can't comment on it. But I can comment on the needless hard line put forth by the two gents referenced above.

I might learn something from Mongoose - but you two hardliners offer nothing.

To suggest that you can only make money by trading the daily charts is absolutely ludicrous.
  • #18
  • Jun 19, 2007 9:23am Jun 19, 2007 9:23am
Quoting edge540
Best post on any thread to date.
You need to read more threads!
  • #19
  • Jun 19, 2007 8:44pm Jun 19, 2007 8:44pm
Quoting billbss
We had a thread on this subject just a few days ago.

Where did you gather your statistics that show 90% of traders losing money?
You only have to google but the sources are far and wide and point to similiar statistics. I havent seen the other thread but someone must have done the same as me and read it plastered over the net. I been trading full time and make a living doing this for over 5 years now. Im far from rich but i survive. I have written dozens of systems, many EA's and none can really cut the mustard for more then a few weeks/months before they crash and burn. Well not linear MT4 stuff anyway. I now use Genetic methods mainly PSO with neural networks. I look for probabilty models and trade options and futures. I very rarely trade spot any more for many reasons. If you trade options the risk is known from the second you place the trade. The profits are unlimited when your right and capped perfectly when your wrong. There are no issues regarding slippages, bad fills and general small cap retail nasties one sees with spot trading.


A couple here

 

  1. Bruce Kovner discussed trying to train perhaps thirty people, and only four or five turned out to be good traders. The other 25 left the business and according to Dennis "it had nothing to do with intelligence."
  2. Marty Schwartz discussed hiring four people but nobody lasted.
  3. According to Brian Gelber five or less out of every 100 people who go to the floor to become traders make at least a million dollars within five years and at least half will end up losing everything they came in with.
  4. Tom Baldwin responded that less than 20 percent of those who come to trade on the floor are still around after five years and one percent are successful to the point of making and keeping at least a couple of million

An article in Elliot Wave International

"

Close to ninety percent of all traders lose money. We're not just talking about amateurs here, either. Whether you trade for a living or "for fun," chances are, you will not succeed. That's a sad, sobering fact.
However, the remaining ten percent of traders somehow manage to either break even or even make money – and more importantly, do it consistently. How do they do that?
That's an age-old question. Thousands of books have been written and countless seminars and interviews have been conducted in an attempt to answer it. To this day, no magic formula has been found.
Still, we thought we'd take a shot at it. Elliott Wave International's Futures Junctures Service editor Jeffrey Kennedy has identified five fundamental flaws that, in his opinion, stop most traders from being consistently successful. well you can go here to read the rest.

http://www.elliottwave.com/features/...articleid=3129

  • #20
  • Edited 10:30pm Jun 19, 2007 8:44pm | Edited 10:30pm
http://img239.imageshack.us/img239/2...l20juneja4.jpg

Profit this month so far $4391.39

That is roughly a 22% gain on a $20,000 account in 2.4 weeks
0 traders viewing now