PowerSM Semi-Martingale EA | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Martingale, Reverse Martingale, Modified Martingale, Maths 1 reply

PowerSM Semi-Martingale EA Trades 54 replies

Safe (PowerSM Modified) EA 72 replies

FXTradepro: Strategy using a “Semi-Martingale” Position Sizing 659 replies

Simple MACD w/ Semi-Martingale MM 8 replies

Attachments: PowerSM Semi-Martingale EA
Exit Attachments

PowerSM Semi-Martingale EA

  • #61
  • Edited 11:17pm Mar 19, 2008 9:59pm | Edited 11:17pm
Yes, I've read numerous articles, posts, and comments about it but a lot of these strategies never took into consideration proper entries times, volatility, support and resistance, fibs, and other numerous factors that can drastically improve performance and reliable. I'm sure roulette players would drool if they knew their odds of winning would increase during certain hours of the day, I know I would.

And like I've said before smart traders who use any sort of martingale systems or strategies only use them with money they can afford to lose, inherently know the risk, and never leave profits in the account.

If I had $100,000 account I would trade a nano account with this in a heartbeat because I would never lose on it, why? I haven't figured it up but I figure I could go 100+ progressions before I would be margin called.
  • #62
  • Mar 20, 2008 12:41am Mar 20, 2008 12:41am
Quoting mer071898
Did it hit your take profit level or did it just stop in the middle of a sequence? If you can post your log maybe we can see what happened.

And remember in post #22 I mentioned a possible bug we're trying to correct where the EA will sometimes not enter a sequence when the time and price entries are set to false.
My log is attached.

Even if I reload the EA, it makes no difference. It just doesn't work any more.
Attached File(s)
File Type: zip 20080319.zip   2 KB | 500 downloads
  • #63
  • Mar 20, 2008 1:16am Mar 20, 2008 1:16am
Quote
And remember in post #22 I mentioned a possible bug we're trying to correct where the EA will sometimes not enter a sequence when the time and price entries are set to false.
If I were you I would hold off trading live until all the kinks are worked out. I sent your log to my coder I'm hoping he can figure it out for us soon. if you want in the mean time you can try out the EA using the timed or price entries to make sure those work as well, thanks. Let me know if you have any other problems.
  • #64
  • Mar 20, 2008 1:21am Mar 20, 2008 1:21am
Quoting mer071898
If I were you I would hold off trading live until all the kinks are worked out. I sent your log to my coder I'm hoping he can figure it out for us soon. if you want in the mean time you can try out the EA using the timed or price entries to make sure those work as well, thanks. Let me know if you have any other problems.
OK, although best forward testing is with a live account (just nano's, and I know, it's money).
Thanks
  • #65
  • Mar 20, 2008 1:41am Mar 20, 2008 1:41am
I agree but I just don't like to see people lose money using an EA that is still being tested but I do appreciate the input especially on a live account as long as your OK with it. The only other thing I can suggest in the mean time while I'm waiting to hear back is to make sure you have the current MT4 build (213) and when you remove the EA make sure to restart your MT4 platform, then add the EA back on the chart. Also maybe try increasing your "ordertries" setting to 10 instead of 5 to give the EA more chances to enter orders when it receives an error code. These may or may not work so let me know.
  • #66
  • Mar 20, 2008 2:09am Mar 20, 2008 2:09am
Quoting mer071898
I agree but I just don't like to see people lose money using an EA that is still being tested but I do appreciate the input especially on a live account as long as your OK with it. The only other thing I can suggest in the mean time while I'm waiting to hear back is to make sure you have the current MT4 build (213) and when you remove the EA make sure to restart your MT4 platform, then add the EA back on the chart. Also maybe try increasing your "ordertries" setting to 10 instead of 5 to give the EA more chances to enter orders when it receives an error code. These may or may not work so let me know.
No worries Mer, I am aware of that.
As soon as I am back from my day job I will try the other things to get the EA running and that it does something again.
I like your strategy and want to help improve it or at least give results.
  • #67
  • Mar 20, 2008 2:22am Mar 20, 2008 2:22am
Thanks for all the help!
  • #68
  • Mar 20, 2008 12:43pm Mar 20, 2008 12:43pm
Here are the daily and cumulative statements for all three pairs. Another successful night.
Attached File(s)
File Type: zip PowerSM Statements.zip   9 KB | 590 downloads
  • #69
  • Mar 20, 2008 1:25pm Mar 20, 2008 1:25pm
Quoting mer071898
Here are the daily and cumulative statements for all three pairs. Another successful night.
I wonder why we wouldn't use this 24/7. Are we just trying to save a couple of entries during the slow times? I've been using it 24/7 (just set the "start" and "stop" time to 0) and it has made over 20% since Monday.

