Discuss ForexRobotTrader.com

[jackdan]

In my reviewing of the past few weeks of trading it is evident that this system passed up major trading oportunities while it was holding losing trades. This is the result of a robot that has been optimized for idiots who want a hands off approach. Unfortunately, trading is a business that generally requires participation, it is possible to systematize it, but then you will have to take your losses with your gains, a program can not adapt and think as you or I can. It can only follow its protocol.

I think Don has the best robot on the market, that I have seen. Considering the price, I think it is a fantastic piece of equipment. It has my full interest and attention, and criticisms.

Thank You

Hello,



i would like to discuss with you the setting of this ea. Have you bought it at the end?

I am more than ready to share and discuss these settings here, or privately, my Email is spacanarias at terra dot es.

I am more than convinced that this ea is the very good of it all.


Regards


jackdan

 

[Ken Long]

jackdan,

I have a copy of the Steinitz robot, its not licensable, it was given to me for testing purposes, but it works fine on demo, and if I can ever manage to control its trading better, with calculated exits/losses, and more precise entries, I would be glad to buy it.

Its a fabulous program for its transparency and user adjustable parameters. All its basic theory, of trends and retracements is very sound. Its MTF Heiken Ashi indicator is great to use to find and confirm manual trading oportunities. But the actual trading of the robot leaves a lot to be desired.

I found it very hard to manage this robot sucessfully. It can open many trades when it gets the signals, and produce enormous profits, initialy. But every trend it finds ends with a final trade that goes into drawdown when the trend changes. As long as there are layers of sucessive trending support in the upper time frames it can simply hold on the eventual continuation, but there is no telling in advance if the upper trends will hold, or how much, or how long, of a drawdown needs to be allowed for.

Without any workable loss controls, other than exiting all trades at a specified time, this robot can and generally does accumulate massive drawdowns. The current volitility in the markets and lack of any continuation in the trend is the overall cause of the problems, but even in a well trending market this robot is designed to hold onto massive drawdowns, all in pursuit of that ellusive no loss strategy.

The problems:

1. The Heiken Ashi Smoothed MTF indicator is nothing more than a series of colorful smoothed moving averages. You can do the same thing following a Guppy moving average fan. The actual benefits of the Heiken Ashi indicator, the visual sizes of the bars, and the trend continuation and change anticipating features are not used by this robot. Thats much more complicated, this robot mearly reacts to indicator color changes, similar to following a moving average crossover system.

The problem occurs from trying to use smoothed moving averages for precise entry signals. It will never happen. A smoothed moving average is probably the most lagging signal you can find. Its great for determining the overall trend, and staying focused in the right direction, but if you look at the actual entry signals and the placement on the price chart, they are almost always on the wrong end of the range, and only a continuation of the trend can save the trades.

Further the robot has a habit of entering during counter trend moves, before confirmation of trend continuation. Therefore its always setting itself up for a failure, with only the upper trends to pull it through.

2. The idea of closing all trades at the months end is senseless. Trading does not stop and start at the months end, it stops and starts at the weeks end. If the robot accumulates a series of bad trades early in the month, it can wipe out your equity by the months end.

The program does have features that allow it to close trades if the upper trend fails, but I dont know how well these work, considering the lagging nature of the indicator. If they did work better I'm sure Don would already be using them.

What this robot lacks is any sort of a second indicator. It is simply a moving average system and needs some sort of oscillator to indicate position in the range and help time entries and exits.
.......

That said, I still think this robot can be workable, I need to go back to testing and attempting to optimize it. I got frustrated with it and went on to other things.

I think stopping all trading early Friday and closing all open trades on Fri afternoon is a better plan than month end. This would allow most trades to close on their own, and would close drawdowns and take losses before they got out of hand.

I also think that using the option to close trades when the upper trend changes could help prevent uncontroled drawdowns and losses. Unfortunately all of these controls require accepting frequent losses that are much greater than the common gains, so I dont know if they will work succesfully.

One of the best, and most profitable, methods that I've heard of for managing this robot was someone who wrote his own program to monitor account equity. Whenever the total account equity gain reached +5% the program closed all trades and stoped trading for the week, therebye locking in the gains for the week. If it did not hit 5% total gains it simply closed all trades on Fri and took whatever losses or gains it had accumulated. The last I heard this system was consistently reaching its 5% target and closing down early every week. Something to consider.

The other thing that can easily be done is to only run this robot in one direction and only when you can see succesive layers of supporting trend. Consider it as a trend following tool, and dont trust it to think or make the right decisions. Dictate which direction you want it to trade in, and set the parameters to close all trades when the trend changes. This requires a bit of manual supervision and managment, and it requires a good supporting trend, but this is the only safe way I can see to easily employ this robot.

