My demo journal | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

How to open demo account with Alpari-demo? 3 replies

GCash Demo Journal 53 replies

Demo Journal 11 replies

Any MIG users? Different candle close times between MIG demo and MT4 demo? 9 replies

IBFX changed their demo policy, where can I get an unlimited demo? 9 replies

Attachments: My demo journal
Exit Attachments

My demo journal

  • #1
  • First Post: Oct 18, 2011 8:17am Oct 18, 2011 8:17am
Many people around the forex world think demo trading is pointless, it doesn't represent or give you any clear idea of what can be expected once you go live with the strategy on a live account.

Demos are given a bad name because it's the dreaded "fake money". However, an EA system requires constant work and modification from time to time. I run these variation demos along side my live account. I am a live trader and have been for many years now. (I am profitable, but I choose not to divulge live trading stats for obvious reasons.)

My EA system can not be backtested so any change I seem fit to make can have good effects or not so good effects. This thread is about showing demo results of slow and steady profits.

It is not meant to be a get rich quick scheme. You treat the demo as real money. Because we treat it like real money it gives clear and very similar results to how a real live money account will perform.

Any discussions here should be helpful and considerate of other traders.
  • #2
  • Oct 18, 2011 8:52am Oct 18, 2011 8:52am
A few stats to get things started. Just to note this particular demo was started Jan 17, 2011. While I do intervene from time to time on my live account this mostly amounts to changing variables and or shutting some functions off.

95% of the time the EA runs the show. The demo is let to run "wild". I want to see the worst effect of any action as this gives me cues to what might or might not need adjustment on the live acct.

I have multiple variations of this running as I am always trying to come up with more ways to either increase profitability while reducing risk if possible. Some of these other variations don't always work out so well...

Attached Image(s) (click to enlarge)
Click to Enlarge Name: PA-6-16-11.jpg Size: 100 KB
Click to Enlarge Name: PA-8-19-11.jpg Size: 117 KB
  • #3
  • Oct 18, 2011 8:55am Oct 18, 2011 8:55am
Ok, this post should bring everything up to date as of October 14, 2011 just a shade under 10 month worth of results off the same demo account.

Attached Image(s) (click to enlarge)
Click to Enlarge Name: PA-9-17-11.jpg Size: 118 KB
Click to Enlarge Name: PA-10-14-11.jpg Size: 121 KB
  • #4
  • Oct 18, 2011 9:30am Oct 18, 2011 9:30am
My question for you would be this. Basically all indicators are based on moving averages with some unique math formula applied to them to get their chart on the screen or a separate window (MACD, STO, Oscillators, ETC) or they are based on some kind of high low formula (pivots, bollinger band, fibs, candles etc). Have you found another type of formula out side these major two in your EAs?
  • #5
  • Oct 18, 2011 10:04am Oct 18, 2011 10:04am
Quoting nanningbob
My question for you would be this. Basically all indicators are based on moving averages with some unique math formula applied to them to get their chart on the screen or a separate window (MACD, STO, Oscillators, ETC) or they are based on some kind of high low formula (pivots, bollinger band, fibs, candles etc). Have you found another type of formula out side these major two in your EAs?

Yes, exactly. The biggest problem or "hurdle" if you will is that an EA is an exact science applied to a in-exact science that being the FX market itself.

When I say "in-exact" or more or less the opposite of exact is that an EA is most commonly coded as a "do or die" only scenario or mentality which is flawed for obvious reasons.

You get plagued by the game of inches... you are either over or under. What I have found is not perfect... you take a little or give up alot... you try to take too much and other times you come up short... the problem more specifically is "Rigidity".

An EA can not think for itself perse, you have to tell it what to do and being the obedient servant it is... it then does what it is told to do.

So how do you get rid of the Rigidity you might ask? Simple you give it some more reasoning to make a decision.

