Jump to content

DragonPips


Recommended Posts

  • Replies 1.4k
  • Created
  • Last Reply

Top Posters In This Topic

Re: DRAGONPIPS

 

ha ha ha

Monday fomc speaker should be low impact and it was not the case I lost money.

Yesterday it should be medium impact so I didn't trade. It was one of the best trading session for the month for sure. 7000 $ profit on the demo account, zero on the real account.

ha ha ha, that's forex !!

Have a nice trip Chankl78

Link to comment
Share on other sites

Re: DRAGONPIPS

 

Yes, tell me if you find something interesting in London session. For me it's totally impossible to trade this session with peace in mind. But I am not sure that everybody do the same tests. I am afraid that some people do a forward test during only one week, get some few nice trades and think it's ok.

For me a backtest must show good results at least from 8 -9 months ( AND at least 100 trades) , but better from 18 months (300 trades) for a scalper (if the EA only takes 20 tardes a year, must test on 5 years). And a forward test gives some indication of reliability after at least 3 months. Before that (few weeks, 20 or 30 trades), it's totally chance, guess ect...

 

For MD, MdPro, DP, caspian I have done backtests from 1-1-2008 (I worked full time on backtest during the first 6 months of this year). They are on demo from 1 july and it's only in september I start Live.

Link to comment
Share on other sites

Re: DRAGONPIPS

 

Yes, tell me if you find something interesting in London session. For me it's totally impossible to trade this session with peace in mind. But I am not sure that everybody do the same tests. I am afraid that some people do a forward test during only one week, get some few nice trades and think it's ok.

For me a backtest must show good results at least from 8 -9 months ( AND at least 100 trades) , but better from 18 months (300 trades) for a scalper (if the EA only takes 20 tardes a year, must test on 5 years). And a forward test gives some indication of reliability after at least 3 months. Before that (few weeks, 20 or 30 trades), it's totally chance, guess ect...

 

For MD, MdPro, DP, caspian I have done backtests from 1-1-2008 (I worked full time on backtest during the first 6 months of this year). They are on demo from 1 july and it's only in september I start Live.

 

 

 

is been longer than a week, more than a few of us [ read this thread ] are trading london and making money.

 

there could be any reason why some get good results and some don't

 

it could be your broker such as: slippage/requotes/bad data/slow data/higher spreads

your internet speeds are slow

your using different GMT offsets to others

 

 

heres todays trades with london

 

649191 2009.10.23 11:20 sell 0.10 eurusd 1.50468 1.50708 0.00000 2009.10.23 11:24 1.50417 0.00 0.00 0.00 5.50

245786345 DragonPips [s1]

649390 2009.10.23 11:37 sell 0.10 eurusd 1.50468 1.50721 0.00000 2009.10.23 12:11 1.50413 0.00 0.00 0.00 5.94

245786345 DragonPips [s1]

Link to comment
Share on other sites

Re: DRAGONPIPS

 

gomarkets

 

 

