100% Win Math Grid Ea | Page 60 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Grid - New indicator that replaces chart grid 60 replies

Grid (not grid system) or horizontal lines indicator 6 replies

100% win grid EA 45 replies

Grid, Burst Grid, Grid Risk and Grid Profit 1 reply

Double Time Grid and Pip Grid 0 replies

Attachments: 100% Win Math Grid Ea
Exit Attachments

100% Win Math Grid Ea

  • #1,181
  • Jun 29, 2010 1:04am Jun 29, 2010 1:04am
Quoting xmph
Viper and Wael,
In addition to taking a scenario-based approach, I'm still investigating a Math_mitigation based on Strong Support/Resistance.

The interest of Strong Sup/Res as I see it is that you more or less predict these ones for ... let's say a last-2weeks or month Sup/Res. What I'm trying to put together is to consider these as BreakEven objective if we've been oscillating for some time.

Thus, instead of keeping the pattern of the grid, I would modify the grid to simply BE@Sup/Res. Yes one would loose time, and as well loose money by not...
Very interested in the idea, can you please explain more with an example if possible!! where would you initate the grid in terms of support & resistance.

how would you identify support and resistance automatically?

would you switch it off in time of news such as interest rates & NFP, Interest rates and so on? how about market spikes?


Please explain more i am very interested in the approach.

Thanks

W
  • #1,182
  • Jul 1, 2010 7:16am Jul 1, 2010 7:16am
hai viper, I have finished 3 week run this EA on live , and it is great resoult, i can make 5% profit every days, but I just get the message from the broker, they say my EA had send more than 4000 messages in a few hours, so they stop my account , they ask me to stop use this ea. actualy I run every days 3 set and every set i think not more than 75 pending order. so, viper can I ask your help to make some modification on the EA, can we hide pending order from the broker? ( we put the pending order in our program ) Thank for your help....
  • #1,183
  • Jul 1, 2010 10:15am Jul 1, 2010 10:15am
Quoting dastum
hai viper, I have finished 3 week run this EA on live , and it is great resoult, i can make 5% profit every days, but I just get the message from the broker, they say my EA had send more than 4000 messages in a few hours, so they stop my account , they ask me to stop use this ea. actualy I run every days 3 set and every set i think not more than 75 pending order. so, viper can I ask your help to make some modification on the EA, can we hide pending order from the broker? ( we put the pending order in our program ) Thank for your help....
Glad you doing well with this one mate, can you provide us more details about your settings and all, and which Broker you are using ?

How about your account size and the % on entry.

Thanks,

W
  • #1,184
  • Edited 10:07pm Jul 1, 2010 8:35pm | Edited 10:07pm
Quoting wael117
Glad you doing well with this one mate, can you provide us more details about your settings and all, and which Broker you are using ?

How about your account size and the % on entry.

Thanks,

W
wael
thanks for your attention, I run in level 2 and increament 12 pips, I test with small fund, 200$ account size and with leverage 1:500 and lot size is 0.01, I use FIGfx Broker, this broker also not permit run pending order EA during The news. wael, can you make small modification on this EA, we hide (stealthmode ) all our pending order, so the broker can only see all open order in the platform.. I thing more powerful our EA , more broker try to protect it..
Wael ,do you have recomended broker for run this EA?
  • #1,185
  • Jul 1, 2010 10:26pm Jul 1, 2010 10:26pm
look great is the download link at the beginning of the thread the latest version of the ea?
  • #1,186
  • Edited 6:22pm Jul 2, 2010 6:20pm | Edited 6:22pm
Quoting wael117
Very interested in the idea, can you please explain more with an example if possible!! where would you initate the grid in terms of support & resistance.

how would you identify support and resistance automatically?

would you switch it off in time of news such as interest rates & NFP, Interest rates and so on? how about market spikes?


Please explain more i am very interested in the approach.

Thanks

W
Hello Wael,
My intention is to have a dynamic grid. For that, I'm thinking of using Murray Math levels as it is dynamic as well.

What I'm trying to do is basically investigate if it makes much sense to keep adding positions if you're expected BreakEven level is beyond that of a strong level (Res/Sup).

So far, I wrote a little formula for BreakEven. I'm about to finish a little function that draws a line on the chart to show the BE-Line and I will further investigate my idea.

For quick integration, the BE-Line function I wrote should be EA independent. I will add it to Viper's code so that we may further discuss this.

Now, (I'm not that of an expert in MT4), I realized that my MT4 spread is blocked at 8 in backtest today ... do you know why that is ?
It was good though, coz although I had small grid setting to work a little on it and have lots of orders triggered, I realized that the behavior of the EA with a grid increment of 10 when the spread is 8 .... is basically non-sense. So this is one more thing that makes me think a dynamic grid is good.

