Hopes and Dreams shattered...
:) Well, you are right in the sense that no mechanical trading system can consistently
make money over the long run, however, there are systems that provide you with
sound clues on the market...making the final call on the trade (in terms of position
size and entry/exit decision) is where human expertise comes in...
Use your EA to provide you 'advice', not make 'decisions'...you'll find many a good EAs that help improve your trading performance...good luck! :)
Use your EA to provide you 'advice', not make 'decisions'...you'll find many a good EAs that help improve your trading performance...good luck! :)
Thanks for taking the time to answer. Consider this: If no system can be programmed
to trade the markets profitably over the long term, then I dare to say that trading
boils down to gambling. You have to try your "luck" every time you enter
into a trade, and hope you were right.
I know that you can develop your skills, but at the end of the day your "system" should be programmable, otherwise it boils down to gambling with your luck.
Good Luck!
I know that you can develop your skills, but at the end of the day your "system" should be programmable, otherwise it boils down to gambling with your luck.
Good Luck!
I don't agree with Maratha at all. Of course, it should be possible, Benna.
A simple example is if a system will do following - buy at the 1st of the month and sell 2 days later. Provided that you have tested this strategy on historical data, and can convince yourself that it works. Then, Maratha, you don't need "advice" and then make a "decision" - it is a mechanical system.
So, Benna, keep looking for a good, historically verified system and you'll be on a way to your dreams.
Good luck!
A simple example is if a system will do following - buy at the 1st of the month and sell 2 days later. Provided that you have tested this strategy on historical data, and can convince yourself that it works. Then, Maratha, you don't need "advice" and then make a "decision" - it is a mechanical system.
So, Benna, keep looking for a good, historically verified system and you'll be on a way to your dreams.
Good luck!
Do you have a manual system that you trade that profits consistently? That would
possibly show you if you can trade profitably at all. Automating what you do is
not always easy, but at least there is potential then. Expecting a magical EA to
do automatically what you don't already do is a bit presumptuous. There's also
a difference between consistently profiting, and profiting as big as you'd like
(or as big as your manual system). Consistency for something running on it's own
sounds pretty good to me, even if it's small profit. One last flame - as much work
as people put in on forums for developing strategies, and the flak the originators
take in that process would probably convince most anyone to withold successful
EAs from so many that would take without contributing.
Ququr,
you seem to know a lot about MQL4. Could you do me a favor and take a look into my post "Please HELP with EA"? I am trying to explain an issue to MQL4 guys, but they seem not getting it. Am I not clear in my post, or ...?
thanks.
you seem to know a lot about MQL4. Could you do me a favor and take a look into my post "Please HELP with EA"? I am trying to explain an issue to MQL4 guys, but they seem not getting it. Am I not clear in my post, or ...?
thanks.
Benna, Even if I'm a newbie in this automated trading stuff, i know about discretionary
trading... that's far worse. Automated trading is the best solution possible. I
think you probably forced the EA parameters too much. I became convinced about
automated trading after reading a book written by a guy who dissavous automation
(Dirk du Toit - "Bird Watching in Lion Country"). The mistake lies in
the bad trading principles. Just make a normal EA to take regular signals, like
MA crosses (the simpler, the better) with different periods on different timeframes
with a 1-2 leverage/trade (i'm not sure about this, i'm working on it now). Losses
made by false positives on longer timeframes are recovered by profits on shorter
timeframes and vice versa. It should be something to work... just keep the positions
small and remember not to be fooled by randomness... Another problem is that MT
automation is not full, but rather incomplete, since it's not server based. The
PC from home can't run MT 24/24 - and that's the last requirement of the automation,
being the moment when all signals can be taken - longer frames and shorter frames.
So the next step after making and testing a succesful EA would be to rent a server
from a datacenter and put MT on it.
I've been attempting to automate my trading for a while... Thus far I have been
able to successfully automate my exit strategies. I still manually choose when to get in. Yes, the exit strategy is much easier.