SetYourBalanceTargetBelow=- Stop if balance target equal or bigger than --

BalanceTarget=10000000.00000000

za=-- Trading Time (GMT) --------------------------------------------

TradingTime=19:00-23:00 GMT is default

OpenHour=07:00

CloseHour=11:00

xa=-- Money Management ----------------------------------------------

lotsInfo=-- balance under $1000 start with 0.01 Lots

MoneyManagement=1

mmm=Mode: 1-FreeMargin 2-Equity 3-Balance

MM_Mode=1

TradeSizePercent=3.00000000

Lots=0.01000000

MaxLots=10.00000000

zb=----------------------------------------------

Aggressive=1

xb=-- Trade Management ----------------------------------------------

MaxTrades=2

MaxTrades_x_Bar=1

xc=---------------------------

BrokerSpread=-- Set spread not over than 4 --

Spread=2.50000000

TakeProfit=5.00000000

StopLoss=25.00000000

x0=---------------------------

ECN_STP=1

x1=---------------------------

Use_TimeProtectionLevels=0

MinutesToClose_Level_1=60

PipsToClose_Level_1=3.00000000

MinutesToClose_Level_2=140

PipsToClose_Level_2=2.00000000

MinutesToClose_Level_3=170

PipsToClose_Level_3=1.00000000

MinutesToClose_Level_4=200

PipsToClose_Level_4=-1.00000000

MinutesToClose_Level_5=240

PipsToClose_Level_5=-4.00000000

s1=---------------------------

Use_RangeFilter=1

Range_PreviousBars=8

Max_Range=45.00000000

s2=---------------------------

Use_Close_AllTrades=0

Close_AllTrades_At=23:30

s3=---------------------------

Use_Strategy1=1

Use_Strategy2=0

Use_Strategy3=1

Use_Strategy4=0

x2=-- Extreme Edition -------------------------

ATR_Threshold=20

Slippage=1

x3=---------------------------

MagicNumber=1213123131

x4=---------------------------

ExpertInfo=-- Change EA name to hide EA identity -----------------

ExpertName=DragonPips

x90=-- Alerts control ----------------------------------------------

MessageAlert=0

SoundAlert=0

SoundFileAtOpen=alert.wav

SoundFileAtClose=news.wav

x91=-- Manual GMT Offset -------------------------------------------

Use_Manual_GMTOffset=0

Manual_GMTOffset=0

x92=-- Backtests use only ------------------------------------------

GMTOffset_for_Backtests=0

Link to comment
Share on other sites

Re: DRAGONPIPS

 

If a backtest means nothing (it's partially true), it's difficult to forward test during several months hundred of Ea's.

I already saw (many) Ea's that perform well on backtest and are poor on forward test, but I never saw (I will be happy if it's the case) an Ea that is a looser in backtest that become wonderfull in live. I have backtested something like 300 Ea's and throw 290 of them in my (big) garbage. Morover, some EA's that shows wonderfull results this last few months but are terrible before makes me anxious to trade them in real because they will fail back sooner or later. Personnaly Forex is not Poker, it's my income, I trade 100 to 200 lots every night, so I can tell you that I backtest seriously each Ea before put it in live, and forward test during several weeks, and I repeat this process everytime I modify a piece of code.

Backtesting is a big problem with Mt4, but I have a method that can show if an EA is reliable in backtest or not :

- Forwart test on Demo during 1 month and save the report

- Backtest on the same mt4 platform for the same period. Save report

- Compare the 2 reports : if the trades are exactly the same (exactly the same entry ans exits, no more that 1 or 2 minutes difference), I seriouly think the Ea is Backtest reliable. Sometimes it's the case, sometimes it's totally different. That's not bulletproof because history datas can differ from your broker demo, but for sure a fair idea of the EA value. If backtest are reliable at 90 %, it's enought for me, knowing that the same Ea in Live and demo with the same broker can often give different results (and I don't say with different brokers.....). There is a tread about backtesting with 99 % quality, ok it's great, but there is so much difference in Live according to brokers that big surprises remains when go live.

>> Syk1kz : Anyway I will start demo with yours params during London. We will compare results.

Link to comment
Share on other sites

Re: DRAGONPIPS

 

Hmmm...

 

For me, I backtest is to find the best strategy for DP every week.. Good part is there are 4 of them & can use the best for each week. Some strategy is good for last week, might not be good for this week... So far it seems very successful as 90% of the times (Except for the weeks of FOMC) , it is very tedious to test it weekly on the pairs I want to trades.

 

hee hee.. Sorry, from this week onwards might not be able to backtest. Need to go to Japan to bring my wife & son back to singapore.

Also take a short 16 days trip to onsen, eat all the ramen, also the sushi .....

 

WIll try to backtest on sunday .. but maybe no time as i still yet to buy gifts for my parents in law in japan, not yet pack things for this trip..

 

Haha.. anyway, will try my best.