You come up with terms or definitions of what you could call a "trend friend" - meaning the fx pair is trending in the direction of your trades. Realize even this won't be perfect everytime... but you are not looking for perfection in an imperfect world if you are then you're in the wrong place, because you will still have losses don't kid yourself if think you won't have any losses -- all that matters in this game is that you win more than you lose month in and month out anyone to tell you any different is a snake oil salesman.

The biggest problem in FX is that the range size is never the same. This is where most EA's fail over the long term. Because ranges of a particular size can exist for several months then they either shrink or expand... catch you and the EA off guard and then all hope is lost the EA discarded as trash or whatever. No one attempts to keep working at the problem let alone try to even understand it fully.

(However for others that may be reading this, there are more problems than just the rigidity problem... so don't think that you solve just the rigidity problem and fail to address the others.... to give you an idea of the complexity involved my EA is well over 2000 lines worth of code... and while that beast (the code itself) likes to think he is all knowing... he is not I can assure you. LOL)

So getting back to the Rigidity problem...

build a multiple profit scenario.... don't always settle for 50 pips... don't be too greedy and never settle for 10 pips.

Set up your rules as to why you are settling for 10 pips as opposed to why you won't hold out for 50 or more. Every time these kinds of decisions should be about profits or the growing of equity.

I think my best ever profit that I show on the demo was over $200 at 50 cents a pip you can do the math on that one... and you can clearly see my avg profit is much less than that. That is how you defeat rigidity.

However, sometimes the decision you give the EA might be wrong. Just realize you won't capture every move... some will get away. I think this should be enough to give you some ideas.
  • #6
  • Oct 18, 2011 3:55pm Oct 18, 2011 3:55pm
Quoting pip_seeker
Yes, exactly. The biggest problem or "hurdle" if you will is that an EA is an exact science applied to a in-exact science that being the FX market itself.

When I say "in-exact" or more or less the opposite of exact is that an EA is most commonly coded as a "do or die" only scenario or mentality which is flawed for obvious reasons.

[size=3]You get plagued by the game of inches... you are either over or under. What I have found is not perfect... you take a little or give up...
If you get a chance read my 10.0 theory and see how that fits your theories. I also have done range studies and this is what I came up with.

http://www.forexfactory.com/showthread.php?t=308304
Attached File(s)
File Type: xls RANGES OF PAIRS.xls   35 KB | 249 downloads
  • #7
  • Oct 18, 2011 6:53pm Oct 18, 2011 6:53pm
Quoting nanningbob
If you get a chance read my 10.0 theory and see how that fits your theories. I also have done range studies and this is what I came up with.

http://www.forexfactory.com/showthread.php?t=308304

Well I don't know that I want to turn this thread into theory discussion for the simple reason theories are never ending always waiting to be proven wrong.

The market changes (range sizes) constantly. Volatility as you already demonstrated in the excel file changes as well.
We already know this by easily using an ATR indicator. This isn't meant to pick on you so please don't take it that way. OK?

I really do nothing more than play the ODDS at opportune times. That in essence is my theory. (opportunity knocks)

So, I think this should be more about coming up with additional ways to make EA's more rational. It sounds like you have a trading strategy that works already but you don't trust the EA as much as you could or something like that is what I remember reading about it.

Alot of this discussion should be focused more on ways to know when to take profits... but much further than just saying oh I want 30 pips or I want 10 pips.

Why do you close for 30 pips? or only settle for 10? List the reasons... Some reasons may be chart specific or timeframe specific... just realize (if you are new to fx) the terms you use on one chart can be much different on another.

Because I want to get something out of the thread too... something I may have overlooked or a different idea to get me to think of something else.

If your strategy works this is about looking for ways to make it even better. Give the EA some more conditions to ponder if you will. That way everyone gets something out of the thread... because in the end what ever idea you have or system, strategy or whatever you have to be able to run it when things go wrong...



  • #8
  • Oct 18, 2011 7:11pm Oct 18, 2011 7:11pm
Quoting pip_seeker
Well I don't know that I want to turn this thread into theory discussion for the simple reason theories are never ending always waiting to be proven wrong.

