100% Win Math Grid Ea | Page 30 | 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

  • #581
  • Oct 1, 2009 7:32pm Oct 1, 2009 7:32pm
rbrtsha,

Have previously looked into what you mentioned but it does require the TP levels to be dramatically extended in order to be profitable. You can extend the original Mgrid levels which will reduce the multiplication of lot sizes but it does sometimes stuggle to take a profit also.

Whilst you are correct it will take longer to hit the 'Death trade' it will also mean (by extending the TP levels) there is more opportunity that more levels will be taken out and the trade will soon turn negative anyway.


Viper/Saint/Plasma

Further thoughts on this 'death trade' scenario. Im more convinced now about balancing the buys/sells at a particular comfort zone. WHY??

Becuase we just don't know which way price is going to lean. We do know eventually it will hit the TP level (within a reasonable setting) but we can't forsee when/where. We don't know if price will retrace or continue. We need to be able to take advantage of price if it does retrace and also take advantage if price continues on without digging any deeper (if possible) but somehow maintaining a constant level between the buys/sell. By weighting one side more than the other, especially when we are at our max/comfort level, we are doomed for failure if it goes against us (which ofcourse it can/will). What i propose will still end with a loss. But the entire point is to reduce that loss as much as possible without picking a side (which could result in a MUCH bigger loss if we are wrong about direction).

So again, how do you do it. Atleast with the way i described in my previous post you know exactly what you will risk per set. And that risk can be increased/decreased by adjusting your High Water Level %.

What do you guys think? Is there a way to minimise the loss more without taking a peference to a side with the large amount of lots opened?
Luck is where preparation meets opportunity.......
  • #582
  • Oct 1, 2009 11:52pm Oct 1, 2009 11:52pm
Hey All... AGAIN.


With what i have been talking about, i will attempt to code it and get back to you. Im learning bits at a time but hopefully i can code enough to see if it will work or not, then hand it over to you TV if it does to do a better/professional job (only if you feel inclined ofcourse).

I know this is typical martingale noob strategy but i just can't let it go.
Im a bit of a dreamer and........what if ??
Luck is where preparation meets opportunity.......
  • #583
  • Edited 4:39am Oct 2, 2009 3:01am | Edited 4:39am
...martingale systems have been taken straight off the roulette wheel and applied to the markets for decades.

Everyone has considered risk mitigation and no one has succeeded. (See the multiple threads on ff and tsd).

May be its just the term risk 'mitigation' i have a problem with. Its a bit like the term 'friendly fire'. Not so friendly when you're on the recieving end of it.

Really what we are talking about is taking a loss on the 'assumption' that we are entering a death trade or what would correctly be termed 'profit mitigation'. Do that a couple of times and you will blow your account any way. Do it once, and because the profits generated in each single set are so small compared to the risk we take, it would takes months to recover using the same system. And aren't we hedging/martingaling because we hate loosing trades?

This is why we need to accept the risk of what we are doing. The death trade will only be known to have existed once it has taken place. There is no predicting it.

So that said all we can do is to try and stack the probability of a close in our favour.

Lets say for example that we have a grid with 3 levels and a 25 pip increment. Thats 100 pips up or down to tp. Historically, in a range bound market for that particular pair, how many complete up down swings can we expect before a breakout. If the answer is 6 (triple top, triple bottom), then we will only run the grid with a lot size and equity that will allow us to survive atleast 8 complete swings (quad top/bottom).

So the answer gentlemen is equity.

No false sense of security here.....
Attached Image (click to enlarge)
Click to Enlarge Name: DetailedStatement.gif Size: 6 KB
bull in a china shop
  • #584
  • Oct 2, 2009 4:08am Oct 2, 2009 4:08am
viper,

i did the testing on geppy using a 25 pip increment and 3 levels on alpari demo.

i had a s1 order taken and pendings were only placed on b2 and b3. when s2 was taken, pendings were correctly placed at b1,b2,b3.

i tried with both the ordrel on and off, and had mgrid running at the same time to compare.

i ran v2.1 on the default settings.

