- #394
- Edited 11:50am Aug 19, 2010 7:23am | Edited 11:50am
You are using the M1 chart to run the EA, which is fine.
However, the EA uses the M15 data to determine re-entries.
If the M15 data is incorrect, then the EA will not re-enter properly, and this is what we are seing here.
Re-entries should occur only when an M15 bar has closed inside the box, which is obviously not the case in your chart if we try to draw 15-minutes "super-bars" out of the M1 chart.
Make sure you correctly (re-)built the M15 FXT files in tester/history from your Dukascopy tick data with the Dukascopy2FXT script.
Otherwize mt4...
works great with M15 Data!
Another thing I found out today: When I have an mini-acc with little margin, then the setting lot = 0.1 & risk = 1% became an error. look at pic
It happened two times with these settings. But some orders with these setting were ok!
I have changed the CT entries due to the fact you are more prone to see reversals at the Target levels rather than the edge of the box. The natural tendency is to see reversals at key Fibonacci extensions and we are not taking advantage of that by using the box edges for entries. We got creamed on the Eur/Usd this week and the CT concept is only a few weeks old, we needed a better solution.
[color=Navy]Look at the way the BO strategy has so many trades that hit Target 1 and retrace back...
An applause
for your revisions. One step closer to greatness. Good Job. I now see a revision to the entries a future possibility. Keep up the good work. Don't take any loss lightly and don't stop eliminating them. Throw out that thought it is impossible to eliminate all of them.It would be nice to see you post the results a little differently and more advantageous to you. The Martingale is good to eliminate prior losses but at this stage of the development process I believe you would see the weaknesses of the system quicker if you showed a column of 1. Losses, 2. Winners, and 3. Martingale applied. This way you see the losses first which is what you want to key on and improve upon.
When I first started writing EA's I also used a very aggressive Martingale system which worked very well but the the program always hit a brick wall as far as the amount of profit over a given period of time in back testing. The only way I could get the EA to produce more pips was to key on the losing trades. At first this proved correct as the profit went through the roof but as each loss was eliminated then my Martingale system didn't come into play because it only came into play after a loss. That's when I realized that another method needed to be added to the EA to maximize the results. The maximizing of the EA became an obsession as the total pip count went through the roof. The question in my mind was just how far could I take this thing.
I then added contracts as profits were won on each new trade and the pip count quadrupled. Again it was an extremely aggressive formula but still the end results were not quite there yet. You see I decided to learn about risk after losing 250K in one week on a back test which was similar to what you just experienced. I went looking for information on the subject and I found what I was looking for in an article on the internet. I applied the new formula to the EA. The end result was very impressive and almost as good as the two very aggressive methods, the contract adder system and the Martingale system put together.
All this with no addition to risk. I eventually took the Martingale out of my EA as it was no longer needed. I know you're not there yet but keep adjusting and you'll eventually get there. Nothing is impossible. You are already way ahead of where I was in the little time you started but I also had to teach myself how to write code.
Sorry for being so long winded.
Good Luck,
Johnny