Finally, I was backtesting in Alpari, and now moved to another broker (FXCM), and I realized that the profits are really dependent on the Broker limits in terms of what it accepts has a minimum Stoploss. I recently provided you with the code I used for trailling. You can see that this code make sure that the broker's max StopLoss (to prevent scalping) is not reach before updating an order .... by switching to different MT4 from different broker, I noticed a clear difference in terms of what happen in the end. ...... just saying.

That's it for now.
  • #1,187
  • Jul 3, 2010 12:59am Jul 3, 2010 12:59am
my broker has complained about too many orders hitting there server from my account... i guess ill have to adjust the values.. its only a demo accoount to so i dont know why they are whinging
  • #1,188
  • Edited 2:16pm Jul 4, 2010 10:43am | Edited 2:16pm
Quoting xmph
Hello Wael,
My intention is to have a dynamic grid. For that, I'm thinking of using Murray Math levels as it is dynamic as well.

What I'm trying to do is basically investigate if it makes much sense to keep adding positions if you're expected BreakEven level is beyond that of a strong level (Res/Sup).
Here is an illustration as I found some time this Week-End to work on it
A classic Death Trade:

Attached Image (click to enlarge)
Click to Enlarge Name: ProblemScenario_Part1.jpg Size: 108 KB
Attached Image (click to enlarge)
Click to Enlarge Name: ProblemScenario_Part2.jpg Size: 61 KB


Please, pay attention to where the BeakEven Line lies.
now what happen if you place a new trade (call it a mitigation order) at a Key support Level:

example of a 0.01 order .... BEline moved ... that's good .... but not quite enough:
Attached Image (click to enlarge)
Click to Enlarge Name: SolutionScenario_0.01Lot.jpg Size: 88 KB


Same, with the Key level illustrated using the Waddah Attar Fib levels and for a 0.02 order ... much better:
Attached Image (click to enlarge)
Click to Enlarge Name: SolutionScenario_0.02Lot@KeySupport S1 at BreakEven.jpg Size: 87 KB


So, why am I in love with the second mitigation (which is what I meant since the begining): Because by placing this order at a key level, I also "moved" the BreakEven level in between two Support Level. Now, that makes sense to me.

Few important notes:
- This is performed with my code and its a little different than Viper's which explain why I got into a death trade quicker than Viper.
- I have a grid increment of 40pips and a BuyLevel1-SellLevel1 separated by 2xIncrement
- Mitigation can be performed by whether adding new positions, or changing the grid increment WHILE adding a multiplier as of course, change the increment will alter the math.

Here are few indis that you may be willing to experience with. They work in Backtest.
Attached File(s)
File Type: mq4 Waddah_Attar_Pivot_Fibo_First.mq4   6 KB | 1,037 downloads

Attached File(s)
File Type: mq4 Waddah_Attar_Pivot_Fibo_Second.mq4   7 KB | 944 downloads

Attached File(s)
File Type: mq4 SR_MultiplePivotsV2.mq4   26 KB | 1,240 downloads


That's all for now.
Test it and let me know.

XmPh.
  • #1,189
  • Edited 10:48pm Jul 4, 2010 8:07pm | Edited 10:48pm
Quoting xmph
Here is an illustration as I found some time this Week-End to work on it
From the sample tests I've run today, the concept seems to work well. I'll keep you posted.

My statement, so that we may compare along the way of the implementation I do and report in this thread. We'll see if the strategy I suggest will get rid of the huge DD .... implementing just a tiny piece of the strategy and only on the Short part helped survived 3 StoppedOut I had during the same period with similar parameters.

The one big issue: The Sunday Gap !

Attached Image (click to enlarge)
Click to Enlarge Name: Mitigation_oneSided_MinimalConcept.jpg Size: 68 KB
  • #1,190
  • Edited Jul 6, 2010 5:44am Jul 5, 2010 8:07am | Edited Jul 6, 2010 5:44am
Warm regards to the oldies here!
Posting after a long gap, once again ! Really not being able to devote more time and testing to it. Darn!