I started it on a $1,000.00 demo account, so I used the nano spreadsheet and doubled the lots. However, then I decided to go with 2 currencies (Eur/Usd and Usd/Chf), and didn't take the time to set the lots back to where they should be. So in essence I have doubled the risk from the spreadsheet.
  • #70
  • Mar 20, 2008 1:36pm Mar 20, 2008 1:36pm
The problem is when I tried doing it 24/7 with the original FXTradepro EA it didn't take long before I hit a ranging period where the EA would go from sequence 1 to 16-17 in a matter of an hour. I'm not saying it can't be done but you are opening yourself up to a lot more risk during the ranging part of the day. So far I haven't gone past level 10 and I would prefer to keep it that way.
  • #71
  • Mar 20, 2008 1:47pm Mar 20, 2008 1:47pm
Yes, that makes perfect sense. What time of the day do you trade (Eastern or GMT)?

Quoting mer071898
The problem is when I tried doing it 24/7 with the original FXTradepro EA it didn't take long before I hit a ranging period where the EA would go from sequence 1 to 16-17 in a matter of an hour. I'm not saying it can't be done but you are opening yourself up to a lot more risk during the ranging part of the day. So far I haven't gone past level 10 and I would prefer to keep it that way.
  • #72
  • Mar 20, 2008 2:00pm Mar 20, 2008 2:00pm
I am CMT which is GMT+6. I just leave the EA with the defaults which is 07:00 GMT which was the Frankfurt opening @ 01:00 my time, but since we are now in daylight savings time here, now 07:00 GMT is now the London open @ 02:00 my time.
  • #73
  • Mar 20, 2008 8:13pm Mar 20, 2008 8:13pm
Mer,

I've read your thread thru so far and watched your video.
Listening to you, I can tell you're an intelligent man and you're
taking this project seriously. I say project because, I too endeavored
to "crack the code" on such a method by integrating statistical and
probability study into the expert and believe me, it became a PROJECT.

Yours is far less complicated and straightforward and simple is better.
I believe choosing well timed entries is still the key to mitigating risk
and keeping it under 10 bounces in a series, no matter the approach.

Besides the video, I also viewed the statements. So far, this is good too.
You are right on track and should make 2 to 3 percent per month with
a fair degree of consistency. Natually, the risk is what the risk is and
the "consistency" part only comes under attack when, for all your efforts
to minimize the risks, they still pop up and put you to the test.

I think this is EXACTLY why an automated execution of such a method is
essential, because, after several bounces in a row and some decent
drawdown, it's way too easy to lose your faith in between bounces and
simply abandon the chase. My opinion from personal experience is to
let the machine do its work. Win, Lose, or Draw. Tweak and test and
continue to make improvements, but I think one should either automate or
go back to manual, so as not to second guess the process in motion.
Just my opinion.

Now, having said that, and testifying to the fact that the method you're
espousing has merit and can be profitable in the right hands, I have
a much larger question for you. As I respect the work you've put into this
and all the help you've offered to all of us here, I don't want you to be
at all defensive if I ask this. Purely a thinking fellow's question:

Don't you think you need a good chunk of cash to do this the RIGHT WAY ?

Using $100,000 USD returning $500 to $1000 per week and quite possibly
2 or 3 thousand a month is definately worth investigating. It could
potentially replace the "9 to 5" income of many a working stiff.
THIS IS POWERFUL and COULD CHANGE SOMEONE's LIFE.

Would not want to discourage anyone from trying to get ahead, but
do you think this is the BEST way to grow a smaller account of say
$1,000 or so at the rate of 2 or 3 dollars a day ? It will take years.

