Thanks for the new EA
I checked out your profile, and I must say it's very impressive. It looks like you've added several systems since September? Are they systems you wrote?
Originally Posted by onu
Update to version 1.1
This update fixes some problems with mini accounts.
I apologize if this isn't the right way to post an update. Version 1.1 fixes problems with mini accounts where the lot step is 0.1. My live account is at MT4Live.com - Share Your MT4 Statements Free if you wanna see how it's doing.
With attachments this time.
Here are the files ... sorry.
I see that this is a martingale-based EA. A word of warning - those EAs may seem to work in the short-term, but in the long term it is statistically proven that they will harm your account. Just a kind warning to the users that may trade this EA.
Yes, it's a martingale-based EA, but I've taken several steps to reduce the risk to your account. Forex trading is inherently risky so is this EA risk free? Of course not. If anyone ever claims to have a risk-free EA don't use it because they are lying. I've seen EAs that make that claim, some verrrry expensive.
I watch several forums where technical analysis is discussed and believe me, this EA has done much better. Why? Because it has no emotion. It doesn't think 'the market *should* be doing this' and trade accordingly, it goes by what the market is actually doing and adjusts.
Some of the things I've done to reduce risk:
- Money Management. If the free margin drops below an amount set in the parameters all open trades will close.
- Use of the indicator Parabolic SAR. Rather than blindly opening the same losing trades over and over it changes according to the market.
- Reduced martingale multiplier. This EA uses a default multiplier of 1.7 rather than 2. It can also be adjusted up or down according to a traders risk tolerance.
- No SLs. This EA uses Limit orders to stop losing trades, so anyone who thinks Brokers stop-hunt, this EA doesn't have 'em.
I know I'm defensive about my EA but I believe it works. I agree people should be aware that it's martingale-based, and that's something I haven't tried to hide. I wouldn't be using it on my live account if I didn't think it worked! My account was at $1626 when I started PentzParabolic. Now it's at $2024.50, a gain of $397.95, for a 24.5% increase in about 2 weeks. I agree that's a short-term result. So if being a martingale-based EA concerns you, just watch my account for a while, let me be the guinea pig.
Originally Posted by XRanger
I got quite impressive results with version 1.1 your EA. Good job!
The EA works very well with a demo account with FXDD, however it generates errors with Alpari and FXCM demo accounts. It seems that is due to the lots of 0.01 generate errors. I tried with lots of 0.1 with 0.1 increments but results are not even close to the one obtained with 0.01 lots.
Any recommendation to make your EA work better with 0.1 lots?
What errors are you seeing? It should work fine on Alpari, although I haven't tested it directly. I have a demo account so I'll try it.
Originally Posted by philippe07024
I have it running on a FXOpen demo account right now that is a mini account so it has default lot size of 0.1. So far the results have been good. It sounds like you manually set the LotSize and LotStep parameters to 0.1? What balance did the account have? What was the setting of the "FreeMarginMin" paramter?
Have you backtested this?
I tested on fiber with default settings since start of year as you suggest. With AutoLotSize = false and trading 0.01 lots its drawdown was over 26k. Whilst it somehow didn't blow the 10k test account, the net loss was 9.9k so it wasn't too far off a blowout.
I then tested it with AutoLotSize = true and, guess what, it did blow the account.
I don't want to be too critical, because I acknowledge that you don't have to share your work, but there are parts of the code that have been poorly thought out. Firstly, you should have used the StringSubstr() function to overcome the various prefixes and suffixes used by brokers. Even so, you don't even need all of those different magic numbers as you should include the Symbol() within the conditions. Finally, just a simple check to see if the Trade Context is busy or not isn't really enough.
Tags for this Thread