PipMaker v1 - Price action based EA - page 6

New comment
 

Program Logic...

Don/Neo,

I have been playing around with this EA ever since Don posted it. I have performed many (visual) backtests to watch how it works. And I have seen every iteration of the EA since, until this last one. When I run the worst case scenario... from Nov 26, 2005 through to present, all of the low timeframe tests (boost on/off), eventually blow the account because of low equity. Even my ADX filter doesn't save the low timeframes.

I understand you don't think that the logic is working quite right for the orphaned open orders. I was thinking about how to nullify the orphaned orders, both long and short. Well, I am also a fan of hedging strategies and was thinking... if there was a way to ensure that the biggest losing open orders are paired with the biggest gaining orders (don't get closed out - and that are close to off-setting the loss), you could continue to make profitable orders inside the envelope (high-low of the market).

I'm not sure if I'm explaining it right, but, it seems that the logic needs to be more intelligent on how it closes profitable orders. It needs to offset the losers somehow, so that additional trades can be entered within the profit envelope (low range consolidations). Then if margin is needed, the EA would close offset orders to free the margin.

So, in essense, if several orders long and short have been established (there is always more of one than the other), and price makes new highs, the short orders would be offset by "close-by" long orders. Those long orders would not be liquidated because they become the hedge for the short orders. The long orders that are left over and are in excess of the hedges for the short orders would be closed for a profit. Price would continue to move. Now if price moves to new lows, the EA logic would ensure that the losing long orders are hedged by appropriate in-profit short orders, thus hedged.

I hope that I'm getting my point across. I wish my programming skills were up to my ideas concerning this logic process.

Great job, Don and Neo.

Carl

 

Original Theory

The original strategy we traded was to simply close all buys or sells once they met the profit level. This was very successful. The problem though was trending. If the price went high, you were stuck with a lot of sells open and conversely, if the price dropped, you were left with a bunch of buys. My partner started taking and closing three profitable orders along with one losing order once he met his profit goal of $20. He had a butt load of orders open and using this technique he was able to significantly reduce the number of open orders. I tried to code this but was unable to correctly identify the top three orders. He managed to come up with the current profitable order routine, so, we modified the technique to include all profitable orders and not just the top three. I think that using the top three orders in conjunction with the losing order might produce more profit and keep the margin tighter. What do you think?

Don

 

I been tinkering w/ this EA the last couple of days. Is the new neo inspired version still to be used on 1m TFs primarily?

 

Thanks

don_forex:
I use 1M time frame as there isn't any indicators involved.

I am trying to incorporate a LWMA and still will use a 1M tf.

Don

Thanks for your answer

 

Yes...

don_forex:
The original strategy we traded was to simply close all buys or sells once they met the profit level. This was very successful. The problem though was trending. If the price went high, you were stuck with a lot of sells open and conversely, if the price dropped, you were left with a bunch of buys. My partner started taking and closing three profitable orders along with one losing order once he met his profit goal of $20. He had a butt load of orders open and using this technique he was able to significantly reduce the number of open orders. I tried to code this but was unable to correctly identify the top three orders. He managed to come up with the current profitable order routine, so, we modified the technique to include all profitable orders and not just the top three. I think that using the top three orders in conjunction with the losing order might produce more profit and keep the margin tighter. What do you think? Don

Don,

It sounds like you should try to stick with your friends strategy. This current strategy will orphan a lot of open losses in a trend. There is no other way around it.

Carl

 

Hi, Don

The issue over closing the 3 most profitable vs all profitable orders will be broker-dependent. I believe FXDD offset longs & shorts so option 1 would be best, whereas most other brokers (my own included ) do not, so option 2 would be best in that case.

Perhaps the best solution would be the facility to select the method - I'll take a look later today as I'm sure it won't be difficult to achieve.

Neo

 
Neo:
Hi, Don

The issue over closing the 3 most profitable vs all profitable orders will be broker-dependent. I believe FXDD offset longs & shorts so option 1 would be best, whereas most other brokers (my own included ) do not, so option 2 would be best in that case.

Perhaps the best solution would be the facility to select the method - I'll take a look later today as I'm sure it won't be difficult to achieve.

Neo

You are definitely a better programmer than I am. It will probably be quite easy for you. It was beyond my capabilities.

Don

 

1M timeframe

lepiricus:
I been tinkering w/ this EA the last couple of days. Is the new neo inspired version still to be used on 1m TFs primarily?

Yes. I will pretty much always use the 1M timeframe.

Don

 

This EA backtests very well even with massive drawdowns, but on forward testing it incures a massive drawdown and is painful to watch sometimes lol. Much potential here though. I just got to learn to walk away as it runs instead of worrinng about a dozen or so open positions.

 

In order to prevent massive draw downs the following system might be adopted which is similar to the system told by don_forex The EA should allow the user determine a an equity target.

If the equity target is reached the EA should close all the open orders at once and the restart again.

Suppose if the initial balance is $500 and the target equity set is +$25. ( not the profit ..it is equity). Once the equity achieves the target +$25 ( Balance $525)in equity, it closes all the open orders. The EA restarts again.

Again when the equity reaches the target profit of another $25 ( Balance $550)it again closes all the orders.

Currently I am struggling to do this manually. Better if it is incorporated in EA. I use version 2 .I find that more profitable with above system

12345678910111213...225
New comment