Granted, if it fails you'd be out 50K or 100K instead of 1K, but I cannot
grow old trading nanos for the next 20 years with a only thousand bucks.
Again, not to discourage, but do you think this is a good idea for small
accounts compared to the reality of time it will take to make the compounding kick in ?

Yes I know it's 25%- 35% per year. Account could triple inside 3 years.
In ten years, assuming 2 percent/mo. compounded (and assuming no blow-out)
it would grow by over 1,000 percent (play with the numbers).
If it's $10,500 after ten years I don't care because I have starved to death
waiting for the 17th bounce !!

So, asked another way, Do you think, considering the time it takes this method
to produce meaningful results, to trade this way on a 3 or 4 figure account
is worth adopting or worth the effort ?

After all that and trying to be careful not to offend you, if you get p*ssed
for me asking, I'm going to give you a nuggie - right here on this forum.

And by the way, Good Work.
  • #74
  • Mar 20, 2008 11:20pm Mar 20, 2008 11:20pm
Toshi, I fully agree with you 100% that a large amount of capital is truly needed to make this EA work to it's full potential. Granted there are only a few individuals on this forum that could possibly dump this much into an account in the first place let alone have the nerve to trade any type martingale or semi-martingale system. But keep in mind there are bank traders and fund managers using automated EA's for trading millions and millions of dollars daily, so the commercial aspect is always possible.

But again, I believe that someone using this EA should also trade other strategies using this EA as a supplement to increase profits to reach that level where you can trade it solely on a larger account if so desired.

My goal when I go live I will be funding about 10k and dumping all profits each month into my primary trading account which as of today is around 48-49k. By using the EA as a separate secondary trading income will help me produce a larger primary trading account, on which I can increase lot sizes faster, which in turn allows me to fund a larger EA account, which in turn...well, you get the idea.

As FXTradepro stated numerous times in his original thread the the EA is just a tool and is not a substitute for proper trading techniques, analysis, and money management.

As for me getting pissed off because someone, like yourself, has a meaningful opinion is crazy, I value every opinion placed on this thread and don't ever feel that you would offend me.
  • #75
  • Mar 21, 2008 1:25am Mar 21, 2008 1:25am
I forgot to switch off the EA when I went to work. When I got back home I saw it still was running. It blew up my nano account. Yes, one must have a huge account...
There was a trading range in the GBPUSD from 12:00 to 13:30 GMT (March 20th) of about 20 pips. It took up to 23 trades.
I didn't expect it so soon, because I wanted to tweak the parameters, but unfortunately it did happen.
Still like the system.
  • #76
  • Mar 21, 2008 10:35am Mar 21, 2008 10:35am
Good intro for the 'ol Martingale discussion! Already stated earler... I too am an ex-D'Alembert trader. This approach here is more conservative, and one thing I like is that it sequences relatively quickly! But, as in all progressive lot systems, it all depends on the percentages of the system. All gaming-based discussions of PLM (progressive lot managment) have pretty much fixed percentages, by game. We don't have that luxury, which can be both good and bad. The better the edge we have, the more successful our PLM will be. But, we don't really know what the future will bring. Vegas does. The roulette wheel will cough up the same house win% today as it did yesterday, and will tomorrow, next week, and next year.

So, as Mer said, previous analysis/discussion of PLM based on data from gaming is of limited use to us. Interesting, but not really applicable. We have a BIG advantage in that we can control our entry, our time-of-day (volatility) etc. This particular method is interesting because it is based on the premise that we will hit a minimum 50p run sometime in 20 or so trades. We can further increase our "odds" so to speak by selected WHEN we start this run. As above, start it at the wrong time, and you blow your account. But looking at past charting, trading the EU on the European session, on non-holidays (today would have been a bad day to trade, for example...), there has ALWAYS been at least 50p movement in the session. (I say 50p because I am assuming we are short, and the market moves long and hits our SL at -10p, then we go long, so we need another 40p to win, total 50p... plus spread...)