i also have an additional request for development. is it possible to have the ea start a new set by placing pending orders using the tp/sl price of the previous sequence as the starting point. currently, if the price spikes the tp/sl and falls away, the new grid pending orders take the current price as the starting point for the pending orders. (hope i am making sense).

thanks viper and best wishes, i appreciate your effort. looking forward to testing v2.2
bull in a china shop
  • #585
  • Edited 12:48pm Oct 4, 2009 4:19am | Edited 12:48pm
Quoting tradingviper
Version 2.2 is almost done. Should be able to post tomorrow...........
Still eagerly Waiting ....TV u got stuck somewhere?
I see we can progress more on this, if you have more time for programming.... but you seem to have not
I hope ur Boss gives u a week of holidays Try hypnotizing him/her


P.S. I brainstormed more and thought of another good idea.. for RISK mitigation after a number of whipsaws.
Kindly check your PM to discuss that!i mean to discuss even its possible or not..before posting here!
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #586
  • Oct 5, 2009 8:42am Oct 5, 2009 8:42am
Ding Dong!

is TradingViper inside?
its been 5 days
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #587
  • Oct 5, 2009 9:56pm Oct 5, 2009 9:56pm
Here is version 2.2, hot off the press - As stated in earlier press release

It does draw lines at the pivot and every level per plasmapants request with user settable colors (this was trickier than it should have been).
It does add ATR (user defined levels) entry filter
It does fix the SetCount bug so if you tell it to do only one set, it should no longer allow 2
It does contain a special surprise for ForexSaint - near the top of the file, look for your own sandbox called allowSaintTrading(){...

It does NOT fix the apparent "math problem" since I thought that was fixed and have not personally seen it since. I will try to track that down.

Trade Well,
Viper

P.S. To my impatient friends, sorry for the delay. I did finish it Friday just as the markets closed and so was unable to do ANY testing. I did not want to release it like that. When the markets did open on Sunday, I realized I had broke the color line drawing for plasma, so I wanted to fix that. :-)
Attached File(s)
File Type: mq4 ViperGridEA_Version2.2.mq4   52 KB | 1,891 downloads
  • #588
  • Oct 5, 2009 10:09pm Oct 5, 2009 10:09pm
Quoting Ozzie Taz
Further thoughts on this 'death trade' scenario. Im more convinced now about balancing the buys/sells at a particular comfort zone. WHY??

Becuase we just don't know which way price is going to lean. We do know eventually it will hit the TP level (within a reasonable setting) but we can't forsee when/where. We don't know if price will retrace or continue. We need to be able to take advantage of price if it does retrace and also take advantage if price continues on without digging any deeper (if possible) but somehow maintaining a constant level...
Ozzie, This was my first high-water mark approach. The problem I discovered with a balanced grid is that you will still get killed. We are likely to have a lot of positions open (say 40) and if we have the same number in each direction, 20 longs and 20 shorts, your loss will be in the neighborhood of 20*3*increment pips. - Still a huge loss - work some scenarios out on paper and play with some ideas. I do agree that we need to find a way to minimize the damages, but plasma is also correct that ideally, we want every trade to run it's full course and win. One loss will take a long time to make up.

Viper
  • #589
  • Oct 5, 2009 10:20pm Oct 5, 2009 10:20pm
I agree - i have added to the original idea which i PM'd you about. I only use 1 level which makes my idea useable.

I have coded most of it (well i think i have) but my problem is that it is opening too many orders. I can't work it out for the life of me
Luck is where preparation meets opportunity.......
  • #590
  • Oct 5, 2009 10:23pm Oct 5, 2009 10:23pm
Quoting plasmapants
...martingale systems have been taken straight off the roulette wheel and applied to the markets for decades.

Everyone has considered risk mitigation and no one has succeeded. (See the multiple threads on ff and tsd).

May be its just the term risk 'mitigation' i have a problem with. Its a bit like the term 'friendly fire'. Not so friendly when you're on the recieving end of it.

