Introducing PivotMagic EA

I have been working on a system that uses pivots with very simple pivot-related instructions. I derived those from an academic who teaches signals and is trader himself.

It seems to be working with almost no losses, but it is still in the very early development stages and needs lots of work. The code is still very very messy as well. So bear with me if you find unexpected results.

I would like to invite you folks to have a look at the backtesting I did for the system for the period May 1-26 for the pair: USDCHF on the 5-M chart. The result was an astounding 2300 pips . Also the modeling quality was acceptable (84%).

Also attached you'll find the EA so you could do your own testing. I believe that you'll have to play with the parameters somewhat to see how each is affecting every pair. I would love to know the optimum setting for that.

However, there are some disadvantages in the system:

1- It uses trailing stops a lot, meaning that one needs to stick with the system and ensures no connection failures, etc. would happen

2- It needs follow-up all the time because it takes long (multi-day) trades. Therefore, any backtest should be at least for a week.

3- Each pair needs to be dealt with separately because there are massive differences in the way they respond to this strategy. In fact, I noticed that some pairs don't like it at all!!!

4- The parameters in the EA are a bit delicate and could cause lots of confusion (particularly as every pair responds differently to each one of them)

5- It only works so far with the 5-M setup. Don't try anything else. I'll try to work on other options in the future.

I did testings for all major pairs and got good results on the backtester. Of course in reality, things are much different.

But with extra work and fine-tuning, I believe this one could be added to the list of winning strategies.

Please provide me with comments and suggestions first about the performance of the strategy and its negatives.

Here's a sneak peak at this specific test result:

IMPORTANT NOTE ABOUT THE TEST:

Given that this is a strategy based on continuity, the fact that the test abruptly stops at an specific point makes potential loss at the end of the test. I'm sure you'r familiar with this. So the best thing is to assume that the last potential loss (a few hundred pips sometimes) should not have been if the test was allowed to continue to a suitable point. In the future, I can also handle this easily with some conditional setup (e.g., ensure trade stops with profit (or minimum loss) before end of specific period).

Waiting for your remarks...

If you want to ask about the strategy itself. send me a PM or email, I'll be happy to show you. But I do hope you could help test it first.

investor_me

Attached Files

	USDCHF-results-PivotMagic.zip (11.6 KB, 681 views)
	PivotMagic_EA.zip (10.0 KB, 1138 views)

This is the illustration of the trades made by the EA in the given test USDCHF 5-Min (17-26 May).


Notice how the EA beautifully executed the trades with the trailing point utilized to avoid any potential losses.

However, I noticed that 23 May had no trades at all. It may be because it did not cross any vital pivot lines based on the given parameters.

But hey, it did work for the other days.

Those who test the EA can post what they found. Thanks...

Interesting EA. I would test it out and have some ideas to improve it but will not touch it without the source code. Sorry.

Listen, there is a lot of talent here that can help you if you provide the source code to the EA. Who know, we could get it working on every currency pair and timeframes allowing everyone to prosper.

Hello Invester me
The file you sent is exe4 can you send the file as mq4?
Thanks

i'm going to try your ea during this week. If i can, later on will post the results. It looks good.

I will test it too. Looks good on backtest so far. Please post mq4 .

it's useless to test an EA for just 3 days. The EA is going to expire on the 31st of this month. Here is what Investor me wrote in the yahoo message group :

"Yes it will not work after May 31, because it is incomplete code and
it must only be used for testing. If you can test it within this
framework that's good. But if you want it extended, I'll be able to
submit a more solid EA as we continue in its development.

Thanks "


Furthermore, he also said that this EA might be commercial in the near futur:

"We're still working on the source code so far and I can tell you it
looks promising. But the professor who provided the scientific backing
to his theory used in the code is reluctant to release it to the
public. I'm not sure if he will agree to that because he is seriously
considering making it a commercial package.

I'll try to see if I can post the code or part of it at least."

Let's wait an see what happens, but I wouldn't help out in testing it at all if this is going to be a commercial package. And if thathappens this thread should be moved to the commercial section of this forum.

Sada

If you can make money with it, then why make it commercial package?

You're asking the community to help you test this then you're going to turn around and charge them to use it?

Thanks for the opportunity, how kind of you...

You might as well post the source code and take ownership of your code because someone will clone this and claim it as their own. It's not too difficult to reverse engineer an EX4 file either by actually decompiling it or just figuring out how it works.