We can further help ourselves by trying to get our first trade right! (which is what everybody does... but for us, it really doesn't matter!! )

Oh... and it actually is very easy to trade the EA entry per our own indicators, without having to deal with the false-false bug. Set the EA on timed Entry, with your session time values, and set to trade long or short first depending on your analysis, but click the "OFF" button for the EA, so you have an X where the smiley face would be. EA is applied and everything,... then, watch your charts... when your indicator says go, you simply click the "ON" button for the EA,... your EA verifies that you are within your set trading session.. (kind of a nice check...) and boom, enters you long or short on the next tick, according to how you set the EA up. Actually, I like this better than trading False-False, because 1) you get a check on the session times so you don't enter a sequece outside your plan, and 2) you get the End-time limit as well.

Anyway, all the above, plus good MM can make this a very viable system... (I think!!)
  • #77
  • Edited 6:22pm Mar 21, 2008 1:10pm | Edited 6:22pm
fewhills, that's why I don't trade Gbp/Usd (or Gbp/Jpy or Gbp/Chf) because of their volatility. These pairs can easily range 20-30 pips before any kind of breakout can occur, using less than a 20-30 pip stop is asking for trouble. Also being Easter weekend I wouldn't have traded due to the expected low volume. Again, this is why I strongly suggest not trading live until the EA is properly tested. Just don't get discouraged over it though, maybe just stay with the pairs for live trading then test out the other pairs on a demo account. I know on the original thread they were having good success on Eur/Jpy with 100/75/25 and Gbp/Jpy with 140/105/35.

Glad to have you here mrkam, the more the merrier.

As expected my trades are on hold because of the ranging but since but these pairs are more stable than Gbp/Usd and my trading was stopped at 17:00 GMT, I'm not anywhere near level 23 as in the pictures.
Attached File(s)
File Type: zip pics.zip   202 KB | 699 downloads
  • #78
  • Mar 21, 2008 2:33pm Mar 21, 2008 2:33pm
First of all, wanted to thank everyone of you for your contributions. Just found this thread while thinking in a very similar direction.

The SL and TP settings obviously must be customized to each particular pair, according to the latest volatility. In my view, a 40-pip TP is too high, and that is what leads to reaching the 10th+ level. A narrow TP I think will reduce the number of levels reached.

As I see it, a good entry only saves us one level, nothing more. A wrong entry will shortly be reversed, thus getting us into the appropriate direction. Unless... we are in a range, but a system like this is based on the assumption that we cannot predict neither a trend nor a range.

Where I want to get here? A 24/5 robot with tight TP's and SL's, always with one trade open. If TP is hit, then a new trade in the same direction is opened. SL hit, then SAR. Suitable pairs are those with tight spreads and high volatility, and the calculation of the SL and TP should be done in addition of the spread.
I still didn't tried Mer's EA, but will start next Monday and post my results stating the used settings, so we can compare and optimize.

One question: Once a trade is closed, either at the SL or TP, does the EA open an additional trade, or you should wait to the next day/candle/manual order?

Two problems to deal with here: News times and spread widening done by the broker. Trade must be disabled during major news releases (let's say 1 hour before and 2 hours after, just to make sure not to catch a spike, or a spike to catch us, that is).
As someone said in the other thread, brokers widen the spreads from time to time to screw people trading these kind of systems. Ideally, the EA should have a safety lock, disabling trading when the spread > X pips.

Another problem I see: When in a strong trend, we would be opening and closing trades in the same direction, thus paying spreads that otherwise we would be earning ourselves. I am thinking in a way to optimize the earnings here, but any comments will be warmly welcome.

Well, sorry for the long post, but my head has been bubbling these last days...

Regards,

Middleastern
  • #79
  • Mar 21, 2008 6:06pm Mar 21, 2008 6:06pm
Quote
The SL and TP settings obviously must be customized to each particular pair, according to the latest volatility. In my view, a 40-pip TP is too high, and that is what leads to reaching the 10th+ level. A narrow TP I think will reduce the number of levels reached.
Middleastern, everyone will have different opinions on TP and stops, that's the reason I put up the EA for everyone so they can play with it as much as they want. With regards to TP and stoploss if you take the calculator on post #1 you can put in whatever TP and stoploss you want and see where your profits will end up. I, personally, have had no trouble so far reaching 40 pip TP profit levels on any pair I'm trading yet. The EA uses a 4-1 TP to stop ratio and specific lot increment sizes that has been tested to ensure you have a profit no matter what level your TP is hit at (again, play with the calculator) When you start reducing TP levels you must lower your stop to match, thus increasing your risk to being stopped out way too often especially in a sideways ranging market like today as fewhills found out. Also trading highly volatile pair like Gbp/Usd, Gbp/Jpy, or Gbp/Chf makes you more prone to stronger whipsaws and thus more stops being hit. In order to use this EA it's all about control, using controllable pairs, controlling entry points and times, and controlling your money management. as said before the EA is just a tool that you are in control of, not the other way around.