Really what we are talking about is taking a loss on the 'assumption' that we are entering a death trade or what would correctly be termed 'profit mitigation'....
Plasma,
Of course the best solution/answer is to have enough equity/margin to weather any sideways movement. I (and I assume you - is your last name Gates by any chance ?), don't have infinite equity/margin. My rationale was, rather than suffer a margin call and have the broker just close everything in an uncontrolled and unfavorable way, do it yourself in the best way possible to minimize the damages. Of course I would rather NEVER have to do that and use small enough position sizes and have enough margin, etc.

It may be futile. I know a lot of people a lot smarter than I have applied brain power to this. I don't know if any have come up with reasonable solutions to minimize the damages or not. I do KNOW that nobody has ever come up with a way win every time!

Viper
  • #591
  • Oct 5, 2009 10:31pm Oct 5, 2009 10:31pm
Quoting plasmapants
viper,

i did the testing on geppy using a 25 pip increment and 3 levels on alpari demo.

i had a s1 order taken and pendings were only placed on b2 and b3. when s2 was taken, pendings were correctly placed at b1,b2,b3.

i tried with both the ordrel on and off, and had mgrid running at the same time to compare.

i ran v2.1 on the default settings.

i also have an additional request for development. is it possible to have the ea start a new set by placing pending orders using the tp/sl price of the previous sequence as the starting point. currently,...
Hey everyone, I did find 2 missed orders, but unfortunately I don't think I can fix it. I seems to be a server issue or something. see below...

00:17:03 ViperGridEA GBPUSD,H1: Alert: GBPUSD - STARTING NEW TRADE SET; Set: 4; To begin in 10 seconds...
00:17:19 ViperGridEA GBPUSD,H1: open #77521530 buy stop 0.10 GBPUSD at 1.6005 sl: 1.5830 tp: 1.6110 ok
00:17:19 ViperGridEA GBPUSD,H1: open #77521531 sell stop 0.10 GBPUSD at 1.5935 sl: 1.6115 tp: 1.5835 ok
00:17:20 ViperGridEA GBPUSD,H1: open #77521532 buy stop 0.10 GBPUSD at 1.6040 sl: 1.5830 tp: 1.6110 ok
00:17:20 ViperGridEA GBPUSD,H1: open #77521533 sell stop 0.10 GBPUSD at 1.5900 sl: 1.6115 tp: 1.5835 ok
00:17:23 ViperGridEA GBPUSD,H1: shutdown by timeout
00:17:40 ViperGridEA GBPUSD,H1: terminated by timeout

I'll have to investigate what causes a "shutdown by timeout" error... at least it's a lead. Maybe this will lead to fixing the "math bug" somehow. Let's hope :-)

Viper
  • #592
  • Edited 11:17pm Oct 5, 2009 10:34pm | Edited 11:17pm
Thanks viper.. yes impatient !

1) Thanks for the surprise..!Understanding it correctly... it means we can add any code filter ourselves above there after "bool tradingIsAllowed = true;"if we want. Alright !
Can you give us a snippet of code for easy understanding..for example..
lets say a basic filter... Start Grid after MA10 crosses MA 40 from below in TF ea is working on.
I have started coding a bit myself these days and want just basic idea!

2) In ATR filter you have kept RangeFilterPeriod and RangeFilterthreshold only. SO does that mean it will calculate this based on the TF chart , we are using ViperEA on...right? (actually thats why i asked for a TF setting also..any way not a big thingy..we can use the appropriate chart TF)

3)Thanks for Adding Starting price (Inital Price) as comment. Really important if in case any issue is detected!

4)Would be waiting for a level 2 or Modded EA (if not in main versions) about the Grid Conversion thingy i wrote in previous posts.EAGERLY

5) It would be very handy , if EA can list OPEN Buy Orders ( Total No. of orders and Total lot size) and OPEN Sells ( Total No. of orders and Total lot size) as comment on screen (of one instance of EA only, as one Ea can be played on 1 chart at time. imo Code shd have magic number based filter i think to correctly count). It will also help in Analysing Any math problems if they happen...by looking at it and the current price . (Math level problem can also occur when grid size is small and broker doesnt allow stop orders that close to price...or price moves instantly to any side after Starting) I hope u understood.

6) As per normal math logic... I havnt expereinced as snag as yet. And i tell you... math logic posted by FFlash in 1st post is not correct as he has explained. and neither hanover was able to fully explain it.its good that you kept logic from original EA itself.