Ken

 

[jackdan]

Hi Ken,


i have an equity bot, and will try it this week. If you feel like wanting the result,i'll post them here, with some setting i have. i use the 3.23 hedge version.
I like the idea of the 5% a week. I sounds a good idea, and could then give the results don aims.
my email is spacanarias at terra dot es.


regards

 

[Ken Long]

JackDan,

I would be happy to read any results you post.

I'm not running the Steinitz right now. There are only so many robots I can run at once. I've also been getting a little tired of robot testing, so far none have made consistant profits, and I really need to work more on my own trading and show some profit there.

The robots I currently run are FAPT and FAPS. FAPT I've been running since it came out in Nov, and its a good scalper, but not perfect. One big loss can wipe out a lot of gains. FAPS is somewhat questionable, like the Steinitz. It looks good for a while, until it gets stuck in a large drawdown. The other robot that I have waiting, and really want to get running is called the TopProfit EA. I was free from one of the development sites, and so far has a great track record. I cant post an attachment here, but if your interested PipCop.com has one of the best testing sites that I know of, and the files for TopProfit are posted there under the new robot testing thread. As well, he ran the Steinitz on demo untill it destroyed the account. All the discussion is still there. Check PipCop if your interested in robot testing... You can also contact me through that site, I try to avoid direct e-mails, I get way to many, and I try to keep my inbox clear.

 

[Ken Long]

Steinitz HAS MTF Demo Test

The risk setting you are running on this bot is much too high for this market. The Steinitz is a no stop loss time bomb in a volitile market such as we are currently experiencing. It is easily capable of running 1000 pip drawdowns across all the pairs if the market changes early in the month. This robot really needs a strong and stable trend to survive, if it gets that it can be immensly profitable, but only if it gets that and avoids any major changes.

 

[Paul Walton]

IT does not work

I must be wrong! Don set up my account (£5000) and was aware of my risk preference. I paid him to set up the robots.... and it blew my whole account. Don's take on all of this ' you have been over trading' A request for a refund was declined. so who is telling the truth here?

This Robot is very very risky

 

[Ken Long]

I must be wrong! Don set up my account (£5000) and was aware of my risk preference. I paid him to set up the robots.... and it blew my whole account. Don's take on all of this ' you have been over trading' A request for a refund was declined. so who is telling the truth here?

This Robot is very very risky

Whats new? I dont mean to be rude, but this is an old story. Don Steinitz doesnt give out any refunds, for the obvious reason. The robot only runs profitably in certain conditions, and he's here to sell it and make money. This is an unfortunate senario, but Steinitz is more of a marketer than a trader. He created his products to sell and make money on.

If you read through the reviews, there are some people who have made money with his creations, but its all short term. There is virtualy no one who will say these things are safe to use long term or without manual oversight, (other than Don).

This is an older robot with a failed signal system, on paper its theory look good, but its timing and trade managment are horrible. It ALWAYS enters its trades late, this is just the nature of using nothing but a smoothed moving average for the signal. This robot pays no attention to the actual price it trade at. It simply reacts to the changes in a smoothed Heiken Ashi average. The only thing that makes this robot money is that it follows the trends, and as long as those trends are still in place it can pull a win out of a bad trade, but beware when the trend changes because the robot doesnt close its bad trades, it holds them indefinately, untill you close them manualy, or get a margin call.

Don's a personable guy. He's got a lot of fans, and enemies, mostly because a lot of people want to believe him and have someone promise them success, and others get disgusted with his lies and false promises.

The bottom line is, when you fail its all your fault, never his robots. His robots are perfect creations, they never fail, but his customers are flawed. Unless of course they praise him and buy more products. In which case they are quoted and used to promote his empire.

Always forward test robots on a demo account, and a live micro account before comitting substantial capital. Better yet, always remove your initial capital from the account as soon as possible and let the robot live or die on its own dime.

 

[ernest8fingers]

Crazy not to listen

Whats new? I dont mean to be rude, but this is an old story. Don Steinitz doesnt give out any refunds, for the obvious reason. The robot only runs profitably in certain conditions, and he's here to sell it and make money. This is an unfortunate senario, but Steinitz is more of a marketer than a trader. He created his products to sell and make money on.

If you read through the reviews, there are some people who have made money with his creations, but its all short term. There is virtualy no one who will say these things are safe to use long term or without manual oversight, (other than Don).