Anyway, for those trying to reinvent the wheel or waste their precious time, testing same/minor things (for this EA with lot of variables involved (to trade Fx)) :
1) More the levels of grid, more close you are to a Death Trade (from each new grid start). In other words, relative speed moving towards the death trade increases with no. of levels.
2) Fixed Grid Size is doomed for Death Trade. Market conditions are variable and so should be each grid.
3) Point (2) can be considered a "LAW" & for sure when there are no time filters and someone likes this to trade 24*6.
4) Good Entry filters always increase the odds in your favor than the Death Trade.(which ones? I think i posted some food for thought, around page count of 60's. Even after improving further, i still have more 'better' work to do regarding entry filters)
5) Thinking Death trade will not come early and trading it live, praying, you will mint before that is 'Donation' (in long term for sure). Help the poor rather & help you karma!
6) Entry filters/strategies are most important. More logical/market-condition-based they are, lesser the need for exit strategies.
7) Black Swan/News events will always be a nemesis for this strategy. Exit strategies won't be of much use!
8) With more odds in our favor (with above), i still support "Withdrawal is the best MM for this strategy" but add "but only without compounding". With compounding, over a period of time, it has no statistical edge and has a random result (Only exception is if one has found something significant enough to delete phenomenon called "Death trade" from his/her dictionary)
9) I am not sure what fate this strategy has, over a long term, as its all based on fate of hedging and related rules Fx will have , in future.
10) Multi Grids- I don't even want to think about them. Point (7) again. Still interested - PlasmaPants is playing with those kinda of "multi-grid plasma-rays" even now.

Hope this helps.
Regards


Quoting xmph
From the sample tests I've run today..........
Attachment
You need to check your back testing setup first before trying/inferring things based on half-cooked/poor-data results(as can be seen from your screen shot)
Just an advise. No offense.
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #1,191
  • Jul 5, 2010 9:01am Jul 5, 2010 9:01am
Quoting forexsaint
You need to check your back testing setup first before trying/inferring things based on half-cooked/poor-data results(as can be seen from your screen shot) Just an advise. No offense.
Thanks for the advice. Now please, provide me with better data-quality for testing and it will be my pleasure to run my system on them, as like most ppl complaining about poor data quality they never provide "the good ones". Until then, I'll work with what I have and it has plenty of "high-quality" death trades scenario for testing.

As per the trying/inferring things .... I suggest modding the grid with respect to the market conditions, namely pivot/Sup/Res ..... eventually inferring something close to your
Quote
More logical/market-condition-based they are, lesser the need for exit strategies.
Although I disagree in that Entry is as important as Exit.

Quote
No offense.
I love forum discussions, it's like a sweet and sour mix of ppl leveraging their anonymity to let go their anger of the day ! Eventually not that of a Saint after all ;-)

That said, I will keep posting my investigation and results. If anyone believes this is a waste of "thread space", let me know and I'll stop posting. That is not much of an issue for me. I don't need to prove anything to anyone, I was just willing to share my work/ideas with the thread that first give me most of it.

XmPh.
  • #1,192
  • Edited 11:45pm Jul 7, 2010 11:42pm | Edited 11:45pm
Quoting forexsaint
You need to check your back testing setup first before trying/inferring things based on half-cooked/poor-data results(as can be seen from your screen shot) Just an advise. No offense.
99% quality, 0 chart issue .... if I can do better you'll have to detail precisely how ? I followed this tutorial to convert the dukascopy tickdata: http://eareview.net/tick-data#prepareddata

Viper, that would be excellent to compare your results as well on these datasets if you keep on improving ur EA.

So, for the purpose of comparing to the future improvements I'll make based on my "suspicious and misjudged" comments (I'm joking Saints, No Offense ;-) oooops, I did it again), and considering I only implemented a one-sided minimal notion of support with respect to BreakEven level:

Attached Image (click to enlarge)
Click to Enlarge Name: Mitigation_oneSided_MinimalConcept_M1.jpg Size: 71 KB

Attached Image (click to enlarge)
Click to Enlarge Name: Mitigation_oneSided_MinimalConcept_M5.jpg Size: 66 KB


I'm currently investigating how chaging timeframe could affect so much performances. I do use (NewBar() and few other "market description" thingy so I'll digg a little further).

Viper: Did you take this into account: http://www.100forexbrokers.com/margin-call-vs-stop-out
Each Broker as a different rule to "automatically" close loosing positions depending on floating loss. GoMarkets has a 50% floating loss triggering the closing of the highest loss-position. This must be taken into acct by the EA. This dramatically affect "real-life" death trade mitigation !

Cheers.
  • #1,193
  • Jul 8, 2010 3:44am Jul 8, 2010 3:44am
Quoting xmph
99% quality,...
Hi Xmph,
I am trying to understand what kind of expert is "FCD_GridStrategy" from your test and where could I found it,
Can you please share the code ?

Thanks, you all doing here great job.
  • #1,194
  • Jul 8, 2010 8:14am Jul 8, 2010 8:14am
Quoting benseg
Hi Xmph,
I am trying to understand what kind of expert is "FCD_GridStrategy" from your test and where could I found it,
Can you please share the code ?
Thanks, you all doing here great job.
FCD_GridStrategy is the grid strategy I coded based on the knowledge spread around this thread for the most part. I coded the lot strategy I was comfortable trading with, with the math that makes sense to me (considering lots have been described), and so far I added barely one of the concepts I mentioned earlier.