Looking forward
Regards
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #593
  • Edited 11:34pm Oct 5, 2009 10:57pm | Edited 11:34pm
Quoting tradingviper
I do agree that we need to find a way to minimize the damages, but plasma is also correct that ideally, we want every trade to run it's full course and win. One loss will take a long time to make up.
Frankly, i dont agree we should allow a margin call to happen. Taking loss is part of trading.
Even if this is a Martingale (Modification not true martingale), We are here to TRY modifying it more for our use and not to allow it to whipsaw us to death.
If our entry is so bad that we get whipsawed close to margin call, we should better take the loss saving as much equity we can and better try to make up for that loss in next sets (ofcourse will take time) if we are still interested/have confidence in his entries/strategy

There is no point modifying it here and there with small functions, if we are still trying to trade based on martingale psychology. Then its 100% sure..day will come!
The effort if any we shd put can either be to CUT the loss (which yes takes time to fill up again though it wont be much disheartening if comes later after a no. of winner trading sets)
or
By other Mods at methodology level ( to decrease that "100% death trade once" probabaility) ...keeping the PROFIT math still same.

Quoting tradingviper
I know a lot of people a lot smarter than I have applied brain power to this. I don't know if any have come up with reasonable solutions to minimize the damages or not. I do KNOW that nobody has ever come up with a way win every time!
I feel Grid Conversion from level 2 to level 1 is a lead to a solution if not the solution. We are Doubling the time to Death Trade ,and still keeping 100% win scenario intact!
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #594
  • Edited 8:07am Oct 6, 2009 3:44am | Edited 8:07am
1) A bug ?
Seems like while editing Code of SetCount..it happened.
After end of a Grid with SetCount=1 ( 2 also) i.e, Tp gets hit, its not closing trades on opposite end. i.e.Pending orders. i Waited long ! In following case Buy Tp was triggered but still Sell stop orders are there !
Dont know why.. may be close all code has been bypassed?
EDIT ..just noticed... its closing ordes ..but after some 15minutes or so..

http://i35.tinypic.com/11lhmwx.jpg

2) Further, i See EA is waiting for whopping 10 seconds after its dropped on chart and says so in log too. 10seconds is a big time i think when enteries are to very important for this EA and fully managed. May be a Flexible setting for this too (any way i edited the code and its ok now ..ok in the sense for my taste)
Infact i was thinking ... we can have something like start grid at this price. (as extern parameter)

3) Overall SetCount function is working bcoz i tried closing pending grids with "close all pending order"scripts. and it comments appropriately . Say after 1 close...it says Max Setcount reached thingy on screen.

4) Also wanted to ask..what happens in your opinion if we change things like increment adjust,sl,tp, setcount , time period while ea is within a working set?

tia
Best regards
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #595
  • Oct 6, 2009 4:20am Oct 6, 2009 4:20am
Saint: My point is that we do not know that a whipsaw is about to happen. What we do, (by using any criteria), is assume that its about to happen. On the basis of this assumption we go in to fire fighting mode closing positions and inevitably taking a loss. Do this a couple of times and you will have the same result as a death trade. Your account will be wiped out. Alternatively we could wait a couple of hours only to find that our assumption was a mistake.

When i first started trading the grid live, i knew the possibility of a death trade existed and so wrote of the starting capital in my mind. Just pretended it didn't exist. Assumed it was lost already etc. So i was prepared for risk. It took 10 weeks to double the account. My perspective then changed. It became about capital preservation. Didn't want to loose my winnings etc. So i became more conscience (have i spelled that right?) of the death trade and assumed on no less then 3 occasions that today was gonna be the day. Each time cost me big time. You can see it on the equity curve from my last posting. As it happened, if i had waited a couple of hours or in one case just 15 minutes, i could have closed netting a few dollars.

If you can't embrace risk the net result is a death trade at your own hands. (Talking from experience here).