This is an older robot with a failed signal system, on paper its theory look good, but its timing and trade managment are horrible. It ALWAYS enters its trades late, this is just the nature of using nothing but a smoothed moving average for the signal. This robot pays no attention to the actual price it trade at. It simply reacts to the changes in a smoothed Heiken Ashi average. The only thing that makes this robot money is that it follows the trends, and as long as those trends are still in place it can pull a win out of a bad trade, but beware when the trend changes because the robot doesnt close its bad trades, it holds them indefinately, untill you close them manualy, or get a margin call.

Don's a personable guy. He's got a lot of fans, and enemies, mostly because a lot of people want to believe him and have someone promise them success, and others get disgusted with his lies and false promises.

The bottom line is, when you fail its all your fault, never his robots. His robots are perfect creations, they never fail, but his customers are flawed. Unless of course they praise him and buy more products. In which case they are quoted and used to promote his empire.

Always forward test robots on a demo account, and a live micro account before comitting substantial capital. Better yet, always remove your initial capital from the account as soon as possible and let the robot live or die on its own dime.

This is exactly what everyone should do when testing the Forex waters with ANY ExpertAdvisor(or other AutoBot). Back tests are as meaningful as Indicators are accurate-NOT.

 

[DayTraderTed]

Use Reverse Engineering

I read all of the complaints and all of the wonderful capablities of fine tuning this EA, why not have the robot do the opposite. Go long when it initially was suppose to go short and go short when it was progragamed to long. Just reprogram the robot to do the opposite. I am new to robot having to do the trading and I am not a programmer but I can tell you if you can have a program that always loses then you have a program that will always win by just doing the opposite.
Someone tell me it isn't so.

 

[Ken Long]

I read all of the complaints and all of the wonderful capablities of fine tuning this EA, why not have the robot do the opposite. Go long when it initially was suppose to go short and go short when it was progragamed to long. Just reprogram the robot to do the opposite. I am new to robot having to do the trading and I am not a programmer but I can tell you if you can have a program that always loses then you have a program that will always win by just doing the opposite.
Someone tell me it isn't so.

It isnt so..

First, and most obvious, we arent given the code, just the robot. Most commercial robots dont give out their code. Otherwise real programers would tear them apart and just steal their code, which they already do with decompilers. The only robots that offer their code are the free ones on the development forums.

Some of these are good robots, and all are great learning tools, simply because the code is there for us to examine, even if I cant properly read it all.

Other than that the idea of simply reversing the signals is a misstake. This robot is a trend follower. Its designed to be a no loss robot, no stop loss robot.. When it has a strong upper trend it can go for months without a loss simply by holding the trades untill they hit the profit target. But when the trend changes it doesnt react quickly enough and can get stuck holding a bad trade. If you dont actively manage that trade and take your loss quickly you can end up giving back all your gains and more.

If you were to reverse the proccess you would be selling in an up trend and buying in a down trend. With a tight stop loss and no profit target. You would likely see months without a win, and the occaisional reversal which could potentialy make your system. Could you work this way? Do you really think this would be profitable?

Without going as far as reversing the system, simply do your own trend analysis and designate a fixed direction for all the trades. Once a day reevaluate your trend analysis and check any open trades. Set stops, if neccessary. Close all open trades if the upper trend fails, and wait for a confirmed trend before starting new trading.

Its a little bit of work, but this robot did very well when it had a trend.

Still, I think the basic system is flawed. Just like the original ForexAutoPilot, it fails when it tries to time pullbacks with a laging indicator system, often buying the wrong side of the range. But, it often does very well when it ticks off trade after trade in a momentum trend.

I have yet to succeed with this robot or FAPS. (I'm not currently testing them.) No matter how carefull I think I am, they always try to think for themselves and end up doing something stupid when I'm not looking. Both of these robots run on the no stop loss, no losing trades philosophy. As soon as you change that, you change the win loss ratio, and suddenly all those little wins dont add up to the same total. The only way I can think of to use these safely is with a very selective analysis, only putting them on a chart when everything lines up. You need at least two trends in alignment, primary and intermediate, and then you can let the robot play with the minor trends. What constitutes primary and intermediate depends on the time frame your trading.

The basic indicator behind the Steinitz is the Multi Time Frame Smoothed Heiken Ashi Indicator. The problem here is that it is deceiving. When you set it on a low time frame and look at the upper time frame indicator it gives the impression that your seeing multiple bars of that upper time frame, when the actual expanse of time hasnt changed. If your showing 20 bars of time it may be just a few hours, even on the monthly indicator. ie, 20 bars of time on the upper indicator may just be showing a few hours of real time, deceiving you into thinking you have a long term trend when its just one bar. Best to do your own chart analysis, and only use this robot as an aid.