Link to comment
Share on other sites

Re: DRAGONPIPS

 

>> Chankl78

Interesting... but if strategy is always good for last week but might not be good for this week, we have a problem...always a week late...

I prefer not to have change every week and find a more consistent system even if it's not the maximum settings.

Anyway your research about Dp is nice. (I remember we have both wash the code).

Hope you enjoy your travel. Long time I didn't stay to Japan and Singapore. I like this countries...

See you soon on this forum...

Link to comment
Share on other sites

Re: DRAGONPIPS

 

>> Chankl78

Interesting... but if strategy is always good for last week but might not be good for this week, we have a problem...always a week late...

I prefer not to have change every week and find a more consistent system even if it's not the maximum settings.

Anyway your research about Dp is nice. (I remember we have both wash the code).

Hope you enjoy your travel. Long time I didn't stay to Japan and Singapore. I like this countries...

See you soon on this forum...

 

 

i would agree, your a week late and your settings are for the last 5 days trading, not the current 5 days

 

Juinerty your comment above further concretes my point that backtests are crap

 

you say above -> Interesting... but if strategy is always good for last week but might not be good for this week, we have a problem...always a week late...

 

so you say you optimize on 6 months backtests

 

well i say you are 6 months late with your settings, making them even more redundant than Chankl78's

 

i would say Chankl78 strategy is the best, last week is a fairly good indication of CURRENT market conditions, last august does not compare to this week ,why would you optimize your settings from 6 months ago until now? it makes no sense when the markets are choppy and volatile during asian and trending during london.

 

 

next you say -> so I can tell you that I backtest seriously each Ea before put it in live, and forward test during several weeks, and I repeat this process everytime I modify a piece of code.

 

this is the first time i have seen you mention forward tests :) my impression was you backtested for 6 months or whatever, found good results then went live with no other testing. thats why i was :berbusa: :o :o :o

 

anyway i personally thik if you trade forex for a living you should be trading manually not EA

 

but thats my opinion

 

 

*edit*

 

wait on caspian you say you trade 100-200lots daily

 

is this with EA ?

Link to comment
Share on other sites

Re: DRAGONPIPS

 

>> Syk1kz.

I think there is misunderstanding

I am not 6 months late with my parameters because I never try to find the best parameters for an EA. This is curve fitting and it's why so many people fail and do not understand why an Ea works good for few weeks and fail later.

Rather to have to change parameters every weeks that seems to be the best for that week, I try to find parameters that are "all road". Our approach is completly opposite. Obviously I have sometimes to adjust some param because market changes, Starthour and Endhours for example.

My backtest on 18 months do not try to over optimise parameters (anyway on 18 months it's hard), I only want to know intimately the behavior of the EA and adjust parameters to get a smooth curve, not the maximum balance.

This is the case, and the reason why this month I have results with DP that are not so good than you, but I will continue with this parameters next month and next month because I know perfectly the behavior of my EA. (I often know few minutes before DP will open a trade).

>>I always run 3 months minimum on forward demo prior to go live.

>>I trade forex for 6 years and I can tell you that if I play with EA it's because manual trading is too difficult. No time today to develop my opi,ion on manual trading.

>>I play 2 to 8 lots for every trades with MD1.11, MDPro, DP, Caspian and EuropeWaves, so at the end of the night 100-200 lots

Link to comment
Share on other sites

Re: DRAGONPIPS

 

>> Syk1kz.

I think there is misunderstanding

I am not 6 months late with my parameters because I never try to find the best parameters for an EA. This is curve fitting and it's why so many people fail and do not understand why an Ea works good for few weeks and fail later.

Rather to have to change parameters every weeks that seems to be the best for that week, I try to find parameters that are "all road". Our approach is completly opposite. Obviously I have sometimes to adjust some param because market changes, Starthour and Endhours for example.

My backtest on 18 months do not try to over optimise parameters (anyway on 18 months it's hard), I only want to know intimately the behavior of the EA and adjust parameters to get a smooth curve, not the maximum balance.