Anyway i don't want to carry on ranting about it so i'll leave the subject until my next big loss.



vGrid2.2: Testing on gu, 3 levels, 22 pips +2, everything else on default. the math seems to be working! will continue testing to see if the math continues working when you change the lot size. It also seems that the ea is sucking a lot of juice out of my cpu. anyone else experiancing this.
bull in a china shop
  • #596
  • Oct 6, 2009 5:41am Oct 6, 2009 5:41am
correction: anti virus was running. ea does not suck juice.
bull in a china shop
  • #597
  • Oct 6, 2009 8:23am Oct 6, 2009 8:23am
@Plasma
its just that i am putting things with another perspective.
Thinking that death trade will not come..wont make it inevitable. i am also not stressing radical mitigation strategies ... But We need to use them when margin used up/orders open are too much.
I agree to the fact that this stuff is risky... and thinking that this money is over at start of a set only!
But Everything is invented new. We shd STILL try to develop an optimal exit strategy in case that DEATH TRADE is close....further when our analysis/market knowledge is telling us that markets will remain ranging whipsawing us to death. Yes we cant know it will hit TP after so many whipsaws too.

Thats why in fact the Idea i put up is not a Real RISK MITIGATION. Its a WAY to make more odds in our Favour. It will not be interefering /balancing/closing orders..( these have been added as watermark levels by TViper already in ea)! It will be Placing orders at most peripheral level after a no. of whipsaws ( That no. will be something which would have already used up quite a margin already)
I have tried explaining the idea before.. and still waiting for reply from Tviper and you guys.

To culiminate , I feel no point thinking that Death trade will never come !
and when due to bad luck/entries, it comes so early for a new starter (with live account) he/she is over ! I will always try to Find and search ideas where we can close at Break even or delay Death trade as much as we can. Shifting the probability %age to hit TP will always help us.
In any way Forex is a business of probabilities only !
Regards


@ozzie..
how r u doing..long time?
what you think abt the Level Conversion Strategy, i tried explaining in vipergrid ea?
Regards


@ all.
Are you also experiencing the "very very slow" order deletion after Tp is reached, when selecting MaxSetCount other than 0?
100 Fold Challenge->Interested? ->https://www.forexfactory.com/thread/32152
  • #598
  • Oct 6, 2009 5:06pm Oct 6, 2009 5:06pm
test
  • #599
  • Oct 6, 2009 9:40pm Oct 6, 2009 9:40pm
Quoting plasmapants
viper,
i also have an additional request for development. is it possible to have the ea start a new set by placing pending orders using the tp/sl price of the previous sequence as the starting point. currently, if the price spikes the tp/sl and falls away, the new grid pending orders take the current price as the starting point for the pending orders. (hope i am making sense).
I need further clarification: You ALWAYS want a new grid to start where the previous grid TP/SL was ? What if price has already moved thru that point and thru one increment away (e.g. S1), then what ? I did remove the delay starting a new set, so it should be faster grabbing the initialPrice... think about it and let me know what you think...

Viper
  • #600
  • Oct 6, 2009 9:48pm Oct 6, 2009 9:48pm
Quoting forexsaint
Thanks viper.. yes impatient !

1) Thanks for the surprise..!Understanding it correctly... it means we can add any code filter ourselves above there after "bool tradingIsAllowed = true;"if we want. Alright !
Can you give us a snippet of code for easy understanding..for example..
lets say a basic filter... Start Grid after MA10 crosses MA 40 from below in TF ea is working on.
I have started coding a bit myself these days and want just basic idea!
1. Sample code - adding 3 lines of code...
bool
allowSaintTrading()
{
bool tradingIsAllowed = true;

// ************* Enter test code here ************* //
// Should set variable 'tradingIsAllowed' to false if you want to restrict
// trading due to whatever test you conceive
// e.g. if( forexSaint == Sad ) tradingIsAllowed = false;

// Start new trade filter code here...

// Require the 10 day simple MA to be greater than the 40 day
double ma10 = iMA(NULL,0,10,0,MODE_SMA,PRICE_CLOSE,1);
double ma40 = iMA(NULL,0,40,0,MODE_SMA,PRICE_CLOSE,1);

if( ma10 > ma40 ) tradingIsAllowed = true;

// End new trade filter code here.

return (tradingIsAllowed);
}
2 traders viewing now