I am now attempting to put together a formula for my entry strategies. Unfortunately,
my entry strategy is based largely on an external program (4XMadeEasy) and I have no idea what specific calculations they are doing to come up with their
buying and selling pressure lines. I'm working on something similar. At any rate,
I believe you CAN create an automated trading system, but that it will never be as effective as a live human trader. That being said,
I'm going to keep working on my entry strategy scripts and see if I can come up
with something that works. At least with an exit strategy if I decide to sleep once in a while I don't miss the take profit moment! =) Good
luck!
Lance...
Lance...
Yes Yes, automated trading is the only way to go... haven't you stayed up all night playing, guessing
and losing, then take a nap and find you missed the 100 pip run an hour after you
left the screen??? this automation thingy stays at the monitor all day and night.
. You can make the program so tight it will only jump on (trade) one trade a week
or month or day...and be successful. I believe that following two or three indicators
that you would normally use to make manual trades.. (and you should) and plugging
them in with a little "fudge factor", set your stop-loss and take profit
correctly, you will come out ahead...Go ahead and watch your expert, would you
have sold there? Would you have bought there? would you have gotten out there?.
. test, watch, redo, reset your program settings, (how far does the ... I donno.
.. perhaps the momentum change before you buy... 5? 10? how much does it have to
change to close the sell? 2? 8? tweeking the program by observation is the only
way I would go..
I believe the main thing is just like in manual trading... stay out of the ranging market.... don't guess the top or bottom and never double up to catch up... automated trading does not "guess".. they are not emotional... these are the things that made me lose money... all the time... ah I have rambled on too long... you get the idea..
I believe the main thing is just like in manual trading... stay out of the ranging market.... don't guess the top or bottom and never double up to catch up... automated trading does not "guess".. they are not emotional... these are the things that made me lose money... all the time... ah I have rambled on too long... you get the idea..
I believe OldMan is basically right - but there's no way (and not even recommended)
to stay out of ranging markets. For example not taking a position when distance
between Bollinger Bands is under "n" pips. That's a mistake, it's precisely the period announcing
a big market movement , just the Bollinger Bands don't tell you which way. It would be better to have multiple small positions which
hedge themselves during ranging markets (notice the moving averages - half going
up and half going down) and restart making money by increasing position as the
markets become more and more trendy. Now I don't agree the main thing is the same as manual trading - even if I don't
trade at all at this moment - there is the trader's Hell: market excitement , guessing,
leverage and breaking the rules resulting in heavy losses. The dangerous thing
about automated trading is - beyond fears about power outage , computer crashes or internet connection
being shut - and this thing needs to be discussed - The MYTH that after a while,
markets "desyncronize" with the trading strategy (urging the completion
of a brand new one) whatever being that strategy...
Please don't be discouraged. You can indeed profit from autotrading. But you are also partially correct in that there is no ONE ea that will always
make money. ALL strategies/systems draw down. ALL do. It's a fact of life.
It's a gamble whether you begin trading your new system at the beginning of a drawdown, or at the beginning of an upswing.
The answer you seek is deep-rooted in statistics. A friend of mine (also a system-trader) told me recently, he no longer takes "trades", they are statistical events.
I also recently spoke with another friend. She consults with the top traders in the country. She mentioned that almost ALL of them are statistical system traders, and not discretionary traders. When you speak with them, all they say are numbers. "This trade has a 37% probability of hitting target, and a 15.5% chance of hitting stop, but a 57% chance of catching profit with a trailing stop, but she needs to breathe at least 15% drawdown before showing that."
There is another 40 year trading veteran. He has a few favorite strategies (that are also listed in the link below). But do you think he takes every single trade? Nope. He watches the statistics. He paper trades the strategies. When one signals it is starting to become profitable, he begins trading it. When the statistics start to go sour, he STOPS. By then another strategy is beginning to kick in. He uses a very powerful strategy management system. Look for references to GB007 in the link below. Spreadsheets included.
You have to think the same way. For each strategy (ea) you are considering, you must be intimately familiar with the historical statistical behaviour of it. What are the largest historical drawdowns. This will clue you in to when your strategy isn't performing as historically predicted. Then you can turn it off. Can you anticipate drawdowns by putting a simple moving average on your equity curve? Does the average trade give you a positive expectations, albeit small?