This is the case, and the reason why this month I have results with DP that are not so good than you, but I will continue with this parameters next month and next month because I know perfectly the behavior of my EA. (I often know few minutes before DP will open a trade).

>>I always run 3 months minimum on forward demo prior to go live.

>>I trade forex for 6 years and I can tell you that if I play with EA it's because manual trading is too difficult. No time today to develop my opi,ion on manual trading.

>>I play 2 to 8 lots for every trades with MD1.11, MDPro, DP, Caspian and EuropeWaves, so at the end of the night 100-200 lots

^:)^ u have bought some great insight into EA trading that I will learn from :shand: =D>

Link to comment
Share on other sites

Re: DRAGONPIPS

 

If a backtest means nothing (it's partially true), it's difficult to forward test during several months hundred of Ea's.

I already saw (many) Ea's that perform well on backtest and are poor on forward test, but I never saw (I will be happy if it's the case) an Ea that is a looser in backtest that become wonderfull in live. I have backtested something like 300 Ea's and throw 290 of them in my (big) garbage. Morover, some EA's that shows wonderfull results this last few months but are terrible before makes me anxious to trade them in real because they will fail back sooner or later. Personnaly Forex is not Poker, it's my income, I trade 100 to 200 lots every night, so I can tell you that I backtest seriously each Ea before put it in live, and forward test during several weeks, and I repeat this process everytime I modify a piece of code.

Backtesting is a big problem with Mt4, but I have a method that can show if an EA is reliable in backtest or not :

- Forwart test on Demo during 1 month and save the report

- Backtest on the same mt4 platform for the same period. Save report

- Compare the 2 reports : if the trades are exactly the same (exactly the same entry ans exits, no more that 1 or 2 minutes difference), I seriouly think the Ea is Backtest reliable. Sometimes it's the case, sometimes it's totally different. That's not bulletproof because history datas can differ from your broker demo, but for sure a fair idea of the EA value. If backtest are reliable at 90 %, it's enought for me, knowing that the same Ea in Live and demo with the same broker can often give different results (and I don't say with different brokers.....). There is a tread about backtesting with 99 % quality, ok it's great, but there is so much difference in Live according to brokers that big surprises remains when go live.

>> Syk1kz : Anyway I will start demo with yours params during London. We will compare results.

 

 

What EA's do you use live?

 

I think backtest and demo test are quite usless. I do backtest and demo only to see if a particulary EA works correctly. Then I go Live with minimum lots' (usually 0.01 )

If it works good for a few days I go to trade bigger lots.

Also I think backtest and optimization is usless most of time. You can optimized an EA for the last 6 month and put then live but tomorrow is another day and marked may change..

Link to comment
Share on other sites

Re: DRAGONPIPS

 

Still trying to find a manual system.. Hahaha..

 

But so far at least DP is profitable as long as I stay away from news.

 

For my backtesting, all the points mentioned above, I never thought of that!!!!

Thanks for enlightening me.

:shand: chankl78 enjoy u holiday & many GREEN PIPS =D>

Link to comment
Share on other sites

Re: DRAGONPIPS

 

It's very important to understand what to do and what don't with optimisation.

I think that most parameters should not be optimised, especially MA RSI MACD etc...periods. I let them fix. If we change every week the MA period, we can get nice results with a simple Ma crossover EA for this week (past), not the next week. A good example of crap EA like this is Forex Funnel.

So I DO NOT agree to reoptimise Ea every weeks (nor every month).

I use Md1.11, MdPro, Dragon since May-June. The initial optimisation done only changed TP, SL, starthour and endhour for every pair. And that's all. In September I changed starthour and endhour because Asian session was more volatile. Nothing more. So I can trust my EA's because I know them perfectly

>> Mograst. Please read the previous posts : I use Md1.11, MdPro, Dragon , Caspian Live

Live return : 15% in september, 6 % in october so far (my fault ,didn't stop monday during FOMC speaking, 6000$ loss).