The market changes (range sizes) constantly. Volatility as you already demonstrated in the excel file changes as well.
We already know this by easily using an ATR indicator. This isn't meant to pick on you so please don't take it that way. OK?

[size=3]I really do nothing more than play the ODDS at opportune times. That in essence is my theory....
There are several topics here for discussion. I wanted you to see 10.0 and compare it to what you are thinking to see how close we think. The Range excell is based on two things Oct 2008 was a 1 time event therefore I used a 2nd basis for range trading and that was ranges since then. If that is the most a currency will move until the next super story then we can range trade safely in those areas. That is the thinking behind the excel. From that info I rated the different currencies according to their ranges and from that I chose pairs to range trade, counter trade and trend or BO trade. Obviously BO trades work better on pairs that are more volatile and range trading with low volatile pairs. So that is what I use to trade those pairs. Now you talk about range trading with and EA and I would point out that the lower volatile pairs would be better to range trade with and EA. That is my point on that area.
  • #9
  • Oct 18, 2011 7:21pm Oct 18, 2011 7:21pm
I dont set my EAs to TP except my 4H EAs I use to range trade. They go from top of a MA high to the bottom of a MA low. All the rest I tp manually or set a TP near S/R line. How do I TP I usually go by the session.

JPY session 20-40 pips I grab because price will usually retrace back before the Euro session.

Eur session I will also grab profits unless I know a fundamental is driving the markets.

USA session I like to grab profits 3-4 hours in or let them run until the end. So I dont have a set pip level but will try to catch profit at S/R lines during a session. I will then reset my EA to reenter if the S/R line is broken and let it run.
  • #10
  • Oct 18, 2011 8:40pm Oct 18, 2011 8:40pm
Quoting nanningbob
There are several topics here for discussion. I wanted you to see 10.0 and compare it to what you are thinking to see how close we think. The Range excell is based on two things Oct 2008 was a 1 time event therefore I used a 2nd basis for range trading and that was ranges since then. If that is the most a currency will move until the next super story then we can range trade safely in those areas. That is the thinking behind the excel.


Ok the only problem is you never get much if any warning of a new super event... it's always hindsight which can be catastrophic. I know because many years ago I did quite a bit of looking for these "super events" as you call them. I've found that you can never really prepare for them you are likely always caught in the middle of them.


Quoting nanningbob
From that info I rated the different currencies according to their ranges and from that I chose pairs to range trade, counter trade and trend or BO trade. Obviously BO trades work better on pairs that are more volatile and range trading with low volatile pairs. So that is what I use to trade those pairs. Now you talk about range trading with and EA and I would point out that the lower volatile pairs would be better to range trade with and EA. That is my point on that area.
While it's true certain pairs are more likely candidates for range trading just as much as certain pair are more geared toward BO trading... but here in lies the rub... what is range and what is not. Because range to me is likely not the same to you and so on.

You could have range on any chart and those ranges can be quite big so big in fact you'll get some people saying "If I didn't know this was range trading I swear it was trending."

Because if anything changes in FX frequently it's the size of the range err trend. Whichever you prefer... that's why I didn't give a description of what "trend" or "range" is to me personally because they mean different things to different people and the strategy they employ.

But when I get some free "reading" time I'll try to read some more of that thread and see if anything stands out to comment on.

  • #11
  • Oct 18, 2011 8:51pm Oct 18, 2011 8:51pm
Quoting pip_seeker
[/size]

Ok the only problem is you never get much if any warning of a new super event... it's always hindsight which can be catastrophic. I know because many years ago I did quite a bit of looking for these "super events" as you call them. I've found that you can never really prepare for them you are likely always caught in the middle of them.