I'm sharing my ideas/understanding, with what I believe are illustrations so as to make things move further (I hope). I will try to illustrate the impact of the mitigation strategies I'm suggesting/inferring. However, I do not share my code. I do not believe in free beer, I believe in free experience-sharing, free-discussion, and eventually I'll share piece of codes like I did, but not my EA. Viper is a better man than me on this subject, and he already did an excellent work. It should not be too difficult to take part in the discussion and mod Viper's code if you wish .... but a little effort is mandatory for every success ;-) ....

Wael is a perfect example as I believe he is not coding, but his "trailing" ideas are good and based on sound analysis. Although I believe trailing cut short most of potential profits (from my experience), it reduces tremendously the risks (of further bouncing, eventually leading to a death trade, who knows!) while guaranting profits. I personally use trailing (as I described it, quite similar to Wael) in my code.

regards,
  • #1,195
  • Jul 9, 2010 12:40am Jul 9, 2010 12:40am
See the results!

Now you are on right track , as far as testing is concerned.
Make sure you don't have any order related errors (like order send , etc) in your expert log (after testing) or any other weird issue.
Then you have something (to be called a 'real' Test RESULT) for statistical testing, in your hand.
Try out the variable spread loader, in case you are trading this EA(or going to) on variable spread broker.

Regards

P.S. Been busy with other things, so don't check in here often. Anything urgent , PM me.
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #1,196
  • Jul 9, 2010 10:00am Jul 9, 2010 10:00am
Quoting xmph
From the sample tests I've run today, the concept seems to work well. I'll keep you posted.

My statement, so that we may compare along the way of the implementation I do and report in this thread. We'll see if the strategy I suggest will get rid of the huge DD .... implementing just a tiny piece of the strategy and only on the Short part helped survived 3 StoppedOut I had during the same period with similar parameters.
RESPECT to all your efforts man,

and yes you are true about me.. I am not coding i just believe in the logic of the EA.. many people did quit on this EA, but i am going to post a result somehow similar to yours very shortly .... i am generating it right now.. and this is one of the reasons i am still sticking with this..

btw congratulations.. i think you are the first one to post a full year back test result without an account blow up..

i'll be second soon ..

Take care X.

Note.. man i would love to put my hands on your code to run some tests on it.. i know that you wont grant me this wish.. but once again great effort man.. i wish i knew how to code...

keep looking out for my results.. will be waiting for your feedback..

Viper i will be looking forward for yours too...
  • #1,197
  • Jul 9, 2010 10:20am Jul 9, 2010 10:20am
couple of points

1 - Yes i remove the settings
2 - Yes, this is the original viper 2.6.2, code hasn't been touched.. i just played with the settings to get this result
3- Yes i think this is the 2nd result (after xmph) for a full year with no blow out.
4 - Yes i think this is the first one with the original code ( hope i made Viper proud )

5 - Account started with $100,000.00 USD
6 - Initial lot is 10 standard lots (i.e. 1,000,000.00) Entry is fixed.
7- yes you can play with the settings to work with $1,000.00 account with micro lots
8- leverage used is only 1:100 not to 1:500
9 - Account ended up balance of $951,105.31 i,e, 851% profit in 12 months

ENJOY!!!

Productive feedback is more than welcome!!!
Attached Image (click to enlarge)
Click to Enlarge Name: go1.jpg Size: 160 KB
  • #1,198
  • Jul 10, 2010 12:22pm Jul 10, 2010 12:22pm
Hello to all

This is really a great thread, still it would be awesome if someone could make a summary post or something.

I tried to read the whole of it but my eyes started to hurt on page 20.. can anyone post an update on how the EA discussed in the beginning evolved and what is the current stage reached?

Thanks a bunch!
  • #1,199
  • Edited 9:06pm Jul 11, 2010 8:36pm | Edited 9:06pm
Quoting blank_jack
I tried to read the whole of it but my eyes started to hurt on page 20..
Lol .... can you come to my house my fingers hurt and I need to pull out some programming !
  • #1,200
  • Jul 11, 2010 9:00pm Jul 11, 2010 9:00pm
Quoting wael117
ENJOY!!!
Productive feedback is more than welcome!!!
Week-End backtesting in order to have some comparables (next 2 posts). Ticks come from Dukascopy (see the previous link I gave), with a fixed spread of 4 (a worst case scenario I believe). Modeling quality 99%:

Here are Viper's results withdefault settings (thus original method with Viper's way to add positions), with increment 40 for:

2008-2009
Attached Image (click to enlarge)
Click to Enlarge Name: Viper_2008-2009.jpg Size: 118 KB


and for 2009-2010
Attached Image (click to enlarge)
Click to Enlarge Name: Viper_2009-2010.jpg Size: 118 KB
0 traders viewing now