Quote
One question: Once a trade is closed, either at the SL or TP, does the EA open an additional trade, or you should wait to the next day/candle/manual order?
The EA only opens a new trade in the opposite direction when the stop is triggered. When the EA hits it's TP the sequence stops and will not start until the next time or price is reached (if you use the time or price entry). The EA is suppose to enter trades immediately on the next tick when placed on the chart if both settings are set to false but currently there is a bug (which is be looked at by my coder) that's causing it not to enter trades.


Quote
Two problems to deal with here: News times and spread widening done by the broker. Trade must be disabled during major news releases (let's say 1 hour before and 2 hours after, just to make sure not to catch a spike, or a spike to catch us, that is).
As someone said in the other thread, brokers widen the spreads from time to time to screw people trading these kind of systems. Ideally, the EA should have a safety lock, disabling trading when the spread > X pips.
The EA does have an input for slippage built in and an "ordertries" function to help with news events and spikes. Besides you can always turn off the EA during these times by clicking on the "Expert Advisors" button.

Quote
Another problem I see: When in a strong trend, we would be opening and closing trades in the same direction, thus paying spreads that otherwise we would be earning ourselves. I am thinking in a way to optimize the earnings here, but any comments will be warmly welcome.
Since you haven't used the EA yet you should understand the EA will open a position and if stopped out will open a new trade in the sequence in the opposite direction. Before you start using it I would suggest reading through the original thread on post #1 to get more info about the concept of what the EA does and how the calculator works to help you out.
  • #80
  • Mar 21, 2008 6:40pm Mar 21, 2008 6:40pm
Thanks for taking the time to answer all that

I'll start using the EA this week, and I'm sure things will get clearer. I was just trying to adapt the EA features to what I had in mind, which is a little different of what was proposed here. Anyway, I'm eager to see the EA in action, anyway...

Thanks again, Mer

Quoting mer071898
Middleastern, everyone will have different opinions on TP and stops, that's the reason I put up the EA for everyone so they can play with it as much as they want. With regards to TP and stoploss if you take the calculator on post #1 you can put in whatever TP and stoploss you want and see where your profits will end up. I, personally, have had no trouble so far reaching 40 pip TP profit levels on any pair I'm trading yet. The EA uses a 4-1 TP to stop ratio and specific lot increment sizes that has been tested to ensure you have a profit no matter what level your TP is hit at (again, play with the calculator) When you start reducing TP levels you must lower your stop to match, thus increasing your risk to being stopped out way too often especially in a sideways ranging market like today as fewhills found out. Also trading highly volatile pair like Gbp/Usd, Gbp/Jpy, or Gbp/Chf makes you more prone to stronger whipsaws and thus more stops being hit. In order to use this EA it's all about control, using controllable pairs, controlling entry points and times, and controlling your money management. as said before the EA is just a tool that you are in control of, not the other way around.

The EA only opens a new trade in the opposite direction when the stop is triggered. When the EA hits it's TP the sequence stops and will not start until the next time or price is reached (if you use the time or price entry). The EA is suppose to enter trades immediately on the next tick when placed on the chart if both settings are set to false but currently there is a bug (which is be looked at by my coder) that's causing it not to enter trades.


The EA does have an input for slippage built in and an "ordertries" function to help with news events and spikes. Besides you can always turn off the EA during these times by clicking on the "Expert Advisors" button.

Since you haven't used the EA yet you should understand the EA will open a position and if stopped out will open a new trade in the sequence in the opposite direction. Before you start using it I would suggest reading through the original thread on post #1 to get more info about the concept of what the EA does and how the calculator works to help you out.
0 traders viewing now