[size=3]

[size=3]While it's true certain pairs are more likely candidates for range trading just as much as certain pair are more geared toward BO trading... but here in lies the...
That is what I attempted in 10.0 to give definitions based on a daily chart. Your definitions obviously would change based on what chart you use. I agree. I tend to think of the market as ranging with spurts to a new ranging area. In other words there is a high or low PA acts for a season then it spurts to a new session of a high and low. Survive those spurts and range trading is very profitable.
  • #12
  • Oct 18, 2011 9:21pm Oct 18, 2011 9:21pm
Quoting nanningbob
That is what I attempted in 10.0 to give definitions based on a daily chart. Your definitions obviously would change based on what chart you use. I agree. I tend to think of the market as ranging with spurts to a new ranging area. In other words there is a high or low PA acts for a season then it spurts to a new session of a high and low. Survive those spurts and range trading is very profitable.

Yeah I agree with that... although there is always the probability of getting suckered in situations like SNB / BOJ intervention and in some cases heaven forbid joint CB intervention.

But there is nothing more profitable than range trade IMHO, provided you avoid the obvious.
  • #13
  • Oct 18, 2011 9:50pm Oct 18, 2011 9:50pm
Quoting pip_seeker
Yeah I agree with that... although there is always the probability of getting suckered in situations like SNB / BOJ intervention and in some cases heaven forbid joint CB intervention.

But there is nothing more profitable than range trade IMHO, provided you avoid the obvious.
Yup I agree the market ranges more than trend and most trends are short lived. Some are medium and can be handled and a couple of times a year you have a real challenge. Ok we should have some good discussions, hopefully you may have an angle I dont know.
  • #14
  • Oct 19, 2011 6:32am Oct 19, 2011 6:32am
Here is the start of a new idea I have been tossing around for awhile that I just started placing some manual trades... because I already have my base system coded and the code is very long I tend to code in modules.

This module in particular will eventually become apart of the existing code base. I never back test any more except to verify that the code is operating as it should, error and bug checking etc. So the manual part of this is to get a glimpse of how profitable or not the new addition could be.

Kinda like a "toe in the water" before I invest all the time and energy of developing and writing the code. I develop the code in such a way that is reminiscent of an onion... one layer stacked upon another.

This adds sophistication to the code among other things that can work to your advantage more often than not. (except for finding bugs within the code.)

With that said, I have kept lot size very small with this module... no trickery like martingale stuff every trade is 1 micro lot or 10 cents a pip. Currently up over $100 in profits at 10 cents a pip you can easily figure the pips gained on it.

Once I start to wrap up coding of this particular idea it will be a system on it's own initially so as to compare returns against the current system and eventually against it's bigger brother (Current base + Drive module) . In spelling this out it's to give you an idea of my work flow and lend you some possible ideas on how to construct an EA for yourself... the right way. (not trying to cheat, looking to trade automatically!)

IMHO, Preparation is the key to success. Overlooking a few things can cause massive differences in levels of success. Hopefully you can take something positive away from this.

Attached is a snapshot of initial manual test results.


Attached Image (click to enlarge)
Click to Enlarge Name: Drive_Module_10-19-11.jpg Size: 86 KB
  • #15
  • Oct 19, 2011 7:12am Oct 19, 2011 7:12am
Not sure what you are doing but the results look good. I have the same idea. I backtest only to see if it works, I dont trust the info in back testers, I have them very unreliable. Then I trade live instead of demo but at a penny a pip not much risk there. The one thing a demo does not give that a live will give is the funny stuff brokers do to PA, like spread widening near TP and SL parts. One of the main reasons my SL are not on the screen. I promise I will stay away from that thread here. LOL So I have an EA that has lots of coding but I keep it basically simple and try to use EAs in markets i think they will do best in.
  • #16
  • Oct 19, 2011 5:06pm Oct 19, 2011 5:06pm
Quoting nanningbob
Not sure what you are doing but the results look good. I have the same idea. I backtest only to see if it works, I dont trust the info in back testers, I have them very unreliable. Then I trade live instead of demo but at a penny a pip not much risk there. The one thing a demo does not give that a live will give is the funny stuff brokers do to PA, like spread widening near TP and SL parts. One of the main reasons my SL are not on the screen. I promise I will stay away from that thread here. LOL So I have an EA that has lots of coding but I keep...

Spread widening can be dealt with because it doesn't last that long... the EA has all the time in the world and I have found that just waiting I end up making much more... as price will move further in my direction and spread shrinks. Patience is just as important as entry & exit sometimes.