According to each EA/pair I trade 2 to 8 lots

best performers 8 lots :

MdPRO : EurUSd, UsdChf, UsdJpy, USdCad

Dragon : EurUSd, UsdCHf

Caspian: UsdJpy

For more details read the others post I give my param and results

Link to comment
Share on other sites

Re: DRAGONPIPS

 

It's very important to understand what to do and what don't with optimisation.

I think that most parameters should not be optimised, especially MA RSI MACD etc...periods. I let them fix. If we change every week the MA period, we can get nice results with a simple Ma crossover EA for this week (past), not the next week. A good example of crap EA like this is Forex Funnel.

So I DO NOT agree to reoptimise Ea every weeks (nor every month).

I use Md1.11, MdPro, Dragon since May-June. The initial optimisation done only changed TP, SL, starthour and endhour for every pair. And that's all. In September I changed starthour and endhour because Asian session was more volatile. Nothing more. So I can trust my EA's because I know them perfectly

>> Mograst. Please read the previous posts : I use Md1.11, MdPro, Dragon , Caspian Live

Live return : 15% in september, 6 % in october so far (my fault ,didn't stop monday during FOMC speaking, 6000$ loss).

According to each EA/pair I trade 2 to 8 lots

best performers 8 lots :

MdPRO : EurUSd, UsdChf, UsdJpy, USdCad

Dragon : EurUSd, UsdCHf

Caspian: UsdJpy

For more details read the others post I give my param and results

Thank you very much for your sharing invaluable trading ideas!

 

I am backtesting Caspian on USDJPY and Dragon on EURUSD & EURCHF for the past 18-month period, with trading time 19:00 - 22:00 GMT. However, the results are rather disappointing. Am I missing something here? Or else, is there more to it than simply running EAs 24/5, i.e., turning off EAs on big news days, etc?

Link to comment
Share on other sites

Re: DRAGONPIPS

 

It's very important to understand what to do and what don't with optimisation.

I think that most parameters should not be optimised, especially MA RSI MACD etc...periods. I let them fix. If we change every week the MA period, we can get nice results with a simple Ma crossover EA for this week (past), not the next week. A good example of crap EA like this is Forex Funnel.

So I DO NOT agree to reoptimise Ea every weeks (nor every month).

I use Md1.11, MdPro, Dragon since May-June. The initial optimisation done only changed TP, SL, starthour and endhour for every pair. And that's all. In September I changed starthour and endhour because Asian session was more volatile. Nothing more. So I can trust my EA's because I know them perfectly

>> Mograst. Please read the previous posts : I use Md1.11, MdPro, Dragon , Caspian Live

Live return : 15% in september, 6 % in october so far (my fault ,didn't stop monday during FOMC speaking, 6000$ loss).

According to each EA/pair I trade 2 to 8 lots

best performers 8 lots :

MdPRO : EurUSd, UsdChf, UsdJpy, USdCad

Dragon : EurUSd, UsdCHf

Caspian: UsdJpy

For more details read the others post I give my param and results

 

 

i would agree with 90% of this post. we obviously have our different opinions and we are both making money so we are both correct.

 

i am surprised you run caspian on usdjpy. is there such as EA or are you using eurchf on usdjpy or eurgbp or usdjpy?

 

do you have a mt4stats / myfxbook available?

Link to comment
Share on other sites

Re: DRAGONPIPS

 

hmm...I tried demo in The collective Fx broker platform, no trades at all.

But it can be backtested.

Anybody knows why ??

 

:-? :-?

 

I am having the same issue with "The Collective FX", nither dragonpips nithere MD1.2 are trading in the same way as on other brokers. I am not sure why, spread is good but for some reasons eas do not trade!

 

Please anyone more expirenced to share opinion!

 

Cheers

Kudus is what gives me power....
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.




×
×
  • Create New...