MetaTrader is primarily an execution tool. What you need is a strategy development/statistical analysis tool. I recommend Tradestation for your development, and idea exploration. It is mature, and does backtesting incredibly well. There are also lots of third-party analysis tools. But when it comes to execution, leave tradestation, and move to MetaTrader for sure. But, you will come to metatrader with an EA that you know, and intimately understand. YOu know how it should behave, and when it's misbehaving.
Start here:
http://www.woodiescciclub.com/start.htm
Look for Clint_TX's talks on system development and statistics gathering. He put together some incredible, incredible tools for finding your EA's. Don't skip this step. All will be meaningless without it. I promise. Do THIS first. It's all free, honest and heartfelt from Clint.
Note that I said "finding your EA's". As in plural. Becuase the secret lies in the plural. When you have more than one strategy, they begin to complement eachother. There is no ONE ea that can always make money by itself, but a few strong EA's combined. Well, now you're on the path.
Becoming a system trader is indeed a worthy path to take, but there is a great deal of learning and sophistication involved. Stay on the path, just realize that you need to take the next learning step on that path.
Enjoy!
PS.
Try also googling:
equity curve trading
kelly money management
genetic optimization tradestation
It's a gamble whether you begin trading your new system at the beginning of a drawdown, or at the beginning of an upswing.
The answer you seek is deep-rooted in statistics. A friend of mine (also a system-trader) told me recently, he no longer takes "trades", they are statistical events.
I also recently spoke with another friend. She consults with the top traders in the country. She mentioned that almost ALL of them are statistical system traders, and not discretionary traders. When you speak with them, all they say are numbers. "This trade has a 37% probability of hitting target, and a 15.5% chance of hitting stop, but a 57% chance of catching profit with a trailing stop, but she needs to breathe at least 15% drawdown before showing that."
There is another 40 year trading veteran. He has a few favorite strategies (that are also listed in the link below). But do you think he takes every single trade? Nope. He watches the statistics. He paper trades the strategies. When one signals it is starting to become profitable, he begins trading it. When the statistics start to go sour, he STOPS. By then another strategy is beginning to kick in. He uses a very powerful strategy management system. Look for references to GB007 in the link below. Spreadsheets included.
You have to think the same way. For each strategy (ea) you are considering, you must be intimately familiar with the historical statistical behaviour of it. What are the largest historical drawdowns. This will clue you in to when your strategy isn't performing as historically predicted. Then you can turn it off. Can you anticipate drawdowns by putting a simple moving average on your equity curve? Does the average trade give you a positive expectations, albeit small?
MetaTrader is primarily an execution tool. What you need is a strategy development/statistical analysis tool. I recommend Tradestation for your development, and idea exploration. It is mature, and does backtesting incredibly well. There are also lots of third-party analysis tools. But when it comes to execution, leave tradestation, and move to MetaTrader for sure. But, you will come to metatrader with an EA that you know, and intimately understand. YOu know how it should behave, and when it's misbehaving.
Start here:
http://www.woodiescciclub.com/start.htm
Look for Clint_TX's talks on system development and statistics gathering. He put together some incredible, incredible tools for finding your EA's. Don't skip this step. All will be meaningless without it. I promise. Do THIS first. It's all free, honest and heartfelt from Clint.
Note that I said "finding your EA's". As in plural. Becuase the secret lies in the plural. When you have more than one strategy, they begin to complement eachother. There is no ONE ea that can always make money by itself, but a few strong EA's combined. Well, now you're on the path.
Becoming a system trader is indeed a worthy path to take, but there is a great deal of learning and sophistication involved. Stay on the path, just realize that you need to take the next learning step on that path.
Enjoy!
PS.
Try also googling:
equity curve trading
kelly money management
genetic optimization tradestation
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
After a search of many, many months I came to the conclusion that there is no Expert Advisor that can actually consistently make money over the long term. Can anybody prove me wrong?