I haven't had problems like this in a long time...probably cause the brokers know I will just simply program around their trickery. I been doing this too long to deal with that non-sense and typically the only time you get jacked like that is if you aren't holding trades that long.

Obviously by the demo stats shown I can be in trades for hours and hours sometimes... since going to this system I have found less and less BS type tactics from the brokers... little random short term movements don't really give me that many concerns.

Every once in awhile I've taken a slight hit due to a high volatile close, but more often than not the live outperforms the demo in relation to lots traded.... so I can't complain too much in that regard right now. Knock on wood.
  • #17
  • Oct 22, 2011 9:48am Oct 22, 2011 9:48am
Spread widening only concerns me when it is time for TP. That is the only time it aggravates me. I often post my TP far from where I want to take it so I dont have to deal with it. Most of my trades are closed manually during lulls in the market. I see you are still in there in the SL thread. Dont see much use in posting there anymore. It is more a do things our way or get lost dummy thread. I did pick up one new idea there though and that was the ATF MA. Works kind of cool. Do you know how you would code that. I can see the straight line it causes on a lower TF and would be interested in how to code that.
  • #18
  • Oct 22, 2011 9:04pm Oct 22, 2011 9:04pm
ATF MA?

No can't say that I saw that... what does it stand for?

All time frame moving average? lol.

I don't know I'm only guessing. I can't say that I need another indy right now but I might take a look at it when I got some spare time.

Yeah I probably waste too much time in that SL -No SL thread. It's worse than crack cocaine I can't stay away from it. Probably the only way to break this habit is the mods come in an delete posts to break up the next "punked fight".

Kinda surprised that it hasn't happened yet.
  • #19
  • Oct 22, 2011 9:18pm Oct 22, 2011 9:18pm
Quoting pip_seeker
ATF MA?

No can't say that I saw that... what does it stand for?

All time frame moving average? lol.

I don't know I'm only guessing. I can't say that I need another indy right now but I might take a look at it when I got some spare time.

Yeah I probably waste too much time in that SL -No SL thread. It's worse than crack cocaine I can't stay away from it. Probably the only way to break this habit is the mods come in an delete posts to break up the next "punked fight". ...
Oh there were a lot of posts deleted in that thread. It was really really nasty there for a while. I finally quit posting there, in my opinion, the other side had nothing to post but criticism. In other words, except for the ATF-MA (all time frame-moving average) it was just a bunch of people butting heads. I dont mind a good debate but these guys respect no one except their own little narrow definitions. If you want to see the ATF-MA in action you can go to

http://www.forexfactory.com/showthre...79#post5063679

Pages 31 and 32 show how I have started to use it. Basically it plots the MA of a TF on every TF. Using a Daily MA on a lower chart makes it a nice straight line all day to mark your trades from. So on a 15M chart the 2MA daily will be a straight line across the screen all day. I think it has some use but it really helps in conjunction with a S/R system. Anyway you can look.

Now if you like getting hit on the head all day just go back to the SL thread. LOL
  • #20
  • Oct 23, 2011 12:03am Oct 23, 2011 12:03am
LOL... yeah ok I just kinda guessed at that name. I can't believe I actually got it right.

I'll have to think about it if I can use it or not I'm not sure.

I've been working on this new code and finally was able to pull off a reasonable working back test... but this is just a functional test and a peek at possible performance. (never trust a back test folks. Even forward test on demo can give mixed results sometimes depending on the abilities of the programmer.)

I can tell you running this bad boy should be really interesting. I hope to get it ready to run sometime this week --- hopefully tomorrow in some form or another.

If demo forward gives me half what backtest shows it should be phenominal.

Here's a little peek.... (test was run from Sept 12, 2011 to Oct 12,2011 on GU -- Lotsize was 1 mini lot)
Attached Image (click to enlarge)
Click to Enlarge Name: Drive_backtest-10-25-11.jpg Size: 62 KB
0 traders viewing now