well, it looks like an edge so far. If it test out, it wont be hard to go live for mini tests. The real question is can a certain type of market move beat it?
and how much will it take when it happens... weird chops ..that never happen in forex trading ya know.

I think it is all going to depend on how it is configured. There is going to be so many options to test. But, like you say... it seems like there is always some kind of market situation to throw a wrench in a system. What are you're ideas to deal with those situations?

If I knew that I d be a rich and quit my day job. Kayvan had an initial entry idea so we have to stick with that. I think your ea, if working like he is saying is robust enough to play around with some adjustments. No matter what you put out there, Markets always have losers no matter what you program. The idea is to get a lot of wins before that happens. A close that takes into consideration a % of max profit as well as a max drawdown when the orders are open as essential to a martingale ea. At some point, fear goes to panick or you blow it out.

To stop the blow out I allow 9 trades and simply close all after that..There seems to be enough wins thus far with doubling without this hitting below (has not hit yet). We shall see...There is a log in here to my test account somewhere..

MaxTrades: 9
CloseAllMax: True

thank you wolfe . you are really a nice guy .

i read the last few posts . what i understood , is that,there is a fear of using this method in real .all the fears are about ,how we must do if the market goes to a range hours . they are right ,As you know, markets only trend for around 25-30% of the time and during non-trending conditions the use of a moving average is greatly diminished the false orders. we must use an ea to reduce the false orders . if you mind there is a rule published by Tom De Mark. may be you want to use it . he devised a couple of simple rules for defining the likelihood that a trend is in place and then how to calculate when it may be ending...he says in his rules ,a trend will only appear on a chart of the market has closed outside the range of the previous 8 price bars (this works on any time frame). Once this condition is met, he uses a simple moving average of the highs for a downtrend and lows for an uptrend. So, for example, when I say that GBP USD has suggested a short position from 1.9500 this was when we made a close outside the previous 8 session range to the downside.The average stays on the chart for a downtrend until we have*achieve a price bar which doesn't manage to close below any of the previous 4 bars lows, implying the downtrend has lost it's momentum.
It is a very simple idea, but does have fantastic results .

any way you are the boss.

I have been forward-testing this on H1 Cable (see log-in) for the last 8 trading days/24hrs per day with autorestart set to true in an IBFX Demo Mini Acct. The EA is posted in post 432. There have been many range periods (I have seen a 6.40 lotsize)...As long as you set your TP to be conservative (5 bucks per 1k with a 0.1/0.1) you should be able to weather more "range-storms" before maxing out your trade progression of 9 with doubling (0.1, 0.2, 0.4, 0.8, 1.6, 3.2, 6.4, 12.8, 25.6).

Login : 1854641
Investor : se1waqn (read only password)
Server: InterbankFX-Demo Accounts

When 1.6 gets out the only adjustments I will tinker with is to use a lookback period of 2 on the M30 with a 3 pip lag. I will run that on a Demo and put the above on a live account.

ES

P.S. always use a safety:

MaxTrades: 9
CloseAllMax: True

Attached Images

DetailedStatement.gif (5.7 KB, 186 views)

ES-

I logged in to your IBFX account to see your test results. They look encouraging. The 1.6 lot size was a little scary, but at least it pulled out. I would hate to see it go past that 1.6 progression.

Yeh, I would be "sweating" on that. -- I think though you almost have to do this in order to make it with a martingale. Like roulette, its a double down mentality. It would take a strong stomache to do nine deep. I am not sure if there is much safety at that increment. Although on one martengale, I had an error one time and openied up well over 100 orders in an hour. Got home from work and was up a lot of money..but dam I was closing those fast. I am not yet sure how to turn kayvans ma theory from demark into an indicator but definetely something to work out.

Interesting thoughts kayvan.

I don't know if you realize this yet or not, but the latest version I am about to release has a user option you can set for how many bars you want the EA to look back on. For example, if you choose 8 bars, it will use the high and the low of the last 8 bars to implement your method (now known as the kayvan method).

Here is what the parameters for the kayvan method are-

I think you will be busy for a while.