Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Better alternative for Print() for debugging 29 replies

Debugging EAs 3 replies

Need help debugging 0 replies

Debugging EAs 2 replies

Help debugging my EA I feel like the solution is too simple. 1 reply

Attachments: Gambler's Pips EA needs debugging
Exit Attachments

Gambler's Pips EA needs debugging

Thats the one I am using, it does dont seem to work all the time for me , maybe it is related to the pairs and broker demo i am using
I did have some good results with it , so I will keep playing with it.

Quoting jaws810
{quote} The modified version in post 11 in this thread does seem to be working right. I just ran it with the martingale turned off in visual mode. It opens a single trade and another as soon as TP or SL is hit.
If anyone is interested I found a more accurate method of choosing trades with a better w/l ratio. Going back to 2007 it wins more than 60% of the trades using PSAR to choose the trades and a modified martingale that is limited to prevent a death spiral. It's profitable back to 2007. It isn't unstoppable because it does fail if you go back to 2000.

Attached File(s)
File Type: mq4 PSAR Martingale.mq4   6 KB | 413 downloads


Attached Image (click to enlarge)
Click to Enlarge Name: 2007-2014 backtest.JPG Size: 96 KB
Quoting toddanderson
Thats the one I am using, it does dont seem to work all the time for me , maybe it is related to the pairs and broker demo i am using I did have some good results with it , so I will keep playing with it. {quote}
I designed it for the euro and the broker I use is OandA, it might be the lotdigits setting, if your broker only allows .1 lot increments it would need to be set to 1 or it would stop trading due to invalid lot size.
Quoting jaws810
........ Here's how it works, it makes a trade starting with as many lots as you want and then increases lot size on a loss and decreases lot size by the same amount with a win. So if you started with 1 lot and increase/decrease by 1 lot the progression for this series( L,L,W,L,L,W,W,L,W,L,W,W) lots for those trades would look like this: (1,2,3,2,3,4,3,2,3,2,3,2,1) {file}

I found this idea interesting and thought i would see how it could be modelled in MS Excel.

The attached spreadsheet (assuming I did the maths correctly) takes 500 random events (call them trades) and using the scaling identified by Jaws to set the next trade lot size, all lots based on a single lot size of 1% of account balance, reveals an interesting outcome.

If you compare the scale in/out approach when compared to a fixed lot (ie a fixed percentage of account balance) the results are fairly dramatic - particularly with any strategy that delivers a 60% or better reward to risk ratio.

Just open up the spreadsheet and keep hitting the [F9] key to recalculate a new set of random trades.

Have a play with the other settings also.
Attached File(s)
File Type: xlsx Gambler calcs.xlsx   104 KB | 407 downloads
" check out The Traders Outpost "
Quoting jaws810
{quote} I know Martingale will always lose but this isn't really Martingale, and believe me I'll be the first to post any problems once I get the results, but I can't do that unless the EA is working right. Everyone knows EA trading is risky, especially with systems like mine but what's the harm in trying, nothing ventured nothing gained. I copied the results from the flawed backtest and I'm manually correcting the lot sizes and I'm going to see what it could do but it will take a while and I only have 400 trades ATM, but it's a start until someone...
Martingale: 4. A system of gambling in which the stakes are doubled or otherwise raised after each loss. (dictionary.com)

By definition this is martingale, but I would be keen to see what this can do. I haven't tried a progressive lot size system: all other martingale systems are simply a double up or multiplication of the previous trade lot size, which causes a compounding effect that blows every account.

1, 2, 4, 8, 16, 32, 64, 128, etc.

A progressive lot size limits increases in the new trades to a fixed amount, and so is less likely to suffer a margin call as compared with a compounding martingale strategy.

1, 2, 3, 4, 5, 6, 7, 8, 9, etc.

Here is another thread that may interest you on this subject: http://www.forexfactory.com/showthread.php?t=497448
Quoting jaws810
I used to be a gambler and I used a step progression for betting on roulette(it made lots of money until the inevitable death spiral due to inadequate funds or table betting limits), and I think I finally figured out how to apply it successfully to forex. By idea is for the EA is to make it buy and sell based on the last candle basically random, red candle it sells blue it buys, then it uses a slow step progression/regression MM strategy to make it profitable. I'm not a programmer but I understand enough to tinker with an existing program, I used...
May look impressive, but the forex coin is not balanced. If one's longer term prediction is correct, there will be many winners in a row and if longer term direction is wrong, many losers in a row.
Attached Image (click to enlarge)
Click to Enlarge Name: Untitled1.jpg Size: 57 KB
The Thief of Wall Street
Quoting iwjw
this one should do the job {file}
I copy the file to Experts and restart MT4, but MT4 seems to delete this one as it doesn't appear in Navigator and no longer appears in Experts. I tried this on two separate terminals with latest build 745 (24 October).
  •  FerruFx
  • Joined May 2007 | Status: MT4/MT5 EAs/Indicators/Alerts coder | 6,502 Posts | Online Now
Quoting bluepanther
{quote} I copy the file to Experts and restart MT4, but MT4 seems to delete this one as it doesn't appear in Navigator and no longer appears in Experts. I tried this on two separate terminals with latest build 745 (24 October).
Because the code uses the old MT4 language standard so when your platform try to compile it in background, it has many "warnings" so it removes it.
MT4/MT5 EAs/Indicators/Alerts coder
Quoting Arctic4x
Hi Jaws. Listen to what FerruFx is telling you. It will not work. A endless number of traders and also gamblers have tested this before you, myself included. I did not use EA's for testing but i prefer Excel and the result is always the same. It do not work. But like Spaceduck say if you have a trading system with better than 50% hitrate and 1:1 RR and also overcome the trading cost like spread/slippage, then there are som MM systems that will work. But if you have a tradingsystem like that you do not need this type of MM because you already have...

Remember I read about one computer modelling of Martingale betting (in wikipedia), there was a supposition that sharp unexpected increment of bet during losses increase your chances to profit in about 11 times (!), and the computer test confirmed that. I believe that Martingale works, but not with that strict math as being discussed there..maybe it lacks some freedom or suddeness of results in calculation.
  • Post #30
  • Quote
  • Last Post: Nov 16, 2014 11:55am Nov 16, 2014 11:55am
Quoting FerruFx
{quote} Because the code uses the old MT4 language standard so when your platform try to compile it in background, it has many "warnings" so it removes it.
Figures... Metaquotes can't get their own shit working properly. Heck, they erased my old EAs when they force-upgraded everyone to the new 600 builds which altered the directory structure of MT4. They are not thinking of backward-compatibility for the end user - only their own agenda to move pps to MT5.

I ain't a programmer: if someone has the smarts and the generosity, please consider converting this EA to compatibility with newer MT4 builds.
0 traders viewing now